basic_usage.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6a3e0c30-c009-424e-ae3f-0e75d6fcd164",
   "metadata": {},
   "outputs": [],
   "source": [
    "from jaxosqp import osqp\n",
    "import jax\n",
    "import jax.numpy as jnp\n",
    "from jax import random\n",
    "\n",
    "from jax.experimental import sparse\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0f26913e-290c-468e-9d85-8daac0afa00e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-05-31 17:26:19.816114: W external/xla/xla/service/gpu/nvptx_compiler.cc:564] The NVIDIA driver's CUDA version is 12.0 which is older than the ptxas CUDA version (12.1.105). Because the driver is older than the ptxas version, XLA is disabling parallel compilation, which may slow down compilation. You should update your NVIDIA driver or use the NVIDIA-provided CUDA forward compatibility packages.\n"
     ]
    }
   ],
   "source": [
    "# Generate some random problem data.\n",
    "from jax.config import config; config.update(\"jax_enable_x64\", False)\n",
    "B = 50\n",
    "n = 20\n",
    "m = 200\n",
    "\n",
    "outer = lambda A: A @ A.T + 1e-2 * sparse.eye(n)\n",
    "key = random.PRNGKey(1)\n",
    "key, subkey = random.split(key)\n",
    "P = sparse.random_bcoo(key, (B, n, n), nse=0.95, n_batch=1, generator=random.normal).todense()\n",
    "P = jax.vmap(outer)(P) # Ensure P >= 0.\n",
    "\n",
    "key, subkey = random.split(key)\n",
    "q = random.normal(subkey, (B, n))\n",
    "key, subkey = random.split(key)\n",
    "\n",
    "# Add some random constraints (a la OSQP benchmarks\n",
    "A = sparse.random_bcoo(subkey, (B, m, n), nse=0.95, n_batch=1, generator=random.normal).todense()\n",
    "key, subkey = random.split(key)\n",
    "l = -random.uniform(subkey, (B, m))\n",
    "key, subkey = random.split(key)\n",
    "u = random.uniform(subkey, (B, m))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "35427198-6eaa-496b-ae5f-db6f5a00510d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59.3 ms ± 136 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)\n",
      "59.5 ms ± 349 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)\n"
     ]
    }
   ],
   "source": [
    "def solve_osqp(P, q, A, l, u):\n",
    "    prob, data, state = jax.vmap(osqp.OSQPProblem.from_data)(P, q, A, l, u)\n",
    "    return jax.vmap(prob.solve)(data, state)\n",
    "\n",
    "# %timeit sol_mine = solve_osqp(P, q, A, l, u)\n",
    "\n",
    "# prob, data, state = osqp.OSQPProblem.from_data(\n",
    "    # P, q, A, l, u)\n",
    "# sol = jax.vmap(prob.solve)(data, state)\n",
    "solve_jitd = jax.jit(sparse.sparsify(solve_osqp))#(P, q, A, l, u)\n",
    "sol = solve_jitd(P, q, A, l, u)\n",
    "%timeit sol = solve_jitd(P, q, A, l, u)\n",
    "%timeit sol = solve_jitd(P, q, A, l, u)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f56d6cc1-6ca3-41d8-aff5-80b7659b44ca",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array(0., dtype=float32)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "jnp.zeros(())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 470,
   "id": "d2ded574",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 209 ms, sys: 0 ns, total: 209 ms\n",
      "Wall time: 209 ms\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Array([[ 1.38864145e-01,  1.70732021e-01, -2.10882723e-01,\n",
       "        -2.00839490e-01, -9.35341835e-01,  2.08578989e-01,\n",
       "         1.92606375e-02,  3.96503538e-01,  2.00927362e-01,\n",
       "         7.57957622e-02, -8.98166597e-02,  2.51939952e-01,\n",
       "        -3.11227024e-01,  6.90020770e-02,  2.50960812e-02,\n",
       "         1.03611052e-01, -2.88685888e-01, -2.51004636e-01,\n",
       "        -1.39635980e-01,  2.48320192e-01],\n",
       "       [-3.77256930e-01,  2.43834347e-01, -4.94570434e-02,\n",
       "         2.16760844e-01,  4.66569886e-02,  2.27107108e-02,\n",
       "        -1.70008302e-01,  2.50580851e-02, -1.61573857e-01,\n",
       "        -2.12607995e-01, -8.02972168e-02, -6.36534169e-02,\n",
       "         1.63530290e-01,  4.94016707e-02, -2.20710650e-01,\n",
       "        -2.69344568e-01, -4.81215417e-02, -4.17932897e-04,\n",
       "        -5.91501534e-01, -3.37985128e-01],\n",
       "       [ 3.31494361e-02,  8.65903199e-02,  1.92397296e-01,\n",
       "         5.06616607e-02,  5.25929213e-01, -2.05326714e-02,\n",
       "        -2.41110176e-01,  5.46169639e-01, -9.09369528e-01,\n",
       "        -3.22042927e-02,  3.56898680e-02, -3.58549580e-02,\n",
       "        -2.98611045e-01, -3.20148021e-01, -1.56980783e-01,\n",
       "         1.77667066e-02, -3.26480567e-01, -4.57195342e-01,\n",
       "        -2.96711214e-02, -2.20282748e-01],\n",
       "       [-2.23300651e-01,  1.74173098e-02,  2.18200386e-01,\n",
       "         3.83497655e-01,  2.14775540e-02,  2.03369856e-02,\n",
       "         9.08104926e-02,  9.66768414e-02, -7.79651627e-02,\n",
       "        -3.86673212e-01,  5.89835346e-02,  2.55940735e-01,\n",
       "         6.57068491e-02, -2.20432028e-01, -2.90740550e-01,\n",
       "         4.72505242e-02, -2.95580924e-01, -2.62521863e-01,\n",
       "        -4.23580967e-02,  2.41101757e-02],\n",
       "       [-2.14482039e-01, -1.07427016e-02, -8.86005014e-02,\n",
       "        -1.33087218e-01,  6.13390915e-02,  8.33242238e-02,\n",
       "        -2.17291772e-01, -5.00651076e-03,  1.36117011e-01,\n",
       "        -1.25282139e-01,  3.30504328e-01, -1.06036820e-01,\n",
       "        -2.45477676e-01,  5.94223775e-02,  1.02496676e-01,\n",
       "         7.37428069e-02, -1.18937761e-01, -6.21621013e-01,\n",
       "         3.06168854e-01,  2.86824554e-02]], dtype=float32)"
      ]
     },
     "execution_count": 470,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%time solve_osqp(P, q, A, l, u)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "id": "2247cc20-b4e4-4d59-9563-7ffdd71ea16a",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "       False, False,  True,  True,  True,  True,  True, False, False,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True,  True,  True,  True,  True,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False, False,  True,\n",
      "       False, False,  True, False,  True,  True,  True, False,  True,\n",
      "        True,  True,  True,  True, False, False,  True, False, False,\n",
      "       False,  True,  True, False,  True, False, False,  True, False,\n",
      "       False,  True,  True, False,  True, False, False,  True, False,\n",
      "        True, False, False,  True,  True, False, False,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.39740178, 0.39740178, 0.39740178, ..., 0.39740178, 0.39740178,\n",
      "        0.39740178]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True, False,  True,\n",
      "        True, False,  True,  True,  True,  True,  True, False, False,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True, False, False, False,  True,\n",
      "       False, False,  True, False,  True,  True,  True, False,  True,\n",
      "        True,  True,  True,  True,  True, False,  True, False, False,\n",
      "        True,  True,  True, False,  True, False, False,  True, False,\n",
      "        True,  True,  True, False,  True, False, False,  True, False,\n",
      "        True, False, False,  True,  True, False,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.39637244, 0.39637244, 0.39637244, ..., 0.39637244, 0.39637244,\n",
      "        0.39637244]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True, False,  True,\n",
      "        True,  True,  True,  True, False,  True,  True, False, False,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True, False,  True, False,  True,  True,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False, False,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True,  True,\n",
      "        True, False,  True,  True,  True, False, False,  True,  True,\n",
      "       False,  True,  True, False,  True, False, False,  True, False,\n",
      "        True,  True,  True, False,  True, False, False,  True, False,\n",
      "        True, False, False,  True,  True,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.39462486, 0.39462486, 0.39462486, ..., 0.39462486, 0.39462486,\n",
      "        0.39462486]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True, False,  True,\n",
      "        True, False, False,  True,  True,  True,  True, False, False,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True,  True,  True, False,\n",
      "        True, False, False,  True, False, False,  True,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False, False,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True,  True,\n",
      "       False, False,  True, False,  True, False, False, False, False,\n",
      "       False,  True,  True, False,  True, False, False,  True, False,\n",
      "        True,  True,  True, False,  True, False, False,  True, False,\n",
      "        True, False, False,  True, False,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.39303362, 0.39303362, 0.39303362, ..., 0.39303362, 0.39303362,\n",
      "        0.39303362]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False,  True,  True,  True, False,  True,\n",
      "        True, False, False,  True,  True,  True,  True, False, False,\n",
      "       False,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False,  True, False,\n",
      "        True, False, False,  True, False, False,  True,  True, False,\n",
      "        True,  True,  True,  True, False, False, False, False,  True,\n",
      "       False, False,  True, False,  True, False,  True,  True,  True,\n",
      "       False,  True,  True, False, False, False, False, False, False,\n",
      "       False,  True,  True, False,  True,  True, False,  True, False,\n",
      "        True,  True,  True, False,  True, False, False,  True, False,\n",
      "        True, False, False,  True, False,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ],\n",
      "       [0.3624757, 0.3624757, 0.3624757, ..., 0.3624757, 0.3624757,\n",
      "        0.3624757],\n",
      "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ],\n",
      "       ...,\n",
      "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ],\n",
      "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ],\n",
      "       [0.391664 , 0.391664 , 0.391664 , ..., 0.391664 , 0.391664 ,\n",
      "        0.391664 ]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False,  True,  True,  True, False,  True,\n",
      "        True, False, False,  True, False,  True,  True, False, False,\n",
      "       False,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False, False,\n",
      "        True, False,  True, False, False, False,  True,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False, False,  True,\n",
      "       False, False,  True, False,  True, False,  True,  True,  True,\n",
      "       False,  True,  True, False, False, False, False, False, False,\n",
      "       False,  True,  True, False,  True, False, False,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True, False,\n",
      "        True,  True, False,  True, False, False,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.39003807, 0.39003807, 0.39003807, ..., 0.39003807, 0.39003807,\n",
      "        0.39003807],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.3910205 , 0.3910205 , 0.3910205 , ..., 0.3910205 , 0.3910205 ,\n",
      "        0.3910205 ]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False,  True,  True,  True, False,  True,\n",
      "        True,  True, False,  True, False,  True,  True, False, False,\n",
      "       False,  True, False,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False, False,\n",
      "        True, False,  True, False, False,  True,  True,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False, False,  True,\n",
      "       False, False,  True, False,  True, False,  True,  True,  True,\n",
      "       False,  True, False, False, False, False, False, False, False,\n",
      "       False,  True,  True, False,  True, False, False,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True, False,\n",
      "        True,  True, False,  True,  True, False,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.30522406, 0.30522406, 0.30522406, ..., 0.30522406, 0.30522406,\n",
      "        0.30522406],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.39154044, 0.39154044, 0.39154044, ..., 0.39154044, 0.39154044,\n",
      "        0.39154044]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True, False,  True, False,  True,  True, False, False,\n",
      "       False,  True, False,  True, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False, False,\n",
      "        True, False,  True, False, False, False,  True,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False, False,  True,\n",
      "       False, False,  True, False,  True, False,  True,  True,  True,\n",
      "       False,  True,  True, False, False, False, False, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True, False,\n",
      "        True,  True, False,  True,  True, False,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.3536634 , 0.3536634 , 0.3536634 , ..., 0.3536634 , 0.3536634 ,\n",
      "        0.3536634 ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.39212692, 0.39212692, 0.39212692, ..., 0.39212692, 0.39212692,\n",
      "        0.39212692]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True, False,  True, False,  True,  True, False, False,\n",
      "       False,  True, False,  True, False,  True, False,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False,  True, False, False, False,  True,  True, False,\n",
      "        True,  True,  True,  True, False, False, False, False,  True,\n",
      "       False, False,  True, False,  True, False,  True,  True,  True,\n",
      "       False,  True, False, False, False, False, False, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True, False,\n",
      "        True,  True,  True, False,  True, False, False,  True, False,\n",
      "        True,  True, False,  True,  True,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.35829112, 0.35829112, 0.35829112, ..., 0.35829112, 0.35829112,\n",
      "        0.35829112],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.23123246, 0.23123246, 0.23123246, ..., 0.23123246, 0.23123246,\n",
      "        0.23123246]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True, False,  True, False,  True,  True, False, False,\n",
      "       False,  True, False,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True,  True, False, False, False, False,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True, False, False, False,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True, False,\n",
      "        True,  True,  True, False,  True, False, False,  True,  True,\n",
      "        True,  True, False,  True,  True,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.14139801, 0.14139801, 0.14139801, ..., 0.14139801, 0.14139801,\n",
      "        0.14139801]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True, False,  True, False,  True, False, False, False,\n",
      "       False,  True, False,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True,  True, False, False, False, False,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True, False, False, False,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True, False,\n",
      "        True,  True,  True, False,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.15358686, 0.15358686, 0.15358686, ..., 0.15358686, 0.15358686,\n",
      "        0.15358686]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "       False,  True, False, False, False,  True, False, False, False,\n",
      "        True,  True, False,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False,  True, False, False,  True,  True,  True,  True,\n",
      "        True,  True,  True, False, False, False, False, False,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "       False,  True,  True, False, False, False,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True,  True,\n",
      "        True,  True,  True, False,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.14639837, 0.14639837, 0.14639837, ..., 0.14639837, 0.14639837,\n",
      "        0.14639837]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False, False,  True,  True,  True,  True,\n",
      "       False,  True, False, False, False,  True, False, False,  True,\n",
      "        True,  True, False,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False,  True, False, False,  True,  True,  True,  True,\n",
      "       False,  True,  True,  True, False, False, False,  True,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "       False,  True,  True, False, False, False,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.20456332, 0.20456332, 0.20456332, ..., 0.20456332, 0.20456332,\n",
      "        0.20456332]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False, False,  True,  True,  True,  True,\n",
      "       False,  True, False, False,  True,  True, False, False,  True,\n",
      "        True,  True, False,  True,  True,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False,  True, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True, False, False, False, False,  True,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True, False, False, False,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.3764569 , 0.3764569 , 0.3764569 , ..., 0.3764569 , 0.3764569 ,\n",
      "        0.3764569 ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.22221577, 0.22221577, 0.22221577, ..., 0.22221577, 0.22221577,\n",
      "        0.22221577]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False, False,  True,  True,  True,  True,\n",
      "       False,  True, False, False,  True,  True, False, False,  True,\n",
      "        True,  True,  True,  True,  True,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False,  True, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True,  True, False, False, False,  True,  True,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True, False, False, False,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False,  True, False,\n",
      "        True,  True, False,  True,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.2662226 , 0.2662226 , 0.2662226 , ..., 0.2662226 , 0.2662226 ,\n",
      "        0.2662226 ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.25207174, 0.25207174, 0.25207174, ..., 0.25207174, 0.25207174,\n",
      "        0.25207174]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False,  True, False, False,  True,  True,  True,  True,\n",
      "       False,  True, False,  True,  True,  True, False, False, False,\n",
      "       False,  True,  True,  True,  True,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "        True, False, False, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True,  True,  True, False, False,  True, False,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True, False, False,  True,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False, False, False,\n",
      "        True,  True, False,  True,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True, False,  True,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.31785852, 0.31785852, 0.31785852, ..., 0.31785852, 0.31785852,\n",
      "        0.31785852],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.29898584, 0.29898584, 0.29898584, ..., 0.29898584, 0.29898584,\n",
      "        0.29898584]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False, False, False, False,  True,  True,  True,  True,\n",
      "       False,  True, False,  True,  True,  True, False, False, False,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True, False, False, False,\n",
      "        True, False, False, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True,  True,  True, False, False,  True, False,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True, False, False,  True,  True, False, False,\n",
      "       False,  True,  True, False, False, False, False, False, False,\n",
      "        True,  True, False,  True,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True, False, False,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.26435053, 0.26435053, 0.26435053, ..., 0.26435053, 0.26435053,\n",
      "        0.26435053],\n",
      "       [0.33781144, 0.33781144, 0.33781144, ..., 0.33781144, 0.33781144,\n",
      "        0.33781144],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.30723712, 0.30723712, 0.30723712, ..., 0.30723712, 0.30723712,\n",
      "        0.30723712]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False, False, False, False,  True,  True,  True,  True,\n",
      "       False,  True, False,  True,  True,  True, False, False, False,\n",
      "       False,  True,  True, False,  True,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True,  True, False, False,\n",
      "        True, False, False, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True,  True,  True, False, False,  True, False,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "       False,  True,  True, False, False, False,  True, False, False,\n",
      "       False,  True,  True,  True, False,  True, False, False, False,\n",
      "        True,  True, False,  True,  True, False, False,  True,  True,\n",
      "        True,  True,  True,  True, False, False,  True,  True,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.30324855, 0.30324855, 0.30324855, ..., 0.30324855, 0.30324855,\n",
      "        0.30324855],\n",
      "       [0.25751436, 0.25751436, 0.25751436, ..., 0.25751436, 0.25751436,\n",
      "        0.25751436],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.34611166, 0.34611166, 0.34611166, ..., 0.34611166, 0.34611166,\n",
      "        0.34611166]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False, False, False, False,  True,  True,  True,  True,\n",
      "       False,  True, False,  True,  True,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True,  True, False,  True,\n",
      "       False,  True,  True, False,  True,  True,  True, False, False,\n",
      "        True, False, False, False,  True,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True, False,  True,  True, False,\n",
      "       False, False,  True, False,  True,  True,  True,  True, False,\n",
      "       False,  True, False, False, False, False,  True, False, False,\n",
      "       False,  True,  True,  True, False, False, False, False, False,\n",
      "        True,  True, False,  True,  True, False, False,  True,  True,\n",
      "        True, False,  True,  True, False, False,  True, False,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.2983901 , 0.2983901 , 0.2983901 , ..., 0.2983901 , 0.2983901 ,\n",
      "        0.2983901 ],\n",
      "       [0.23150992, 0.23150992, 0.23150992, ..., 0.23150992, 0.23150992,\n",
      "        0.23150992],\n",
      "       ...,\n",
      "       [0.3564913 , 0.3564913 , 0.3564913 , ..., 0.3564913 , 0.3564913 ,\n",
      "        0.3564913 ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.2958321 , 0.2958321 , 0.2958321 , ..., 0.2958321 , 0.2958321 ,\n",
      "        0.2958321 ]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False, False, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False, False,  True,  True, False,  True,\n",
      "       False,  True,  True, False,  True,  True,  True, False, False,\n",
      "        True, False, False, False,  True,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True, False,  True,  True, False,\n",
      "       False,  True,  True, False,  True,  True,  True,  True, False,\n",
      "       False,  True,  True, False, False, False,  True, False, False,\n",
      "        True,  True,  True, False, False, False, False, False, False,\n",
      "        True,  True, False,  True,  True, False, False,  True, False,\n",
      "        True, False,  True,  True, False, False,  True, False,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.34823954, 0.34823954, 0.34823954, ..., 0.34823954, 0.34823954,\n",
      "        0.34823954],\n",
      "       [0.25255427, 0.25255427, 0.25255427, ..., 0.25255427, 0.25255427,\n",
      "        0.25255427],\n",
      "       ...,\n",
      "       [0.2788724 , 0.2788724 , 0.2788724 , ..., 0.2788724 , 0.2788724 ,\n",
      "        0.2788724 ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.29584515, 0.29584515, 0.29584515, ..., 0.29584515, 0.29584515,\n",
      "        0.29584515]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True, False, False, False,  True,  True,  True,  True, False,\n",
      "       False,  True,  True,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False, False,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True,  True, False, False,\n",
      "        True, False, False, False,  True,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True, False,  True,  True, False,\n",
      "       False,  True,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True, False, False, False,  True, False, False,\n",
      "        True,  True,  True, False, False,  True, False, False, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True, False,\n",
      "        True, False,  True,  True, False, False,  True, False,  True,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.38074422, 0.38074422, 0.38074422, ..., 0.38074422, 0.38074422,\n",
      "        0.38074422],\n",
      "       [0.2855162 , 0.2855162 , 0.2855162 , ..., 0.2855162 , 0.2855162 ,\n",
      "        0.2855162 ],\n",
      "       ...,\n",
      "       [0.2818961 , 0.2818961 , 0.2818961 , ..., 0.2818961 , 0.2818961 ,\n",
      "        0.2818961 ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.303846  , 0.303846  , 0.303846  , ..., 0.303846  , 0.303846  ,\n",
      "        0.303846  ]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True, False, False,  True,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False, False,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True, False,  True, False, False,\n",
      "        True, False, False, False, False,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True, False,  True,  True, False,\n",
      "       False,  True,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True, False, False, False, False,  True, False, False,\n",
      "        True,  True,  True, False, False,  True, False, False, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True, False,\n",
      "        True, False,  True,  True, False, False,  True, False, False,\n",
      "       False], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.34661806, 0.34661806, 0.34661806, ..., 0.34661806, 0.34661806,\n",
      "        0.34661806],\n",
      "       ...,\n",
      "       [0.3088773 , 0.3088773 , 0.3088773 , ..., 0.3088773 , 0.3088773 ,\n",
      "        0.3088773 ],\n",
      "       [0.37164894, 0.37164894, 0.37164894, ..., 0.37164894, 0.37164894,\n",
      "        0.37164894],\n",
      "       [0.38017794, 0.38017794, 0.38017794, ..., 0.38017794, 0.38017794,\n",
      "        0.38017794]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True,  True, False,\n",
      "        True,  True,  True,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False, False,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True,  True, False, False,\n",
      "        True, False,  True, False, False,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True, False,  True,  True, False,\n",
      "       False,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True, False, False, False,  True,  True, False, False,\n",
      "        True,  True,  True, False, False,  True, False, False, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True,  True,\n",
      "        True, False,  True,  True, False, False,  True, False, False,\n",
      "        True], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ],\n",
      "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ],\n",
      "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ],\n",
      "       ...,\n",
      "       [0.3852069, 0.3852069, 0.3852069, ..., 0.3852069, 0.3852069,\n",
      "        0.3852069],\n",
      "       [0.2589275, 0.2589275, 0.2589275, ..., 0.2589275, 0.2589275,\n",
      "        0.2589275],\n",
      "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
      "        0.       ]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True,  True, False,  True,  True,\n",
      "       False,  True,  True, False, False,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True,  True,  True, False, False,\n",
      "        True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "       False,  True,  True, False,  True, False,  True,  True, False,\n",
      "       False,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True, False, False, False,  True,  True, False, False,\n",
      "        True,  True,  True, False, False,  True, False, False, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True,  True,\n",
      "        True, False,  True,  True, False, False,  True,  True, False,\n",
      "        True], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.30585882, 0.30585882, 0.30585882, ..., 0.30585882, 0.30585882,\n",
      "        0.30585882],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ]], dtype=float32)\n",
      "  batch_dim = 0\n",
      "Traced<ShapedArray(bool[])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([ True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True,  True,  True,  True,  True, False,  True,  True,\n",
      "        True,  True,  True, False, False,  True, False, False,  True,\n",
      "       False,  True,  True, False,  True, False,  True, False, False,\n",
      "        True,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True,  True, False,  True, False, False,  True, False,\n",
      "       False,  True,  True, False,  True,  True,  True,  True,  True,\n",
      "        True,  True, False, False, False, False,  True, False, False,\n",
      "        True,  True,  True, False, False,  True, False, False, False,\n",
      "        True,  True,  True,  True,  True, False, False,  True,  True,\n",
      "        True, False,  True,  True, False, False,  True, False,  True,\n",
      "        True], dtype=bool)\n",
      "  batch_dim = 0 Traced<ShapedArray(float32[200])>with<BatchTrace(level=1/0)> with\n",
      "  val = Array([[0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       ...,\n",
      "       [0.31013155, 0.31013155, 0.31013155, ..., 0.31013155, 0.31013155,\n",
      "        0.31013155],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ],\n",
      "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
      "        0.        ]], dtype=float32)\n",
      "  batch_dim = 0\n"
     ]
    }
   ],
   "source": [
    "for ii in range(25):\n",
    "    state = jax.vmap(prob.step)(data, state)\n",
    "    state = jax.vmap(prob.update_rho)(data, state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "id": "3177162c-79c9-4120-a910-914cdfe8aebf",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Array([0, 0, 0, ..., 0, 0, 0], dtype=int32, weak_type=True),\n",
       " OSQPData(P=Array([[[ 1.        ,  0.29094326,  0.4284288 , -0.34115893,\n",
       "          -0.13439813],\n",
       "         [ 0.29094326,  1.        , -0.37179992,  0.2923568 ,\n",
       "           0.10243393],\n",
       "         [ 0.42842877, -0.37179995,  1.        , -0.4718293 ,\n",
       "           0.53118366],\n",
       "         [-0.34115896,  0.2923568 , -0.47182932,  0.9999999 ,\n",
       "           0.2352445 ],\n",
       "         [-0.13439815,  0.10243393,  0.53118366,  0.2352445 ,\n",
       "           1.        ]],\n",
       " \n",
       "        [[ 0.9999999 ,  0.31993186,  0.12916589, -0.8673883 ,\n",
       "           0.00755942],\n",
       "         [ 0.31993178,  0.5342728 , -0.32173502, -0.43970987,\n",
       "          -0.03762052],\n",
       "         [ 0.12916589, -0.3217351 ,  0.6585733 , -0.1761025 ,\n",
       "          -0.23804316],\n",
       "         [-0.8673883 , -0.43970987, -0.17610252,  0.9999999 ,\n",
       "           0.34611914],\n",
       "         [ 0.00755942, -0.03762053, -0.23804317,  0.34611917,\n",
       "           0.7202737 ]],\n",
       " \n",
       "        [[ 1.        ,  0.81315273, -0.02081607, -0.696986  ,\n",
       "          -0.6499888 ],\n",
       "         [ 0.8131528 ,  1.        ,  0.52887547, -0.30325064,\n",
       "          -0.47972485],\n",
       "         [-0.02081607,  0.52887547,  1.        ,  0.6119919 ,\n",
       "          -0.01262505],\n",
       "         [-0.6969859 , -0.30325055,  0.6119918 ,  0.9999999 ,\n",
       "           0.18316314],\n",
       "         [-0.6499888 , -0.4797249 , -0.01262505,  0.18316317,\n",
       "           1.        ]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[ 0.9999659 , -0.24518023, -0.7956639 ,  0.10509838,\n",
       "           0.31698224],\n",
       "         [-0.24518022,  0.9999659 ,  0.43681908,  0.22357923,\n",
       "          -0.29762137],\n",
       "         [-0.79566395,  0.43681908,  0.99996597, -0.467378  ,\n",
       "          -0.31369808],\n",
       "         [ 0.10509838,  0.22357923, -0.46737796,  0.9999659 ,\n",
       "           0.4038898 ],\n",
       "         [ 0.3169822 , -0.29762134, -0.31369808,  0.40388983,\n",
       "           0.999966  ]],\n",
       " \n",
       "        [[ 1.        , -0.4374399 , -0.24088432, -0.25732374,\n",
       "          -0.3241695 ],\n",
       "         [-0.43743992,  1.        ,  0.85643524, -0.10894669,\n",
       "          -0.11844584],\n",
       "         [-0.2408843 ,  0.8564352 ,  0.9999999 ,  0.29016516,\n",
       "          -0.19155811],\n",
       "         [-0.25732374, -0.10894667,  0.29016516,  0.9999999 ,\n",
       "           0.28801963],\n",
       "         [-0.3241695 , -0.11844584, -0.19155811,  0.28801963,\n",
       "           0.9999999 ]],\n",
       " \n",
       "        [[ 1.        ,  0.16587065,  0.2749946 , -0.43078205,\n",
       "          -0.5456807 ],\n",
       "         [ 0.16587065,  1.        ,  0.26467654,  0.18799438,\n",
       "           0.24996108],\n",
       "         [ 0.27499464,  0.2646765 ,  1.        ,  0.56462306,\n",
       "          -0.45662427],\n",
       "         [-0.43078208,  0.18799439,  0.5646232 ,  1.        ,\n",
       "          -0.2828055 ],\n",
       "         [-0.5456807 ,  0.24996108, -0.45662424, -0.28280553,\n",
       "           0.99999994]]], dtype=float32), q=Array([[ 0.05292495,  0.06467757,  0.62291723, -0.13877605, -0.86378783],\n",
       "        [ 0.20019338,  1.        ,  0.23693264,  0.01658657, -0.04954969],\n",
       "        [ 0.24112396,  0.04044091, -0.24376109, -0.7822807 ,  0.391238  ],\n",
       "        ...,\n",
       "        [-0.24642737,  0.56672686,  0.7900576 , -1.        ,  0.58952355],\n",
       "        [ 0.25101793,  0.8157204 , -0.4121838 ,  0.11778981, -0.763767  ],\n",
       "        [-0.2629209 ,  0.49513274,  0.7610736 , -0.31655452,  0.19575404]],      dtype=float32), A=Array([[[ 0.24536936, -0.1747192 ,  0.48736784,  0.99851036,\n",
       "          -0.49218485],\n",
       "         [ 0.97722286,  0.21870221, -0.9799229 ,  0.9981341 ,\n",
       "           0.44220614]],\n",
       " \n",
       "        [[ 0.22272494,  0.9999999 , -0.55324817, -0.6239239 ,\n",
       "           0.25659513],\n",
       "         [-0.12133946,  0.2995489 ,  1.        , -0.5669095 ,\n",
       "           0.9999124 ]],\n",
       " \n",
       "        [[-0.02066166, -0.9996262 ,  0.32869336, -0.19786364,\n",
       "           0.10172367],\n",
       "         [-0.00154104, -0.9988116 , -0.83417815, -0.1285774 ,\n",
       "          -0.45431116]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[-0.9985836 , -0.54913986, -0.577118  ,  0.12692778,\n",
       "          -0.6717778 ],\n",
       "         [-0.17959072, -0.02802322, -0.06271502,  0.9993378 ,\n",
       "           0.48143715]],\n",
       " \n",
       "        [[ 0.9991736 , -0.4617665 ,  0.14193504,  0.21468943,\n",
       "          -0.07306388],\n",
       "         [ 0.2361117 ,  0.680679  ,  0.03063584, -0.99816   ,\n",
       "          -0.22197717]],\n",
       " \n",
       "        [[ 0.10928988,  0.86254823,  0.9987163 ,  0.12210178,\n",
       "           0.1501691 ],\n",
       "         [ 0.9982928 , -0.26090524,  0.03786737, -0.9039203 ,\n",
       "          -0.78953576]]], dtype=float32), l=Array([[-1.22731   , -0.11908621],\n",
       "        [-0.37638894, -0.36246392],\n",
       "        [-0.34790877, -1.1731938 ],\n",
       "        ...,\n",
       "        [-0.60085094, -0.45105064],\n",
       "        [-0.40721533, -0.7906791 ],\n",
       "        [-1.031179  , -0.46448508]], dtype=float32), u=Array([[0.7966866 , 0.5497297 ],\n",
       "        [0.4555999 , 0.07994474],\n",
       "        [0.5286612 , 0.2612425 ],\n",
       "        ...,\n",
       "        [1.0012264 , 0.38866115],\n",
       "        [1.0094485 , 0.85047394],\n",
       "        [0.97690284, 1.0118961 ]], dtype=float32), D=Array([[0.4007447 , 0.30443323, 0.44927934, 0.3000256 , 0.52503806],\n",
       "        [0.6630044 , 0.8390461 , 0.5171017 , 0.7468656 , 0.58463943],\n",
       "        [0.336619  , 0.5459523 , 0.31448156, 0.49925548, 0.47010162],\n",
       "        ...,\n",
       "        [0.37495217, 0.35561255, 0.67569715, 0.45578146, 0.6181835 ],\n",
       "        [0.4552195 , 0.5278139 , 0.3716473 , 0.2896845 , 0.3266104 ],\n",
       "        [0.2779031 , 0.33022705, 0.7127735 , 0.35951748, 0.6319413 ]],      dtype=float32), E=Array([[1.3786567, 2.027422 ],\n",
       "        [1.1916087, 0.9052421],\n",
       "        [0.9974981, 1.5840542],\n",
       "        ...,\n",
       "        [2.1123042, 1.1859759],\n",
       "        [1.0228131, 1.8385028],\n",
       "        [1.4970213, 1.499928 ]], dtype=float32), c=Array([1.      , 1.      , 1.      , ..., 0.999966, 1.      , 1.      ],      dtype=float32)),\n",
       " OSQPState(x=Array([[ -0.05501798,  -0.49047273,  -0.75690067,  -0.45751056,\n",
       "           0.9625162 ],\n",
       "        [-24.281647  , -42.62691   , -27.510513  , -47.73124   ,\n",
       "          10.35473   ],\n",
       "        [  1.4127473 ,  -0.2057909 ,  -0.48936078,   1.949922  ,\n",
       "           0.2191805 ],\n",
       "        ...,\n",
       "        [  2.6936488 ,  -2.474353  ,   2.700552  ,   2.3975205 ,\n",
       "          -2.9545927 ],\n",
       "        [ -7.541529  , -19.616177  ,  20.12886   , -14.520721  ,\n",
       "           3.461642  ],\n",
       "        [  5.164899  ,  -1.6517391 ,  -1.1234329 ,   3.8530853 ,\n",
       "           3.1991916 ]], dtype=float32), z=Array([[-1.22731   ,  0.5497297 ],\n",
       "        [-0.37638894,  0.07994474],\n",
       "        [-0.34790877,  0.2612425 ],\n",
       "        ...,\n",
       "        [-0.60085094,  0.38866115],\n",
       "        [ 1.0094485 , -0.7906791 ],\n",
       "        [-1.031179  , -0.46448508]], dtype=float32), chi=Array([[ -0.05498462,  -0.4905034 ,  -0.75695735,  -0.45752653,\n",
       "           0.9625589 ],\n",
       "        [-24.281794  , -42.62716   , -27.510563  , -47.73149   ,\n",
       "          10.354821  ],\n",
       "        [  1.4127568 ,  -0.20574978,  -0.489415  ,   1.9499726 ,\n",
       "           0.21919468],\n",
       "        ...,\n",
       "        [  2.6936674 ,  -2.4744701 ,   2.7006583 ,   2.3976943 ,\n",
       "          -2.9546552 ],\n",
       "        [ -7.541531  , -19.616173  ,  20.128855  , -14.5207205 ,\n",
       "           3.4616408 ],\n",
       "        [  5.1649623 ,  -1.65177   ,  -1.1234708 ,   3.853135  ,\n",
       "           3.1992185 ]], dtype=float32), zeta=Array([[-1.22731   ,  0.5497297 ],\n",
       "        [-0.3773298 ,  0.08020581],\n",
       "        [-0.34791592,  0.26128274],\n",
       "        ...,\n",
       "        [-0.60041445,  0.38983774],\n",
       "        [ 1.0094151 , -0.79070055],\n",
       "        [-1.0313816 , -0.46426752]], dtype=float32), y=Array([[-0.4204607 ,  0.5582306 ],\n",
       "        [-0.2697917 ,  1.2125616 ],\n",
       "        [-0.25412345,  0.2825214 ],\n",
       "        ...,\n",
       "        [-0.01865726,  1.3314339 ],\n",
       "        [ 1.6847588 , -3.1315465 ],\n",
       "        [-1.3079455 , -0.7718598 ]], dtype=float32), dx=Array([[-8.89450312e-05,  8.18371773e-05,  1.51276588e-04,\n",
       "          4.26471233e-05, -1.13904476e-04],\n",
       "        [ 3.92913818e-04,  6.56127930e-04,  1.33514404e-04,\n",
       "          6.63757324e-04, -2.45094299e-04],\n",
       "        [-2.56299973e-05, -1.09672546e-04,  1.44571066e-04,\n",
       "         -1.35064125e-04, -3.78340483e-05],\n",
       "        ...,\n",
       "        [-4.95910645e-05,  3.12089920e-04, -2.83718109e-04,\n",
       "         -4.63485718e-04,  1.66893005e-04],\n",
       "        [ 5.72204590e-06, -7.62939453e-06,  1.14440918e-05,\n",
       "         -3.81469727e-06,  3.33786011e-06],\n",
       "        [-1.68800354e-04,  8.24928284e-05,  1.01208687e-04,\n",
       "         -1.32560730e-04, -7.15255737e-05]], dtype=float32), dy=Array([[ 0.0000000e+00,  0.0000000e+00],\n",
       "        [-1.5056133e-04,  4.1723251e-05],\n",
       "        [-1.1324883e-06,  6.4373016e-06],\n",
       "        ...,\n",
       "        [ 6.9838017e-05,  1.8823147e-04],\n",
       "        [-5.3644180e-06, -3.5762787e-06],\n",
       "        [-3.2424927e-05,  3.4809113e-05]], dtype=float32), converged=Array([ True,  True,  True, ...,  True,  True,  True], dtype=bool), primal_infeas=Array([ True, False, False, ..., False, False, False], dtype=bool), dual_infeas=Array([False, False, False, ..., False, False, False], dtype=bool), kkt_mat=Array([[[ 1.00000095e+00,  2.90943265e-01,  4.28428799e-01, ...,\n",
       "          -1.34398133e-01,  2.45369360e-01,  9.77222860e-01],\n",
       "         [ 2.90943265e-01,  1.00000095e+00, -3.71799916e-01, ...,\n",
       "           1.02433927e-01, -1.74719200e-01,  2.18702212e-01],\n",
       "         [ 4.28428769e-01, -3.71799946e-01,  1.00000095e+00, ...,\n",
       "           5.31183660e-01,  4.87367839e-01, -9.79922891e-01],\n",
       "         ...,\n",
       "         [-1.34398147e-01,  1.02433927e-01,  5.31183660e-01, ...,\n",
       "           1.00000095e+00, -4.92184848e-01,  4.42206144e-01],\n",
       "         [ 2.45369360e-01, -1.74719200e-01,  4.87367839e-01, ...,\n",
       "          -4.92184848e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [ 9.77222860e-01,  2.18702212e-01, -9.79922891e-01, ...,\n",
       "           4.42206144e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000083e+00,  3.19931865e-01,  1.29165888e-01, ...,\n",
       "           7.55941868e-03,  2.22724944e-01, -1.21339463e-01],\n",
       "         [ 3.19931775e-01,  5.34273803e-01, -3.21735024e-01, ...,\n",
       "          -3.76205221e-02,  9.99999881e-01,  2.99548894e-01],\n",
       "         [ 1.29165888e-01, -3.21735114e-01,  6.58574343e-01, ...,\n",
       "          -2.38043159e-01, -5.53248167e-01,  1.00000000e+00],\n",
       "         ...,\n",
       "         [ 7.55941775e-03, -3.76205258e-02, -2.38043174e-01, ...,\n",
       "           7.20274687e-01,  2.56595135e-01,  9.99912381e-01],\n",
       "         [ 2.22724944e-01,  9.99999881e-01, -5.53248167e-01, ...,\n",
       "           2.56595135e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [-1.21339463e-01,  2.99548894e-01,  1.00000000e+00, ...,\n",
       "           9.99912381e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000095e+00,  8.13152730e-01, -2.08160710e-02, ...,\n",
       "          -6.49988770e-01, -2.06616633e-02, -1.54104165e-03],\n",
       "         [ 8.13152790e-01,  1.00000095e+00,  5.28875470e-01, ...,\n",
       "          -4.79724854e-01, -9.99626219e-01, -9.98811603e-01],\n",
       "         [-2.08160710e-02,  5.28875470e-01,  1.00000095e+00, ...,\n",
       "          -1.26250526e-02,  3.28693360e-01, -8.34178150e-01],\n",
       "         ...,\n",
       "         [-6.49988770e-01, -4.79724914e-01, -1.26250517e-02, ...,\n",
       "           1.00000095e+00,  1.01723671e-01, -4.54311162e-01],\n",
       "         [-2.06616633e-02, -9.99626219e-01,  3.28693360e-01, ...,\n",
       "           1.01723671e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [-1.54104165e-03, -9.98811603e-01, -8.34178150e-01, ...,\n",
       "          -4.54311162e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[ 9.99966919e-01, -2.45180234e-01, -7.95663893e-01, ...,\n",
       "           3.16982239e-01, -9.98583615e-01, -1.79590717e-01],\n",
       "         [-2.45180219e-01,  9.99966919e-01,  4.36819077e-01, ...,\n",
       "          -2.97621369e-01, -5.49139857e-01, -2.80232187e-02],\n",
       "         [-7.95663953e-01,  4.36819077e-01,  9.99966979e-01, ...,\n",
       "          -3.13698083e-01, -5.77117980e-01, -6.27150163e-02],\n",
       "         ...,\n",
       "         [ 3.16982210e-01, -2.97621340e-01, -3.13698083e-01, ...,\n",
       "           9.99967039e-01, -6.71777785e-01,  4.81437147e-01],\n",
       "         [-9.98583615e-01, -5.49139857e-01, -5.77117980e-01, ...,\n",
       "          -6.71777785e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [-1.79590717e-01, -2.80232187e-02, -6.27150163e-02, ...,\n",
       "           4.81437147e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000095e+00, -4.37439889e-01, -2.40884319e-01, ...,\n",
       "          -3.24169487e-01,  9.99173582e-01,  2.36111701e-01],\n",
       "         [-4.37439919e-01,  1.00000095e+00,  8.56435239e-01, ...,\n",
       "          -1.18445843e-01, -4.61766511e-01,  6.80679023e-01],\n",
       "         [-2.40884304e-01,  8.56435180e-01,  1.00000083e+00, ...,\n",
       "          -1.91558108e-01,  1.41935036e-01,  3.06358375e-02],\n",
       "         ...,\n",
       "         [-3.24169487e-01, -1.18445843e-01, -1.91558108e-01, ...,\n",
       "           1.00000083e+00, -7.30638802e-02, -2.21977174e-01],\n",
       "         [ 9.99173582e-01, -4.61766511e-01,  1.41935036e-01, ...,\n",
       "          -7.30638802e-02, -1.00000000e+01, -0.00000000e+00],\n",
       "         [ 2.36111701e-01,  6.80679023e-01,  3.06358375e-02, ...,\n",
       "          -2.21977174e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000095e+00,  1.65870652e-01,  2.74994612e-01, ...,\n",
       "          -5.45680702e-01,  1.09289885e-01,  9.98292804e-01],\n",
       "         [ 1.65870652e-01,  1.00000095e+00,  2.64676541e-01, ...,\n",
       "           2.49961078e-01,  8.62548232e-01, -2.60905236e-01],\n",
       "         [ 2.74994642e-01,  2.64676511e-01,  1.00000095e+00, ...,\n",
       "          -4.56624269e-01,  9.98716295e-01,  3.78673710e-02],\n",
       "         ...,\n",
       "         [-5.45680702e-01,  2.49961078e-01, -4.56624240e-01, ...,\n",
       "           1.00000095e+00,  1.50169104e-01, -7.89535761e-01],\n",
       "         [ 1.09289885e-01,  8.62548232e-01,  9.98716295e-01, ...,\n",
       "           1.50169104e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [ 9.98292804e-01, -2.60905236e-01,  3.78673710e-02, ...,\n",
       "          -7.89535761e-01, -0.00000000e+00, -1.00000000e+01]]],      dtype=float32), kkt_lu=(Array([[[ 1.00000095e+00,  2.90943265e-01,  4.28428799e-01, ...,\n",
       "          -1.34398133e-01,  2.45369360e-01,  9.77222860e-01],\n",
       "         [ 2.90942997e-01,  9.15353060e-01, -4.96448278e-01, ...,\n",
       "           1.41536117e-01, -2.46107697e-01, -6.56139329e-02],\n",
       "         [ 9.77221906e-01, -7.16815367e-02, -1.43417907e+00, ...,\n",
       "           5.83688498e-01, -2.57421672e-01, -1.09596672e+01],\n",
       "         ...,\n",
       "         [-1.34398013e-01,  1.54624626e-01, -4.64047223e-01, ...,\n",
       "           1.40167451e+00,  4.88851500e+00, -3.34997702e+00],\n",
       "         [-3.41158628e-01,  4.27828968e-01,  7.89805055e-02, ...,\n",
       "           1.56632677e-01,  4.79173994e+00,  3.67295647e+00],\n",
       "         [ 4.28428352e-01, -5.42357147e-01, -3.81540626e-01, ...,\n",
       "           6.98707461e-01, -7.67292008e-02, -2.37838531e+00]],\n",
       " \n",
       "        [[ 1.00000083e+00,  3.19931865e-01,  1.29165888e-01, ...,\n",
       "           7.55941868e-03,  2.22724944e-01, -1.21339463e-01],\n",
       "         [ 2.22724766e-01,  9.28743124e-01, -5.82016587e-01, ...,\n",
       "           2.54911453e-01, -1.00496063e+01,  2.70253029e-02],\n",
       "         [-1.21339358e-01,  3.64330232e-01,  1.22771919e+00, ...,\n",
       "           9.07957613e-01,  3.68840075e+00, -1.00245686e+01],\n",
       "         ...,\n",
       "         [ 1.29165784e-01, -3.90914708e-01,  3.37513387e-01, ...,\n",
       "          -2.02846393e-01, -5.59413052e+00,  4.10690260e+00],\n",
       "         [ 3.19931507e-01,  4.65055943e-01, -7.52525330e-02, ...,\n",
       "           4.31442350e-01,  8.29530048e+00, -2.20388460e+00],\n",
       "         [-8.67387593e-01, -1.74649939e-01, -1.34977758e-01, ...,\n",
       "          -4.57836479e-01, -5.46411693e-01, -8.12546968e-01]],\n",
       " \n",
       "        [[ 1.00000095e+00,  8.13152730e-01, -2.08160710e-02, ...,\n",
       "          -6.49988770e-01, -2.06616633e-02, -1.54104165e-03],\n",
       "         [-1.54104014e-03, -9.97558475e-01, -8.34210217e-01, ...,\n",
       "          -4.55312818e-01, -3.18404527e-05, -1.00000019e+01],\n",
       "         [-2.06616428e-02,  9.85230625e-01,  1.15015268e+00, ...,\n",
       "           5.36881983e-01, -1.00003958e+01,  9.85227585e+00],\n",
       "         ...,\n",
       "         [-6.49988174e-01, -4.89342175e-02, -5.82328551e-02, ...,\n",
       "           3.24052036e-01,  2.01263809e+00, -5.80947495e+00],\n",
       "         [-6.96985245e-01, -2.64149815e-01,  3.27892870e-01, ...,\n",
       "          -2.84668982e-01, -8.86914432e-01,  2.16559815e+00],\n",
       "         [ 8.13152015e-01, -3.39613318e-01,  2.28224680e-01, ...,\n",
       "          -1.68248769e-02,  8.96972954e-01, -4.06357288e+00]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[ 9.99966919e-01, -2.45180234e-01, -7.95663893e-01, ...,\n",
       "           3.16982239e-01, -9.98583615e-01, -1.79590717e-01],\n",
       "         [-2.45188341e-01,  9.39851582e-01,  2.41731569e-01, ...,\n",
       "          -2.19901025e-01, -7.93980896e-01, -7.20567703e-02],\n",
       "         [-9.98616695e-01, -8.44793975e-01, -1.16746783e+00, ...,\n",
       "          -5.41005075e-01, -1.16679525e+01, -2.40215406e-01],\n",
       "         ...,\n",
       "         [ 3.16992700e-01, -2.33974159e-01,  4.21381649e-03, ...,\n",
       "           5.81556797e-01, -1.21190405e+00,  4.94110823e+00],\n",
       "         [ 1.05101861e-01,  2.65305847e-01,  3.83637965e-01, ...,\n",
       "           2.79557884e-01,  3.98794055e+00,  7.53923225e+00],\n",
       "         [-7.95690298e-01,  2.57201821e-01, -2.60984600e-01, ...,\n",
       "           1.08454101e-01, -8.82844210e-01,  2.43113351e+00]],\n",
       " \n",
       "        [[ 1.00000095e+00, -4.37439889e-01, -2.40884319e-01, ...,\n",
       "          -3.24169487e-01,  9.99173582e-01,  2.36111701e-01],\n",
       "         [-4.37439501e-01,  8.08647454e-01,  7.51062930e-01, ...,\n",
       "          -2.60250390e-01, -2.46885177e-02,  7.83963621e-01],\n",
       "         [ 2.36111477e-01,  9.69475150e-01, -6.40625477e-01, ...,\n",
       "           1.06869251e-01, -2.11981446e-01, -1.08157825e+01],\n",
       "         ...,\n",
       "         [-3.24169189e-01, -3.21834177e-01,  4.35943827e-02, ...,\n",
       "           7.18125105e-01,  1.14027746e-01,  4.03622484e+00],\n",
       "         [ 9.99172628e-01, -3.05306688e-02, -6.33054018e-01, ...,\n",
       "           4.26796257e-01, -1.11882963e+01, -8.62303066e+00],\n",
       "         [-2.40884081e-01,  9.28789020e-01, -3.81496936e-01, ...,\n",
       "          -1.00256570e-01, -1.81913376e-02, -1.19945824e+00]],\n",
       " \n",
       "        [[ 1.00000095e+00,  1.65870652e-01,  2.74994612e-01, ...,\n",
       "          -5.45680702e-01,  1.09289885e-01,  9.98292804e-01],\n",
       "         [ 1.65870488e-01,  9.72487926e-01,  2.19063044e-01, ...,\n",
       "           3.40473413e-01,  8.44420254e-01, -4.26492542e-01],\n",
       "         [ 2.74994373e-01,  2.25260392e-01,  8.75032783e-01, ...,\n",
       "          -3.83260310e-01,  7.78447747e-01, -1.40585646e-01],\n",
       "         ...,\n",
       "         [-5.45680165e-01,  3.50105554e-01, -4.37995344e-01, ...,\n",
       "           2.75416821e-01,  6.52027321e+00, -3.68642360e-01],\n",
       "         [ 9.98291850e-01, -4.38558191e-01, -1.60663247e-01, ...,\n",
       "          -9.63478506e-01,  1.15242376e+01, -1.17253933e+01],\n",
       "         [-4.30781662e-01,  2.66788334e-01,  7.13850141e-01, ...,\n",
       "          -7.63579309e-01, -1.06620990e-01, -1.60240817e+00]]],      dtype=float32), Array([[0, 1, 6, ..., 4, 5, 6],\n",
       "        [0, 5, 6, ..., 6, 5, 6],\n",
       "        [0, 6, 5, ..., 4, 5, 6],\n",
       "        ...,\n",
       "        [0, 1, 5, ..., 4, 6, 6],\n",
       "        [0, 1, 6, ..., 4, 5, 6],\n",
       "        [0, 1, 2, ..., 4, 6, 6]], dtype=int32)), rho=Array([[0.1, 0.1],\n",
       "        [0.1, 0.1],\n",
       "        [0.1, 0.1],\n",
       "        ...,\n",
       "        [0.1, 0.1],\n",
       "        [0.1, 0.1],\n",
       "        [0.1, 0.1]], dtype=float32, weak_type=True)))"
      ]
     },
     "execution_count": 307,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# %time sol = jax.vmap(prob.solve)(data, state)\n",
    "sol_jitd = jax.jit(jax.vmap(prob.solve))\n",
    "sol_sparse_jitd = jax.jit(jax.vmap(sparse.sparsify(prob.solve)))\n",
    "\n",
    "sol_jitd(data, state)\n",
    "sol_sparse_jitd(data, state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "id": "05f00db2-4602-4d43-b019-be5b560f0201",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 3.81 s, sys: 0 ns, total: 3.81 s\n",
      "Wall time: 3.81 s\n",
      "CPU times: user 5.36 s, sys: 16 ms, total: 5.38 s\n",
      "Wall time: 4.4 s\n"
     ]
    }
   ],
   "source": [
    "prob, data, state = osqp.OSQPProblem.from_data(P, q, A, l, u)\n",
    "%time state = jax.vmap(prob.solve)(data, state)[-1]\n",
    "%time state = jax.vmap(prob.solve)(data, state)[-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "id": "26b23f0a-3fdf-49bb-aedb-165698bb9b7e",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 120 ms, sys: 0 ns, total: 120 ms\n",
      "Wall time: 120 ms\n",
      "CPU times: user 118 ms, sys: 27 µs, total: 118 ms\n",
      "Wall time: 118 ms\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(Array([26, 26, 26, ..., 26, 51, 26], dtype=int32, weak_type=True),\n",
       " OSQPData(P=Array([[[ 1.        ,  0.29094326,  0.4284288 , -0.34115893,\n",
       "          -0.13439813],\n",
       "         [ 0.29094326,  1.        , -0.37179992,  0.2923568 ,\n",
       "           0.10243393],\n",
       "         [ 0.42842877, -0.37179995,  1.        , -0.4718293 ,\n",
       "           0.53118366],\n",
       "         [-0.34115896,  0.2923568 , -0.47182932,  0.9999999 ,\n",
       "           0.2352445 ],\n",
       "         [-0.13439815,  0.10243393,  0.53118366,  0.2352445 ,\n",
       "           1.        ]],\n",
       " \n",
       "        [[ 0.9999999 ,  0.31993186,  0.12916589, -0.8673883 ,\n",
       "           0.00755942],\n",
       "         [ 0.31993178,  0.5342728 , -0.32173502, -0.43970987,\n",
       "          -0.03762052],\n",
       "         [ 0.12916589, -0.3217351 ,  0.6585733 , -0.1761025 ,\n",
       "          -0.23804316],\n",
       "         [-0.8673883 , -0.43970987, -0.17610252,  0.9999999 ,\n",
       "           0.34611914],\n",
       "         [ 0.00755942, -0.03762053, -0.23804317,  0.34611917,\n",
       "           0.7202737 ]],\n",
       " \n",
       "        [[ 1.        ,  0.81315273, -0.02081607, -0.696986  ,\n",
       "          -0.6499888 ],\n",
       "         [ 0.8131528 ,  1.        ,  0.52887547, -0.30325064,\n",
       "          -0.47972485],\n",
       "         [-0.02081607,  0.52887547,  1.        ,  0.6119919 ,\n",
       "          -0.01262505],\n",
       "         [-0.6969859 , -0.30325055,  0.6119918 ,  0.9999999 ,\n",
       "           0.18316314],\n",
       "         [-0.6499888 , -0.4797249 , -0.01262505,  0.18316317,\n",
       "           1.        ]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[ 0.9999659 , -0.24518023, -0.7956639 ,  0.10509838,\n",
       "           0.31698224],\n",
       "         [-0.24518022,  0.9999659 ,  0.43681908,  0.22357923,\n",
       "          -0.29762137],\n",
       "         [-0.79566395,  0.43681908,  0.99996597, -0.467378  ,\n",
       "          -0.31369808],\n",
       "         [ 0.10509838,  0.22357923, -0.46737796,  0.9999659 ,\n",
       "           0.4038898 ],\n",
       "         [ 0.3169822 , -0.29762134, -0.31369808,  0.40388983,\n",
       "           0.999966  ]],\n",
       " \n",
       "        [[ 1.        , -0.4374399 , -0.24088432, -0.25732374,\n",
       "          -0.3241695 ],\n",
       "         [-0.43743992,  1.        ,  0.85643524, -0.10894669,\n",
       "          -0.11844584],\n",
       "         [-0.2408843 ,  0.8564352 ,  0.9999999 ,  0.29016516,\n",
       "          -0.19155811],\n",
       "         [-0.25732374, -0.10894667,  0.29016516,  0.9999999 ,\n",
       "           0.28801963],\n",
       "         [-0.3241695 , -0.11844584, -0.19155811,  0.28801963,\n",
       "           0.9999999 ]],\n",
       " \n",
       "        [[ 1.        ,  0.16587065,  0.2749946 , -0.43078205,\n",
       "          -0.5456807 ],\n",
       "         [ 0.16587065,  1.        ,  0.26467654,  0.18799438,\n",
       "           0.24996108],\n",
       "         [ 0.27499464,  0.2646765 ,  1.        ,  0.56462306,\n",
       "          -0.45662427],\n",
       "         [-0.43078208,  0.18799439,  0.5646232 ,  1.        ,\n",
       "          -0.2828055 ],\n",
       "         [-0.5456807 ,  0.24996108, -0.45662424, -0.28280553,\n",
       "           0.99999994]]], dtype=float32), q=Array([[ 0.05292495,  0.06467757,  0.62291723, -0.13877605, -0.86378783],\n",
       "        [ 0.20019338,  1.        ,  0.23693264,  0.01658657, -0.04954969],\n",
       "        [ 0.24112396,  0.04044091, -0.24376109, -0.7822807 ,  0.391238  ],\n",
       "        ...,\n",
       "        [-0.24642737,  0.56672686,  0.7900576 , -1.        ,  0.58952355],\n",
       "        [ 0.25101793,  0.8157204 , -0.4121838 ,  0.11778981, -0.763767  ],\n",
       "        [-0.2629209 ,  0.49513274,  0.7610736 , -0.31655452,  0.19575404]],      dtype=float32), A=Array([[[ 0.24536936, -0.1747192 ,  0.48736784,  0.99851036,\n",
       "          -0.49218485],\n",
       "         [ 0.97722286,  0.21870221, -0.9799229 ,  0.9981341 ,\n",
       "           0.44220614]],\n",
       " \n",
       "        [[ 0.22272494,  0.9999999 , -0.55324817, -0.6239239 ,\n",
       "           0.25659513],\n",
       "         [-0.12133946,  0.2995489 ,  1.        , -0.5669095 ,\n",
       "           0.9999124 ]],\n",
       " \n",
       "        [[-0.02066166, -0.9996262 ,  0.32869336, -0.19786364,\n",
       "           0.10172367],\n",
       "         [-0.00154104, -0.9988116 , -0.83417815, -0.1285774 ,\n",
       "          -0.45431116]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[-0.9985836 , -0.54913986, -0.577118  ,  0.12692778,\n",
       "          -0.6717778 ],\n",
       "         [-0.17959072, -0.02802322, -0.06271502,  0.9993378 ,\n",
       "           0.48143715]],\n",
       " \n",
       "        [[ 0.9991736 , -0.4617665 ,  0.14193504,  0.21468943,\n",
       "          -0.07306388],\n",
       "         [ 0.2361117 ,  0.680679  ,  0.03063584, -0.99816   ,\n",
       "          -0.22197717]],\n",
       " \n",
       "        [[ 0.10928988,  0.86254823,  0.9987163 ,  0.12210178,\n",
       "           0.1501691 ],\n",
       "         [ 0.9982928 , -0.26090524,  0.03786737, -0.9039203 ,\n",
       "          -0.78953576]]], dtype=float32), l=Array([[-1.22731   , -0.11908621],\n",
       "        [-0.37638894, -0.36246392],\n",
       "        [-0.34790877, -1.1731938 ],\n",
       "        ...,\n",
       "        [-0.60085094, -0.45105064],\n",
       "        [-0.40721533, -0.7906791 ],\n",
       "        [-1.031179  , -0.46448508]], dtype=float32), u=Array([[0.7966866 , 0.5497297 ],\n",
       "        [0.4555999 , 0.07994474],\n",
       "        [0.5286612 , 0.2612425 ],\n",
       "        ...,\n",
       "        [1.0012264 , 0.38866115],\n",
       "        [1.0094485 , 0.85047394],\n",
       "        [0.97690284, 1.0118961 ]], dtype=float32), D=Array([[0.4007447 , 0.30443323, 0.44927934, 0.3000256 , 0.52503806],\n",
       "        [0.6630044 , 0.8390461 , 0.5171017 , 0.7468656 , 0.58463943],\n",
       "        [0.336619  , 0.5459523 , 0.31448156, 0.49925548, 0.47010162],\n",
       "        ...,\n",
       "        [0.37495217, 0.35561255, 0.67569715, 0.45578146, 0.6181835 ],\n",
       "        [0.4552195 , 0.5278139 , 0.3716473 , 0.2896845 , 0.3266104 ],\n",
       "        [0.2779031 , 0.33022705, 0.7127735 , 0.35951748, 0.6319413 ]],      dtype=float32), E=Array([[1.3786567, 2.027422 ],\n",
       "        [1.1916087, 0.9052421],\n",
       "        [0.9974981, 1.5840542],\n",
       "        ...,\n",
       "        [2.1123042, 1.1859759],\n",
       "        [1.0228131, 1.8385028],\n",
       "        [1.4970213, 1.499928 ]], dtype=float32), c=Array([1.      , 1.      , 1.      , ..., 0.999966, 1.      , 1.      ],      dtype=float32)),\n",
       " OSQPState(x=Array([[ -0.05501798,  -0.49047273,  -0.75690067,  -0.45751056,\n",
       "           0.9625162 ],\n",
       "        [-24.281647  , -42.62691   , -27.510513  , -47.73124   ,\n",
       "          10.35473   ],\n",
       "        [  1.4127473 ,  -0.2057909 ,  -0.48936078,   1.949922  ,\n",
       "           0.2191805 ],\n",
       "        ...,\n",
       "        [  2.6936488 ,  -2.474353  ,   2.700552  ,   2.3975205 ,\n",
       "          -2.9545927 ],\n",
       "        [ -7.541529  , -19.616177  ,  20.12886   , -14.520721  ,\n",
       "           3.461642  ],\n",
       "        [  5.164899  ,  -1.6517391 ,  -1.1234329 ,   3.8530853 ,\n",
       "           3.1991916 ]], dtype=float32), z=Array([[-1.22731   ,  0.5497297 ],\n",
       "        [-0.37638894,  0.07994474],\n",
       "        [-0.34790877,  0.2612425 ],\n",
       "        ...,\n",
       "        [-0.60085094,  0.38866115],\n",
       "        [ 1.0094485 , -0.7906791 ],\n",
       "        [-1.031179  , -0.46448508]], dtype=float32), chi=Array([[ -0.05498462,  -0.4905034 ,  -0.75695735,  -0.45752653,\n",
       "           0.9625589 ],\n",
       "        [-24.281794  , -42.62716   , -27.510563  , -47.73149   ,\n",
       "          10.354821  ],\n",
       "        [  1.4127568 ,  -0.20574978,  -0.489415  ,   1.9499726 ,\n",
       "           0.21919468],\n",
       "        ...,\n",
       "        [  2.6936674 ,  -2.4744701 ,   2.7006583 ,   2.3976943 ,\n",
       "          -2.9546552 ],\n",
       "        [ -7.541531  , -19.616173  ,  20.128855  , -14.5207205 ,\n",
       "           3.4616408 ],\n",
       "        [  5.1649623 ,  -1.65177   ,  -1.1234708 ,   3.853135  ,\n",
       "           3.1992185 ]], dtype=float32), zeta=Array([[-1.22731   ,  0.5497297 ],\n",
       "        [-0.3773298 ,  0.08020581],\n",
       "        [-0.34791592,  0.26128274],\n",
       "        ...,\n",
       "        [-0.60041445,  0.38983774],\n",
       "        [ 1.0094151 , -0.79070055],\n",
       "        [-1.0313816 , -0.46426752]], dtype=float32), y=Array([[-0.4204607 ,  0.5582306 ],\n",
       "        [-0.2697917 ,  1.2125616 ],\n",
       "        [-0.25412345,  0.2825214 ],\n",
       "        ...,\n",
       "        [-0.01865726,  1.3314339 ],\n",
       "        [ 1.6847588 , -3.1315465 ],\n",
       "        [-1.3079455 , -0.7718598 ]], dtype=float32), dx=Array([[-8.89450312e-05,  8.18371773e-05,  1.51276588e-04,\n",
       "          4.26471233e-05, -1.13904476e-04],\n",
       "        [ 3.92913818e-04,  6.56127930e-04,  1.33514404e-04,\n",
       "          6.63757324e-04, -2.45094299e-04],\n",
       "        [-2.56299973e-05, -1.09672546e-04,  1.44571066e-04,\n",
       "         -1.35064125e-04, -3.78340483e-05],\n",
       "        ...,\n",
       "        [-4.95910645e-05,  3.12089920e-04, -2.83718109e-04,\n",
       "         -4.63485718e-04,  1.66893005e-04],\n",
       "        [ 5.72204590e-06, -7.62939453e-06,  1.14440918e-05,\n",
       "         -3.81469727e-06,  3.33786011e-06],\n",
       "        [-1.68800354e-04,  8.24928284e-05,  1.01208687e-04,\n",
       "         -1.32560730e-04, -7.15255737e-05]], dtype=float32), dy=Array([[ 0.0000000e+00,  0.0000000e+00],\n",
       "        [-1.5056133e-04,  4.1723251e-05],\n",
       "        [-1.1324883e-06,  6.4373016e-06],\n",
       "        ...,\n",
       "        [ 6.9838017e-05,  1.8823147e-04],\n",
       "        [-5.3644180e-06, -3.5762787e-06],\n",
       "        [-3.2424927e-05,  3.4809113e-05]], dtype=float32), converged=Array([ True,  True,  True, ...,  True,  True,  True], dtype=bool), primal_infeas=Array([ True, False, False, ..., False, False, False], dtype=bool), dual_infeas=Array([False, False, False, ..., False, False, False], dtype=bool), kkt_mat=Array([[[ 1.00000095e+00,  2.90943265e-01,  4.28428799e-01, ...,\n",
       "          -1.34398133e-01,  2.45369360e-01,  9.77222860e-01],\n",
       "         [ 2.90943265e-01,  1.00000095e+00, -3.71799916e-01, ...,\n",
       "           1.02433927e-01, -1.74719200e-01,  2.18702212e-01],\n",
       "         [ 4.28428769e-01, -3.71799946e-01,  1.00000095e+00, ...,\n",
       "           5.31183660e-01,  4.87367839e-01, -9.79922891e-01],\n",
       "         ...,\n",
       "         [-1.34398147e-01,  1.02433927e-01,  5.31183660e-01, ...,\n",
       "           1.00000095e+00, -4.92184848e-01,  4.42206144e-01],\n",
       "         [ 2.45369360e-01, -1.74719200e-01,  4.87367839e-01, ...,\n",
       "          -4.92184848e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [ 9.77222860e-01,  2.18702212e-01, -9.79922891e-01, ...,\n",
       "           4.42206144e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000083e+00,  3.19931865e-01,  1.29165888e-01, ...,\n",
       "           7.55941868e-03,  2.22724944e-01, -1.21339463e-01],\n",
       "         [ 3.19931775e-01,  5.34273803e-01, -3.21735024e-01, ...,\n",
       "          -3.76205221e-02,  9.99999881e-01,  2.99548894e-01],\n",
       "         [ 1.29165888e-01, -3.21735114e-01,  6.58574343e-01, ...,\n",
       "          -2.38043159e-01, -5.53248167e-01,  1.00000000e+00],\n",
       "         ...,\n",
       "         [ 7.55941775e-03, -3.76205258e-02, -2.38043174e-01, ...,\n",
       "           7.20274687e-01,  2.56595135e-01,  9.99912381e-01],\n",
       "         [ 2.22724944e-01,  9.99999881e-01, -5.53248167e-01, ...,\n",
       "           2.56595135e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [-1.21339463e-01,  2.99548894e-01,  1.00000000e+00, ...,\n",
       "           9.99912381e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000095e+00,  8.13152730e-01, -2.08160710e-02, ...,\n",
       "          -6.49988770e-01, -2.06616633e-02, -1.54104165e-03],\n",
       "         [ 8.13152790e-01,  1.00000095e+00,  5.28875470e-01, ...,\n",
       "          -4.79724854e-01, -9.99626219e-01, -9.98811603e-01],\n",
       "         [-2.08160710e-02,  5.28875470e-01,  1.00000095e+00, ...,\n",
       "          -1.26250526e-02,  3.28693360e-01, -8.34178150e-01],\n",
       "         ...,\n",
       "         [-6.49988770e-01, -4.79724914e-01, -1.26250517e-02, ...,\n",
       "           1.00000095e+00,  1.01723671e-01, -4.54311162e-01],\n",
       "         [-2.06616633e-02, -9.99626219e-01,  3.28693360e-01, ...,\n",
       "           1.01723671e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [-1.54104165e-03, -9.98811603e-01, -8.34178150e-01, ...,\n",
       "          -4.54311162e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[ 9.99966919e-01, -2.45180234e-01, -7.95663893e-01, ...,\n",
       "           3.16982239e-01, -9.98583615e-01, -1.79590717e-01],\n",
       "         [-2.45180219e-01,  9.99966919e-01,  4.36819077e-01, ...,\n",
       "          -2.97621369e-01, -5.49139857e-01, -2.80232187e-02],\n",
       "         [-7.95663953e-01,  4.36819077e-01,  9.99966979e-01, ...,\n",
       "          -3.13698083e-01, -5.77117980e-01, -6.27150163e-02],\n",
       "         ...,\n",
       "         [ 3.16982210e-01, -2.97621340e-01, -3.13698083e-01, ...,\n",
       "           9.99967039e-01, -6.71777785e-01,  4.81437147e-01],\n",
       "         [-9.98583615e-01, -5.49139857e-01, -5.77117980e-01, ...,\n",
       "          -6.71777785e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [-1.79590717e-01, -2.80232187e-02, -6.27150163e-02, ...,\n",
       "           4.81437147e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000095e+00, -4.37439889e-01, -2.40884319e-01, ...,\n",
       "          -3.24169487e-01,  9.99173582e-01,  2.36111701e-01],\n",
       "         [-4.37439919e-01,  1.00000095e+00,  8.56435239e-01, ...,\n",
       "          -1.18445843e-01, -4.61766511e-01,  6.80679023e-01],\n",
       "         [-2.40884304e-01,  8.56435180e-01,  1.00000083e+00, ...,\n",
       "          -1.91558108e-01,  1.41935036e-01,  3.06358375e-02],\n",
       "         ...,\n",
       "         [-3.24169487e-01, -1.18445843e-01, -1.91558108e-01, ...,\n",
       "           1.00000083e+00, -7.30638802e-02, -2.21977174e-01],\n",
       "         [ 9.99173582e-01, -4.61766511e-01,  1.41935036e-01, ...,\n",
       "          -7.30638802e-02, -1.00000000e+01, -0.00000000e+00],\n",
       "         [ 2.36111701e-01,  6.80679023e-01,  3.06358375e-02, ...,\n",
       "          -2.21977174e-01, -0.00000000e+00, -1.00000000e+01]],\n",
       " \n",
       "        [[ 1.00000095e+00,  1.65870652e-01,  2.74994612e-01, ...,\n",
       "          -5.45680702e-01,  1.09289885e-01,  9.98292804e-01],\n",
       "         [ 1.65870652e-01,  1.00000095e+00,  2.64676541e-01, ...,\n",
       "           2.49961078e-01,  8.62548232e-01, -2.60905236e-01],\n",
       "         [ 2.74994642e-01,  2.64676511e-01,  1.00000095e+00, ...,\n",
       "          -4.56624269e-01,  9.98716295e-01,  3.78673710e-02],\n",
       "         ...,\n",
       "         [-5.45680702e-01,  2.49961078e-01, -4.56624240e-01, ...,\n",
       "           1.00000095e+00,  1.50169104e-01, -7.89535761e-01],\n",
       "         [ 1.09289885e-01,  8.62548232e-01,  9.98716295e-01, ...,\n",
       "           1.50169104e-01, -1.00000000e+01, -0.00000000e+00],\n",
       "         [ 9.98292804e-01, -2.60905236e-01,  3.78673710e-02, ...,\n",
       "          -7.89535761e-01, -0.00000000e+00, -1.00000000e+01]]],      dtype=float32), kkt_lu=(Array([[[ 1.00000095e+00,  2.90943265e-01,  4.28428799e-01, ...,\n",
       "          -1.34398133e-01,  2.45369360e-01,  9.77222860e-01],\n",
       "         [ 2.90942997e-01,  9.15353060e-01, -4.96448278e-01, ...,\n",
       "           1.41536117e-01, -2.46107697e-01, -6.56139329e-02],\n",
       "         [ 9.77221906e-01, -7.16815367e-02, -1.43417907e+00, ...,\n",
       "           5.83688498e-01, -2.57421672e-01, -1.09596672e+01],\n",
       "         ...,\n",
       "         [-1.34398013e-01,  1.54624626e-01, -4.64047223e-01, ...,\n",
       "           1.40167451e+00,  4.88851500e+00, -3.34997702e+00],\n",
       "         [-3.41158628e-01,  4.27828968e-01,  7.89805055e-02, ...,\n",
       "           1.56632677e-01,  4.79173994e+00,  3.67295647e+00],\n",
       "         [ 4.28428352e-01, -5.42357147e-01, -3.81540626e-01, ...,\n",
       "           6.98707461e-01, -7.67292008e-02, -2.37838531e+00]],\n",
       " \n",
       "        [[ 1.00000083e+00,  3.19931865e-01,  1.29165888e-01, ...,\n",
       "           7.55941868e-03,  2.22724944e-01, -1.21339463e-01],\n",
       "         [ 2.22724766e-01,  9.28743124e-01, -5.82016587e-01, ...,\n",
       "           2.54911453e-01, -1.00496063e+01,  2.70253029e-02],\n",
       "         [-1.21339358e-01,  3.64330232e-01,  1.22771919e+00, ...,\n",
       "           9.07957613e-01,  3.68840075e+00, -1.00245686e+01],\n",
       "         ...,\n",
       "         [ 1.29165784e-01, -3.90914708e-01,  3.37513387e-01, ...,\n",
       "          -2.02846393e-01, -5.59413052e+00,  4.10690260e+00],\n",
       "         [ 3.19931507e-01,  4.65055943e-01, -7.52525330e-02, ...,\n",
       "           4.31442350e-01,  8.29530048e+00, -2.20388460e+00],\n",
       "         [-8.67387593e-01, -1.74649939e-01, -1.34977758e-01, ...,\n",
       "          -4.57836479e-01, -5.46411693e-01, -8.12546968e-01]],\n",
       " \n",
       "        [[ 1.00000095e+00,  8.13152730e-01, -2.08160710e-02, ...,\n",
       "          -6.49988770e-01, -2.06616633e-02, -1.54104165e-03],\n",
       "         [-1.54104014e-03, -9.97558475e-01, -8.34210217e-01, ...,\n",
       "          -4.55312818e-01, -3.18404527e-05, -1.00000019e+01],\n",
       "         [-2.06616428e-02,  9.85230625e-01,  1.15015268e+00, ...,\n",
       "           5.36881983e-01, -1.00003958e+01,  9.85227585e+00],\n",
       "         ...,\n",
       "         [-6.49988174e-01, -4.89342175e-02, -5.82328551e-02, ...,\n",
       "           3.24052036e-01,  2.01263809e+00, -5.80947495e+00],\n",
       "         [-6.96985245e-01, -2.64149815e-01,  3.27892870e-01, ...,\n",
       "          -2.84668982e-01, -8.86914432e-01,  2.16559815e+00],\n",
       "         [ 8.13152015e-01, -3.39613318e-01,  2.28224680e-01, ...,\n",
       "          -1.68248769e-02,  8.96972954e-01, -4.06357288e+00]],\n",
       " \n",
       "        ...,\n",
       " \n",
       "        [[ 9.99966919e-01, -2.45180234e-01, -7.95663893e-01, ...,\n",
       "           3.16982239e-01, -9.98583615e-01, -1.79590717e-01],\n",
       "         [-2.45188341e-01,  9.39851582e-01,  2.41731569e-01, ...,\n",
       "          -2.19901025e-01, -7.93980896e-01, -7.20567703e-02],\n",
       "         [-9.98616695e-01, -8.44793975e-01, -1.16746783e+00, ...,\n",
       "          -5.41005075e-01, -1.16679525e+01, -2.40215406e-01],\n",
       "         ...,\n",
       "         [ 3.16992700e-01, -2.33974159e-01,  4.21381649e-03, ...,\n",
       "           5.81556797e-01, -1.21190405e+00,  4.94110823e+00],\n",
       "         [ 1.05101861e-01,  2.65305847e-01,  3.83637965e-01, ...,\n",
       "           2.79557884e-01,  3.98794055e+00,  7.53923225e+00],\n",
       "         [-7.95690298e-01,  2.57201821e-01, -2.60984600e-01, ...,\n",
       "           1.08454101e-01, -8.82844210e-01,  2.43113351e+00]],\n",
       " \n",
       "        [[ 1.00000095e+00, -4.37439889e-01, -2.40884319e-01, ...,\n",
       "          -3.24169487e-01,  9.99173582e-01,  2.36111701e-01],\n",
       "         [-4.37439501e-01,  8.08647454e-01,  7.51062930e-01, ...,\n",
       "          -2.60250390e-01, -2.46885177e-02,  7.83963621e-01],\n",
       "         [ 2.36111477e-01,  9.69475150e-01, -6.40625477e-01, ...,\n",
       "           1.06869251e-01, -2.11981446e-01, -1.08157825e+01],\n",
       "         ...,\n",
       "         [-3.24169189e-01, -3.21834177e-01,  4.35943827e-02, ...,\n",
       "           7.18125105e-01,  1.14027746e-01,  4.03622484e+00],\n",
       "         [ 9.99172628e-01, -3.05306688e-02, -6.33054018e-01, ...,\n",
       "           4.26796257e-01, -1.11882963e+01, -8.62303066e+00],\n",
       "         [-2.40884081e-01,  9.28789020e-01, -3.81496936e-01, ...,\n",
       "          -1.00256570e-01, -1.81913376e-02, -1.19945824e+00]],\n",
       " \n",
       "        [[ 1.00000095e+00,  1.65870652e-01,  2.74994612e-01, ...,\n",
       "          -5.45680702e-01,  1.09289885e-01,  9.98292804e-01],\n",
       "         [ 1.65870488e-01,  9.72487926e-01,  2.19063044e-01, ...,\n",
       "           3.40473413e-01,  8.44420254e-01, -4.26492542e-01],\n",
       "         [ 2.74994373e-01,  2.25260392e-01,  8.75032783e-01, ...,\n",
       "          -3.83260310e-01,  7.78447747e-01, -1.40585646e-01],\n",
       "         ...,\n",
       "         [-5.45680165e-01,  3.50105554e-01, -4.37995344e-01, ...,\n",
       "           2.75416821e-01,  6.52027321e+00, -3.68642360e-01],\n",
       "         [ 9.98291850e-01, -4.38558191e-01, -1.60663247e-01, ...,\n",
       "          -9.63478506e-01,  1.15242376e+01, -1.17253933e+01],\n",
       "         [-4.30781662e-01,  2.66788334e-01,  7.13850141e-01, ...,\n",
       "          -7.63579309e-01, -1.06620990e-01, -1.60240817e+00]]],      dtype=float32), Array([[0, 1, 6, ..., 4, 5, 6],\n",
       "        [0, 5, 6, ..., 6, 5, 6],\n",
       "        [0, 6, 5, ..., 4, 5, 6],\n",
       "        ...,\n",
       "        [0, 1, 5, ..., 4, 6, 6],\n",
       "        [0, 1, 6, ..., 4, 5, 6],\n",
       "        [0, 1, 2, ..., 4, 6, 6]], dtype=int32)), rho=Array([[0.1, 0.1],\n",
       "        [0.1, 0.1],\n",
       "        [0.1, 0.1],\n",
       "        ...,\n",
       "        [0.1, 0.1],\n",
       "        [0.1, 0.1],\n",
       "        [0.1, 0.1]], dtype=float32, weak_type=True)))"
      ]
     },
     "execution_count": 310,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prob, data, state = osqp.OSQPProblem.from_data(P, q, A, l, u)\n",
    "%time sol_jitd(data, state)\n",
    "%time sol_sparse_jitd(data, state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "id": "9c8325fa-ea0b-494c-bafb-332cbdf3149a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'solve time (batch of 100), ms')"
      ]
     },
     "execution_count": 280,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_vals = [4, 8, 16, 32, 64, 128, 256, 512]\n",
    "times = [10.9, 9.01, 9.7, 10.3, 12, 39.6, 149, 534]\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "plt.plot(n_vals, times)\n",
    "# plt.xscale('log')\n",
    "# plt.yscale('log')\n",
    "plt.xlabel('problem size, n')\n",
    "plt.ylabel('solve time (batch of 100), ms')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "id": "5138e842-a608-4d0c-8861-a14b8e95354d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "with jax.profiler.trace(\"/tmp/tensorboard\"):\n",
    "    prob, data, state = osqp.OSQPProblem.from_data(P, q, A, l, u)\n",
    "    sol_mine = jax.vmap(prob.solve)(data, state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "id": "7f20f193-d860-4431-afa2-b72549731ae0",
   "metadata": {},
   "outputs": [],
   "source": [
    "prob, data, state = osqp.OSQPProblem.from_data(P, q, A, l, u)\n",
    "sol_mine = jax.vmap(prob.solve)(data, state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "7aac2bb1-586b-4a25-8295-b3b151308e0a",
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "'OSQPState' object is not subscriptable",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[77], line 3\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mmatplotlib\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m pyplot \u001b[38;5;28;01mas\u001b[39;00m plt\n\u001b[0;32m----> 3\u001b[0m plt\u001b[38;5;241m.\u001b[39mhist(sol[\u001b[38;5;241m0\u001b[39m], bins\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m8\u001b[39m)\n\u001b[1;32m      4\u001b[0m plt\u001b[38;5;241m.\u001b[39mtitle(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mHistogram of solve times, n=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mn\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m, B=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mB\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m, m=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mm\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m      5\u001b[0m plt\u001b[38;5;241m.\u001b[39mxlabel(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mNumber of iterations\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "\u001b[0;31mTypeError\u001b[0m: 'OSQPState' object is not subscriptable"
     ]
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "plt.hist(sol[0], bins=8)\n",
    "plt.title(f'Histogram of solve times, n={n}, B={B}, m={m}')\n",
    "plt.xlabel('Number of iterations')\n",
    "plt.ylabel('Freq')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "54286abe-1c99-4d5a-824d-f9dd7f9851e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array([2.005999 , 2.1144671, 2.004582 , 2.5526845, 2.1723697], dtype=float32)"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "A = sparse.random_bcoo(key, (5, 2, 2), nse=0.15, n_batch=1, generator=random.normal)\n",
    "sum_x = jax.vmap(sparse.sparsify(lambda x: jnp.sum(x @ x.T + sparse.eye(2))))\n",
    "\n",
    "sum_x(A)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "396a92eb-3fec-4dc1-abc2-e4cb9f4da2db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array([[2., 2., 2., 2., 2.],\n",
       "       [2., 2., 2., 2., 2.]], dtype=float32)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import jax_dataclasses as jdc\n",
    "\n",
    "@jdc.pytree_dataclass\n",
    "class Foo:\n",
    "    x: jnp.ndarray\n",
    "    y: jnp.ndarray\n",
    "    z: jdc.Static[int]\n",
    "    \n",
    "def sumup(foo):\n",
    "    return foo.x + foo.y\n",
    "\n",
    "bar = Foo(jnp.ones((2, 5)), jnp.ones((2, 5)), 3)\n",
    "jax.vmap(sumup)(bar)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 359,
   "id": "d5838d92-65d9-4a4b-b35a-b43d95e06c16",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.06 ms ± 31.9 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)\n",
      "450 µs ± 4.38 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)\n"
     ]
    }
   ],
   "source": [
    "# A_sparse = sparse.random_bcoo(subkey, (B, m, n), nse=0.01, n_batch=1, generator=random.normal)\n",
    "# A = A_sparse.todense()\n",
    "\n",
    "A_sparse = sparse.random_bcoo(subkey, (B, m, n), nse=0.01, n_batch=1, generator=random.normal)\n",
    "A_sparse = sparse.BCSR.from_bcoo(A_sparse)\n",
    "\n",
    "A = A_sparse.todense()\n",
    "\n",
    "matmul_jit = jax.jit(sparse.sparsify(lambda A, x: A@x))\n",
    "\n",
    "state = sol[-1]\n",
    "\n",
    "jax.vmap(matmul_jit)(A_sparse, state.x)\n",
    "jax.vmap(matmul_jit)(A, state.x)\n",
    "\n",
    "%timeit jax.vmap(matmul_jit)(A_sparse, state.x)\n",
    "%timeit jax.vmap(matmul_jit)(A, state.x)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "# # Define jit'd matrix-vector product (vmap across batch dim)\n",
    "# bmm = jax.jit(jax.vmap(sparse.sparsify(lambda A, x: A @ x)))\n",
    "\n",
    "# # Run it once to compile\n",
    "# bmm(A, state.x)\n",
    "# bmm(A_sparse, state.x)\n",
    "\n",
    "# # Time dense / sparse naive calls\n",
    "# %timeit bmm(A, state.x) # 23.1 µs ± 1.59 µs per loop\n",
    "# %timeit bmm(A_sparse, state.x) # 32.6 µs ± 2.2 µs per loop\n",
    "\n",
    "# jit a low-level matmul\n",
    "# bmm_sp = lambda A, x: sparse.bcoo_dot_general(A, x, dimension_numbers=((2, 1), (0,0)))\n",
    "# jitbmmsp = jax.jit(bmm_sp) \n",
    "\n",
    "# jitbmmsp(A_sparse, state.x) # Run to compile\n",
    "# %timeit jitbmmsp(A_sparse, state.x) # 30.1 µs ± 2.02 µs per loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "7e29b4ad-389b-4993-8c7f-165e12dc7e3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10, 1000, 100)"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "A_sparse.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "id": "3fb0242c-923d-48f3-b661-30cbd02a5ed5",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'times' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[182], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m times\n",
      "\u001b[0;31mNameError\u001b[0m: name 'times' is not defined"
     ]
    }
   ],
   "source": [
    "times"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 364,
   "id": "f184fd6c-44c4-4f9a-8e5e-8c75052b96d9",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1.83 ms, sys: 159 µs, total: 1.99 ms\n",
      "Wall time: 2 ms\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "5.802953"
      ]
     },
     "execution_count": 364,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import osqp as oo\n",
    "import numpy as np\n",
    "import scipy \n",
    "\n",
    "B, m, n = A.shape\n",
    "\n",
    "results = []\n",
    "probs = []\n",
    "\n",
    "def setup_all(probs):\n",
    "    for ii in range(B):\n",
    "        m = oo.OSQP()\n",
    "        m.setup(\n",
    "            P=scipy.sparse.csc_matrix(np.asarray(P[ii, :, :])), \n",
    "            q=np.asarray(q[ii,:]), \n",
    "            A=scipy.sparse.csc_matrix(np.asarray(A[ii, :, :])), \n",
    "            l=np.asarray(l[ii, :]), \n",
    "            u=np.asarray(u[ii,:]), \n",
    "            verbose=False, \n",
    "            adaptive_rho=True,\n",
    "            scaling=10,\n",
    "            max_iter=4000)\n",
    "            # linsys_solver='mkl paradiso')\n",
    "        probs.append(m)\n",
    "        \n",
    "def solve_all(probs, results):\n",
    "    solve_time = 0.\n",
    "    for m in probs:\n",
    "        results.append(m.solve())\n",
    "        solve_time += results[-1].info.setup_time + results[-1].info.solve_time\n",
    "        \n",
    "    return solve_time\n",
    "\n",
    "setup_all(probs)\n",
    "%time solve_time = solve_all(probs, results)\n",
    "osqp_iters = [rr.info.iter for rr in results]\n",
    "\n",
    "1000 * solve_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 365,
   "id": "64add6ed-18e9-4b51-ba46-e5c86a7e0419",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f1962e1c610>"
      ]
     },
     "execution_count": 365,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "plt.close(\"all\")\n",
    "fig, ax = plt.subplots()\n",
    "ax.hist([sol[0], osqp_iters], label=['jax', 'osqp'], bins=20)\n",
    "# ax.hist(, bins=10, range=(0, max(sol[0])), label='osqp', alpha=0.5)\n",
    "ax.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "id": "dd83219f-5859-49b8-b00a-8681a093843d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f19510a9410>"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh8AAAGdCAYAAACyzRGfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAhtklEQVR4nO3de3BU9f3/8dfmsgmQ7IZAyEUSLqJci9BEYBEUMTaDSqHEqRcGwWGw0gBCipfMVwHREryCOoCKXLRK02JFixeoTQuOJVwMMAWCEWhogiHBCyQQJpvAnt8fne7PFFQ22Xx2NzwfM2fGnD05+96P0Txn9+zGZlmWJQAAAEPCAj0AAAC4vBAfAADAKOIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMCoi0AP8L4/Ho8rKSsXGxspmswV6HAAAcAksy9Lp06eVkpKisLAffm4j6OKjsrJSqampgR4DAAA0Q0VFhbp27fqDxwRdfMTGxkr6z/AOhyPA0wAAgEtRW1ur1NRU7+/xHxJ08fHfl1ocDgfxAQBAiLmUSya44BQAABhFfAAAAKOIDwAAYFTQXfMBAEBrsCxL586d0/nz5wM9SsiKjIxUeHh4i89DfAAA2ryGhgYdP35cZ8+eDfQoIc1ms6lr166KiYlp0XmIDwBAm+bxeFRWVqbw8HClpKTIbrfzIZbNYFmWvvrqKx07dkxXXXVVi54BIT4AAG1aQ0ODPB6PUlNT1b59+0CPE9ISEhJ09OhRNTY2tig+uOAUAHBZ+LGP/MaP89czRvybAAAARhEfAAAEqSlTpmj8+PGBHsPvfLrmY8GCBXr88ceb7Ovdu7c+//xzSVJ9fb1+85vfqKCgQG63W1lZWVq+fLkSExP9NzEAAH7S/ZEPjN3X0cW3+vw9L7zwgizLaoVpAsvnZz769++v48ePe7dPP/3Ue9ucOXO0ceNGrV+/Xlu3blVlZaUmTJjg14EBALhcOJ1OxcXFBXoMv/M5PiIiIpSUlOTdOnfuLEmqqanRqlWr9Pzzz2v06NFKT0/XmjVrtG3bNm3fvt3vgwMA0NZ992WXTZs2acSIEYqLi1OnTp1022236ciRI95j33jjDcXExOjQoUPefb/+9a/Vp0+foPt8E5/j49ChQ0pJSVHPnj01ceJElZeXS5KKi4vV2NiozMxM77F9+vRRWlqaioqKvvd8brdbtbW1TTYAANBUXV2dcnNz9dlnn6mwsFBhYWH6xS9+IY/HI0m65557dMstt2jixIk6d+6cPvjgA7322mt66623gu4txj5d8zF06FCtXbtWvXv31vHjx/X4449r5MiR2r9/v6qqqmS32y94eigxMVFVVVXfe878/PwLriNpTf56fa85r90BANBc2dnZTb5evXq1EhISVFJSogEDBkiSXnnlFQ0cOFCzZs3SO++8owULFig9PT0Q4/4gn+JjzJgx3n8eOHCghg4dqm7duumPf/yj2rVr16wB8vLylJub6/26trZWqampzToXAABt1aFDhzRv3jzt2LFDX3/9tfcZj/Lycm98dOzYUatWrVJWVpaGDx+uRx55JJAjf68WvdU2Li5OV199tQ4fPqykpCQ1NDTo1KlTTY6prq5WUlLS954jKipKDoejyQYAAJoaO3asvv32W61cuVI7duzQjh07JP3nE1y/65NPPlF4eLiOHz+uurq6QIz6o1oUH2fOnNGRI0eUnJys9PR0RUZGqrCw0Ht7aWmpysvL5XK5WjwoAACXq2+++UalpaV69NFHddNNN6lv3746efLkBcdt27ZNTz31lDZu3KiYmBjNmDEjANP+OJ9edpk7d67Gjh2rbt26qbKyUvPnz1d4eLjuuusuOZ1OTZ06Vbm5uYqPj5fD4dDMmTPlcrk0bNiw1pofAIA2r2PHjurUqZNeffVVJScnq7y8/IKXVE6fPq1JkyZp1qxZGjNmjLp27aprr71WY8eO1e233x6gyS/Op/g4duyY7rrrLn3zzTdKSEjQiBEjtH37diUkJEiSlixZorCwMGVnZzf5kDEAANB8YWFhKigo0KxZszRgwAD17t1bL774okaNGuU95oEHHlCHDh20aNEiSdJPfvITLVq0SL/61a/kcrl0xRVXBGj6C9msIPvotNraWjmdTtXU1LTK9R+82wUALi/19fUqKytTjx49FB0dHehxfHLXXXcpPDxcb775ZqBHkfTDa+nL72/+tgsAAEHm3LlzKikpUVFRkfr37x/ocfyO+AAAIMjs379fGRkZ6t+/v+6///5Aj+N3Pl3zAQAAWt+gQYOC7iPR/YlnPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjiA8AAGAUn/MBALh8LXAavK8ac/cV5HjmAwAAGEV8AAAQpNxut2bNmqUuXbooOjpaI0aM0K5duyRJJ0+e1MSJE5WQkKB27drpqquu0po1a7zfu3PnTg0ePFjR0dHKyMjQhg0bZLPZtHfvXknSli1bZLPZ9MEHH2jgwIGKjo7WsGHDtH///lZ/XMQHAABB6qGHHtKf/vQnvf7669q9e7d69eqlrKwsffvtt3rsscdUUlKijz76SAcPHtSKFSvUuXNnSdKZM2d02223qV+/fiouLtaCBQs0d+7ci97Hgw8+qOeee067du1SQkKCxo4dq8bGxlZ9XFzzAQBAEKqrq9OKFSu0du1ajRkzRpK0cuVKffzxx1q1apXKy8s1ePBgZWRkSJK6d+/u/d5169bJ4/Fo1apVio6OVv/+/XXs2DFNnz79gvuZP3++br75ZknS66+/rq5du2rDhg365S9/2WqPjWc+AAAIQkeOHFFjY6Ouu+46777IyEgNGTJEBw8e1PTp01VQUKBBgwbpoYce0rZt27zHHTx40PtSyn+5XK6L3s9398fHx6t37946ePBgKzyi/4/4AAAgBI0ZM0b//ve/NWfOHFVWVuqmm2763pdWgg3xAQBAELryyitlt9v1j3/8w7uvsbFRu3btUr9+/SRJCQkJmjx5st58800tXbpUr776qiSpb9+++uc//6n6+nrv927fvv2i9/Pd/SdPntQXX3yhvn37tsZD8iI+AAAIQh06dND06dP14IMPatOmTSopKdG0adN09uxZTZ06VfPmzdN7772nw4cP68CBA3r//fe90XD33XfLZrNp2rRpKikp0Ycffqhnn332ovezcOFCFRYWav/+/ZoyZYo6d+6s8ePHt+pj44JTAACC1OLFi+XxeDRp0iSdPn1aGRkZ2rx5szp27Ci73a68vDwdPXpU7dq108iRI1VQUCBJiomJ0caNG3X//fdr8ODB6tevn5566illZ2df9D4eeOABHTp0SIMGDdLGjRtlt9tb9XHZLMuyWvUefFRbWyun06mamho5HA6/n7/7Ix/45TxHF9/ql/MAAFpXfX29ysrK1KNHjyYXYF5ujh49qh49emjPnj0aNGiQtmzZohtvvFEnT55UXFzcJZ3jh9bSl9/fvOwCAACMIj4AAIBRXPMBAMBloHv37vrulRajRo1SoK684JkPAABgFPEBAACMIj4AAJeFIHtzZ0jy1xoSHwCANi0yMlKSdPbs2QBPEvoaGhokSeHh4S06DxecAgDatPDwcMXFxenEiROSpPbt28tmswV4qtDj8Xj01VdfqX379oqIaFk+EB8AgDYvKSlJkrwBguYJCwtTWlpai+ON+AAAtHk2m03Jycnq0qWLGhsbAz1OyLLb7QoLa/kVG8QHAOCyER4e3uLrFdByXHAKAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCqRfGxePFi2Ww2zZ4927uvvr5eOTk56tSpk2JiYpSdna3q6uqWzgkAANqIZsfHrl279Morr2jgwIFN9s+ZM0cbN27U+vXrtXXrVlVWVmrChAktHhQAALQNzYqPM2fOaOLEiVq5cqU6duzo3V9TU6NVq1bp+eef1+jRo5Wenq41a9Zo27Zt2r59u9+GBgAAoatZ8ZGTk6Nbb71VmZmZTfYXFxersbGxyf4+ffooLS1NRUVFLZsUAAC0CRG+fkNBQYF2796tXbt2XXBbVVWV7Ha74uLimuxPTExUVVXVRc/ndrvldru9X9fW1vo6EgAACCE+PfNRUVGhBx54QG+99Zaio6P9MkB+fr6cTqd3S01N9ct5AQBAcPIpPoqLi3XixAn99Kc/VUREhCIiIrR161a9+OKLioiIUGJiohoaGnTq1Kkm31ddXa2kpKSLnjMvL081NTXeraKiotkPBgAABD+fXna56aabtG/fvib77r33XvXp00cPP/ywUlNTFRkZqcLCQmVnZ0uSSktLVV5eLpfLddFzRkVFKSoqqpnjAwCAUONTfMTGxmrAgAFN9nXo0EGdOnXy7p86dapyc3MVHx8vh8OhmTNnyuVyadiwYf6bGgAAhCyfLzj9MUuWLFFYWJiys7PldruVlZWl5cuX+/tuAABAiLJZlmUFeojvqq2tldPpVE1NjRwOh9/P3/2RD/xynqOLb/XLeQAAaAt8+f3N33YBAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKN8io8VK1Zo4MCBcjgccjgccrlc+uijj7y319fXKycnR506dVJMTIyys7NVXV3t96EBAEDo8ik+unbtqsWLF6u4uFifffaZRo8erXHjxunAgQOSpDlz5mjjxo1av369tm7dqsrKSk2YMKFVBgcAAKHJZlmW1ZITxMfH65lnntHtt9+uhIQErVu3Trfffrsk6fPPP1ffvn1VVFSkYcOGXdL5amtr5XQ6VVNTI4fD0ZLRLqr7Ix/45TxHF9/ql/MAANAW+PL7u9nXfJw/f14FBQWqq6uTy+VScXGxGhsblZmZ6T2mT58+SktLU1FR0feex+12q7a2tskGAADaLp/jY9++fYqJiVFUVJTuv/9+bdiwQf369VNVVZXsdrvi4uKaHJ+YmKiqqqrvPV9+fr6cTqd3S01N9flBAACA0OFzfPTu3Vt79+7Vjh07NH36dE2ePFklJSXNHiAvL081NTXeraKiotnnAgAAwS/C12+w2+3q1auXJCk9PV27du3SCy+8oDvuuEMNDQ06depUk2c/qqurlZSU9L3ni4qKUlRUlO+TAwCAkNTiz/nweDxyu91KT09XZGSkCgsLvbeVlpaqvLxcLperpXcDAADaCJ+e+cjLy9OYMWOUlpam06dPa926ddqyZYs2b94sp9OpqVOnKjc3V/Hx8XI4HJo5c6ZcLtclv9MFAAC0fT7Fx4kTJ3TPPffo+PHjcjqdGjhwoDZv3qybb75ZkrRkyRKFhYUpOztbbrdbWVlZWr58easMDgAAQlOLP+fD3/icDwAAQo+Rz/kAAABoDuIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIr4AAAARhEfAADAKOIDAAAYRXwAAACjfIqP/Px8XXvttYqNjVWXLl00fvx4lZaWNjmmvr5eOTk56tSpk2JiYpSdna3q6mq/Dg0AAEKXT/GxdetW5eTkaPv27fr444/V2Nion/3sZ6qrq/MeM2fOHG3cuFHr16/X1q1bVVlZqQkTJvh9cAAAEJoifDl406ZNTb5eu3atunTpouLiYl1//fWqqanRqlWrtG7dOo0ePVqStGbNGvXt21fbt2/XsGHD/Dc5AAAISS265qOmpkaSFB8fL0kqLi5WY2OjMjMzvcf06dNHaWlpKioquug53G63amtrm2wAAKDtanZ8eDwezZ49W9ddd50GDBggSaqqqpLdbldcXFyTYxMTE1VVVXXR8+Tn58vpdHq31NTU5o4EAABCQLPjIycnR/v371dBQUGLBsjLy1NNTY13q6ioaNH5AABAcPPpmo//mjFjht5//3198skn6tq1q3d/UlKSGhoadOrUqSbPflRXVyspKemi54qKilJUVFRzxgAAACHIp2c+LMvSjBkztGHDBv3tb39Tjx49mtyenp6uyMhIFRYWeveVlpaqvLxcLpfLPxMDAICQ5tMzHzk5OVq3bp3ee+89xcbGeq/jcDqdateunZxOp6ZOnarc3FzFx8fL4XBo5syZcrlcvNMFAABI8jE+VqxYIUkaNWpUk/1r1qzRlClTJElLlixRWFiYsrOz5Xa7lZWVpeXLl/tlWAAAEPp8ig/Lsn70mOjoaC1btkzLli1r9lAAAKDt4m+7AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMigj0AMAlWeD047lq/HcuAIDPeOYDAAAYRXwAAACjiA8AAGAU8QEAAIwiPgAAgFHEBwAAMIq32jZXqL71M1TnBgC0GTzzAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjPI5Pj755BONHTtWKSkpstlsevfdd5vcblmW5s2bp+TkZLVr106ZmZk6dOiQv+YFAAAhzuf4qKur0zXXXKNly5Zd9Pann35aL774ol5++WXt2LFDHTp0UFZWlurr61s8LAAACH0+/22XMWPGaMyYMRe9zbIsLV26VI8++qjGjRsnSXrjjTeUmJiod999V3feeWfLpgUAACHPr9d8lJWVqaqqSpmZmd59TqdTQ4cOVVFR0UW/x+12q7a2tskGAADaLr/GR1VVlSQpMTGxyf7ExETvbf8rPz9fTqfTu6WmpvpzJAAAEGQC/m6XvLw81dTUeLeKiopAjwQAAFqRX+MjKSlJklRdXd1kf3V1tfe2/xUVFSWHw9FkAwAAbZdf46NHjx5KSkpSYWGhd19tba127Nghl8vlz7sCAAAhyud3u5w5c0aHDx/2fl1WVqa9e/cqPj5eaWlpmj17tp588kldddVV6tGjhx577DGlpKRo/Pjx/pwbAACEKJ/j47PPPtONN97o/To3N1eSNHnyZK1du1YPPfSQ6urqdN999+nUqVMaMWKENm3apOjoaP9NDQAAQpbP8TFq1ChZlvW9t9tsNi1cuFALFy5s0WAAAKBtCvi7XQAAwOWF+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgVESgBwDatAVOP56rxn/nAoAA4pkPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBREYEeAAAuewucfjxXjf/OBbQSnvkAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBTxAQAAjCI+AACAUcQHAAAwivgAAABGER8AAMAo4gMAABhFfAAAAKOIDwAAYBR/1RbAhUL1r6yG6tzApWhDP9888wEAAIwiPgAAgFGtFh/Lli1T9+7dFR0draFDh2rnzp2tdVcAACCEtEp8/OEPf1Bubq7mz5+v3bt365prrlFWVpZOnDjRGncHAABCSKvEx/PPP69p06bp3nvvVb9+/fTyyy+rffv2Wr16dWvcHQAACCF+f7dLQ0ODiouLlZeX590XFhamzMxMFRUVXXC82+2W2+32fl1T858rcGtra/09miTJ4z7rl/PU2iy/nOc/J2udx3pRbuZm7kvA3MyN4BPkPyf//b1tWZcwp+VnX375pSXJ2rZtW5P9Dz74oDVkyJALjp8/f74liY2NjY2Nja0NbBUVFT/aCgH/nI+8vDzl5uZ6v/Z4PPr222/VqVMn2Wy2AE7mX7W1tUpNTVVFRYUcDkegxwkprF3zsXYtw/o1H2vXfKG6dpZl6fTp00pJSfnRY/0eH507d1Z4eLiqq6ub7K+urlZSUtIFx0dFRSkqKqrJvri4OH+PFTQcDkdI/TAFE9au+Vi7lmH9mo+1a75QXDun03lJx/n9glO73a709HQVFhZ693k8HhUWFsrlcvn77gAAQIhplZddcnNzNXnyZGVkZGjIkCFaunSp6urqdO+997bG3QEAgBDSKvFxxx136KuvvtK8efNUVVWlQYMGadOmTUpMTGyNuwsJUVFRmj9//gUvMeHHsXbNx9q1DOvXfKxd810Oa2ezrEt5TwwAAIB/8LddAACAUcQHAAAwivgAAABGER8AAMAo4sOPli1bpu7duys6OlpDhw7Vzp07v/fYAwcOKDs7W927d5fNZtPSpUvNDRqEfFm7lStXauTIkerYsaM6duyozMzMHzy+rfNl7d555x1lZGQoLi5OHTp00KBBg/S73/3O4LTBxZe1+66CggLZbDaNHz++dQcMcr6s39q1a2Wz2Zps0dHRBqcNLr7+7J06dUo5OTlKTk5WVFSUrr76an344YeGpm0F/vmLLigoKLDsdru1evVq68CBA9a0adOsuLg4q7q6+qLH79y505o7d671+9//3kpKSrKWLFliduAg4uva3X333dayZcusPXv2WAcPHrSmTJliOZ1O69ixY4YnDzxf1+7vf/+79c4771glJSXW4cOHraVLl1rh4eHWpk2bDE8eeL6u3X+VlZVZV1xxhTVy5Ehr3LhxZoYNQr6u35o1ayyHw2EdP37cu1VVVRmeOjj4unZut9vKyMiwbrnlFuvTTz+1ysrKrC1btlh79+41PLn/EB9+MmTIECsnJ8f79fnz562UlBQrPz//R7+3W7dul3V8tGTtLMuyzp07Z8XGxlqvv/56a40YtFq6dpZlWYMHD7YeffTR1hgvqDVn7c6dO2cNHz7ceu2116zJkydf1vHh6/qtWbPGcjqdhqYLbr6u3YoVK6yePXtaDQ0NpkZsdbzs4gcNDQ0qLi5WZmamd19YWJgyMzNVVFQUwMmCnz/W7uzZs2psbFR8fHxrjRmUWrp2lmWpsLBQpaWluv7661tz1KDT3LVbuHChunTpoqlTp5oYM2g1d/3OnDmjbt26KTU1VePGjdOBAwdMjBtUmrN2f/7zn+VyuZSTk6PExEQNGDBAixYt0vnz502N7XfEhx98/fXXOn/+/AWf4JqYmKiqqqoATRUa/LF2Dz/8sFJSUpr8x3w5aO7a1dTUKCYmRna7Xbfeeqteeukl3Xzzza09blBpztp9+umnWrVqlVauXGlixKDWnPXr3bu3Vq9erffee09vvvmmPB6Phg8frmPHjpkYOWg0Z+3+9a9/6e2339b58+f14Ycf6rHHHtNzzz2nJ5980sTIraJVPl4dMGXx4sUqKCjQli1bLuuL13wRGxurvXv36syZMyosLFRubq569uypUaNGBXq0oHX69GlNmjRJK1euVOfOnQM9TkhyuVxN/rjo8OHD1bdvX73yyit64oknAjhZ8PN4POrSpYteffVVhYeHKz09XV9++aWeeeYZzZ8/P9DjNQvx4QedO3dWeHi4qqurm+yvrq5WUlJSgKYKDS1Zu2effVaLFy/WX//6Vw0cOLA1xwxKzV27sLAw9erVS5I0aNAgHTx4UPn5+ZdVfPi6dkeOHNHRo0c1duxY7z6PxyNJioiIUGlpqa688srWHTqI+OP/eZGRkRo8eLAOHz7cGiMGreasXXJysiIjIxUeHu7d17dvX1VVVamhoUF2u71VZ24NvOziB3a7Xenp6SosLPTu83g8KiwsbFL6uFBz1+7pp5/WE088oU2bNikjI8PEqEHHXz93Ho9Hbre7NUYMWr6uXZ8+fbRv3z7t3bvXu/385z/XjTfeqL179yo1NdXk+AHnj5+98+fPa9++fUpOTm6tMYNSc9buuuuu0+HDh73BK0lffPGFkpOTQzI8JPFWW38pKCiwoqKirLVr11olJSXWfffdZ8XFxXnfSjZp0iTrkUce8R7vdrutPXv2WHv27LGSk5OtuXPnWnv27LEOHToUqIcQML6u3eLFiy273W69/fbbTd62d/r06UA9hIDxde0WLVpk/eUvf7GOHDlilZSUWM8++6wVERFhrVy5MlAPIWB8Xbv/dbm/28XX9Xv88cetzZs3W0eOHLGKi4utO++804qOjrYOHDgQqIcQML6uXXl5uRUbG2vNmDHDKi0ttd5//32rS5cu1pNPPhmoh9BixIcfvfTSS1ZaWpplt9utIUOGWNu3b/fedsMNN1iTJ0/2fl1WVmZJumC74YYbzA8eBHxZu27dul107ebPn29+8CDgy9r93//9n9WrVy8rOjra6tixo+VyuayCgoIATB0cfFm7/3W5x4dl+bZ+s2fP9h6bmJho3XLLLdbu3bsDMHVw8PVnb9u2bdbQoUOtqKgoq2fPntZvf/tb69y5c4an9h+bZVlWoJ51AQAAlx+u+QAAAEYRHwAAwCjiAwAAGEV8AAAAo4gPAABgFPEBAACMIj4AAIBRxAcAADCK+AAAAEYRHwAAwCjiAwAAGEV8AAAAo/4fPmAIwU/CtV0AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.close(\"all\")\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.hist([sol[-1].rho[:, 0], [rr.info.rho_estimate for rr in results]], label=['jax', 'osqp'])\n",
    "ax.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "09bfea67-ee9c-4bc5-916f-cb1cef13a2fa",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.34 s ± 81.3 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'sol_mine' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[39], line 9\u001b[0m\n\u001b[1;32m      6\u001b[0m sol \u001b[38;5;241m=\u001b[39m solve_qp(P, q, A, l, u)\n\u001b[1;32m      7\u001b[0m get_ipython()\u001b[38;5;241m.\u001b[39mrun_line_magic(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtimeit\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124msol = solve_qp(P, q, A, l, u)\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m----> 9\u001b[0m sol_mine[\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m]\u001b[38;5;241m.\u001b[39mx \u001b[38;5;241m-\u001b[39m sol\u001b[38;5;241m.\u001b[39mprimal[\u001b[38;5;241m0\u001b[39m]\n",
      "\u001b[0;31mNameError\u001b[0m: name 'sol_mine' is not defined"
     ]
    }
   ],
   "source": [
    "from jaxopt import BoxOSQP\n",
    "\n",
    "qp = BoxOSQP(jit=True)\n",
    "solve_qp = jax.jit(jax.vmap(lambda P, q, A, l, u: qp.run(params_obj=(P, q), params_eq=A, params_ineq=(l,u)).params))\n",
    "\n",
    "sol = solve_qp(P, q, A, l, u)\n",
    "%timeit sol = solve_qp(P, q, A, l, u)\n",
    "\n",
    "sol_mine[-1].x - sol.primal[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "id": "2fda995d-e431-4a20-88ab-893026846eff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "osqp stats | mean: 33.30000305175781, median: 25.0, std: 20.520965576171875, max: 325\n",
      "my stats | mean: 32.10000228881836, median: 26.0, std: 14.063783645629883, max: 201\n"
     ]
    }
   ],
   "source": [
    "osqp_iters = jnp.array(osqp_iters)\n",
    "my_iters = sol[0]\n",
    "\n",
    "print(f\"osqp stats | mean: {jnp.mean(osqp_iters)}, median: {jnp.median(osqp_iters)}, std: {jnp.std(osqp_iters)}, max: {jnp.max(osqp_iters)}\")\n",
    "print(f\"my stats | mean: {jnp.mean(my_iters)}, median: {jnp.median(my_iters)}, std: {jnp.std(my_iters)}, max: {jnp.max(my_iters)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "id": "092e2ce1-be60-4c1d-a1e7-0d6598ca6759",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([-0.48711722,  0.05135288]),\n",
       " Array([-0.4871236 ,  0.05135123], dtype=float32),\n",
       " 75)"
      ]
     },
     "execution_count": 291,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results[0].x, sol[-1].x[0], results[0].info.iter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "id": "bdbfea3f-9b4c-4e7d-82f3-242cd05db877",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 572 ms, sys: 11.5 ms, total: 583 ms\n",
      "Wall time: 588 ms\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Array([[0.1]], dtype=float32, weak_type=True)"
      ]
     },
     "execution_count": 251,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Bug hunting\n",
    "P = jnp.array([[[0.61585732]]])\n",
    "q = jnp.array([[0.1930774]])\n",
    "A = jnp.array([[[0.00870719]]])\n",
    "l = jnp.array([[-0.1864503]])\n",
    "u = jnp.array([[0.88669298]])\n",
    "\n",
    "prob, data, state = osqp.OSQPProblem.from_data(P, q, A, l, u)\n",
    "%time sol = jax.vmap(prob.solve)(data, state)\n",
    "sol[-1].z\n",
    "\n",
    "jax_obj = 0.5 * jnp.dot(sol[-1].x[0], P[0] @ sol[-1].x[0]) + jnp.dot(q[0], sol[-1].x[0])\n",
    "# osqp_obj = 0.5 * np.dot(results[0].x, P[0] @ results[0].x) + np.dot(q[0], results[0].x)\n",
    "\n",
    "sol[-1].rho"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "id": "c9f9beb3-37d9-4a9f-adc6-d090df5cd600",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Array([26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26,\n",
       "        26, 51, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 51, 26, 26,\n",
       "        26, 51, 51, 51, 26, 26, 26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26,\n",
       "        51, 51, 51, 26, 26, 26, 26, 51, 51, 26, 51, 26, 26, 51, 26, 51, 26,\n",
       "        26, 26, 26, 51, 26, 26, 51, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 51, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 51, 51, 26, 26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26, 51,\n",
       "        26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 76, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 26, 26, 26, 26, 51, 26, 26, 51, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 51,\n",
       "        26, 26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26, 51, 26, 51, 26, 26,\n",
       "        26, 26, 26, 51, 51, 26, 26, 26, 51, 26, 26, 26, 51, 26, 51, 26, 26,\n",
       "        26, 26, 26, 26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 51,\n",
       "        26, 76, 26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 51, 26, 26, 26,\n",
       "        26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 51, 26, 26, 26, 26,\n",
       "        26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 76, 26, 51, 26, 26, 26, 26,\n",
       "        26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 51, 51, 51, 26,\n",
       "        26, 26, 26, 26, 51, 26, 26, 51, 51, 26, 26, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26,\n",
       "        26, 26, 26, 26, 26, 26, 26, 51, 51, 26, 26, 26, 51, 26, 26, 26, 26,\n",
       "        26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 51, 26, 26,\n",
       "        26, 26, 26, 26, 26, 51, 51, 26, 26, 26, 51, 26, 26, 26, 26, 51, 26,\n",
       "        26, 26, 26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 51, 26, 51, 26, 26,\n",
       "        26, 26, 26, 26, 26, 51, 26, 26, 26, 26, 26, 26, 26, 51, 51, 26, 26,\n",
       "        26, 26, 26, 51, 26, 51, 26, 26, 26, 26, 26, 26, 51, 26, 26, 26, 51,\n",
       "        26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 51, 51, 26,\n",
       "        51, 26, 26, 26, 26, 26, 26], dtype=int32, weak_type=True),\n",
       " [50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  200,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  125,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  100,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  125,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  25,\n",
       "  100,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  100,\n",
       "  50,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  175,\n",
       "  50,\n",
       "  75,\n",
       "  100,\n",
       "  200,\n",
       "  100,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  100,\n",
       "  50,\n",
       "  50,\n",
       "  100,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  125,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  100,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  100,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  100,\n",
       "  50,\n",
       "  200,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  125,\n",
       "  100,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  100,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  100,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  75,\n",
       "  100,\n",
       "  50,\n",
       "  100,\n",
       "  75,\n",
       "  50,\n",
       "  100,\n",
       "  75,\n",
       "  125,\n",
       "  50,\n",
       "  25,\n",
       "  125,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  100,\n",
       "  100,\n",
       "  25,\n",
       "  75,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  125,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  75,\n",
       "  100,\n",
       "  25,\n",
       "  100,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  100,\n",
       "  100,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  150,\n",
       "  75,\n",
       "  75,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  75,\n",
       "  75,\n",
       "  175,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  225,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  150,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  125,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  125,\n",
       "  100,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  100,\n",
       "  100,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  100,\n",
       "  75,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  100,\n",
       "  100,\n",
       "  75,\n",
       "  150,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  100,\n",
       "  50,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  75,\n",
       "  75,\n",
       "  150,\n",
       "  100,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  100,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  125,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  100,\n",
       "  100,\n",
       "  50,\n",
       "  75,\n",
       "  50,\n",
       "  75,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  75,\n",
       "  25,\n",
       "  125,\n",
       "  125,\n",
       "  75,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50])"
      ]
     },
     "execution_count": 308,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sol[0], osqp_iters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "ea731c03-546f-4837-9ba9-686afe18f056",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[159], line 23\u001b[0m\n\u001b[1;32m     21\u001b[0m error \u001b[38;5;241m=\u001b[39m jnp\u001b[38;5;241m.\u001b[39mlinalg\u001b[38;5;241m.\u001b[39mnorm(state\u001b[38;5;241m.\u001b[39mx[jj] \u001b[38;5;241m-\u001b[39m results[jj]\u001b[38;5;241m.\u001b[39mx)\n\u001b[1;32m     22\u001b[0m y_error \u001b[38;5;241m=\u001b[39m jnp\u001b[38;5;241m.\u001b[39mlinalg\u001b[38;5;241m.\u001b[39mnorm(state\u001b[38;5;241m.\u001b[39my[jj] \u001b[38;5;241m-\u001b[39m results[jj]\u001b[38;5;241m.\u001b[39my)\n\u001b[0;32m---> 23\u001b[0m div_x \u001b[38;5;241m=\u001b[39m div_x\u001b[38;5;241m.\u001b[39mat[jj, ii]\u001b[38;5;241m.\u001b[39mset(error)\n\u001b[1;32m     24\u001b[0m div_y \u001b[38;5;241m=\u001b[39m div_y\u001b[38;5;241m.\u001b[39mat[jj, ii]\u001b[38;5;241m.\u001b[39mset(y_error)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/numpy/array_methods.py:485\u001b[0m, in \u001b[0;36mallow_pass_by_position_with_warning.<locals>.wrapped\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    483\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m f(\u001b[38;5;241m*\u001b[39margs[:n_positional], \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mconverted_kwargs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[1;32m    484\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m--> 485\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m f(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/numpy/array_methods.py:532\u001b[0m, in \u001b[0;36m_IndexUpdateRef.set\u001b[0;34m(self, values, indices_are_sorted, unique_indices, mode)\u001b[0m\n\u001b[1;32m    522\u001b[0m \u001b[38;5;129m@allow_pass_by_position_with_warning\u001b[39m\n\u001b[1;32m    523\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mset\u001b[39m(\u001b[38;5;28mself\u001b[39m, values, \u001b[38;5;241m*\u001b[39m, indices_are_sorted\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m, unique_indices\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m,\n\u001b[1;32m    524\u001b[0m         mode\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m):\n\u001b[1;32m    525\u001b[0m \u001b[38;5;250m  \u001b[39m\u001b[38;5;124;03m\"\"\"Pure equivalent of ``x[idx] = y``.\u001b[39;00m\n\u001b[1;32m    526\u001b[0m \n\u001b[1;32m    527\u001b[0m \u001b[38;5;124;03m  Returns the value of ``x`` that would result from the NumPy-style\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    530\u001b[0m \u001b[38;5;124;03m  See :mod:`jax.ops` for details.\u001b[39;00m\n\u001b[1;32m    531\u001b[0m \u001b[38;5;124;03m  \"\"\"\u001b[39;00m\n\u001b[0;32m--> 532\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m scatter\u001b[38;5;241m.\u001b[39m_scatter_update(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39marray, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mindex, values, lax\u001b[38;5;241m.\u001b[39mscatter,\n\u001b[1;32m    533\u001b[0m                                  indices_are_sorted\u001b[38;5;241m=\u001b[39mindices_are_sorted,\n\u001b[1;32m    534\u001b[0m                                  unique_indices\u001b[38;5;241m=\u001b[39munique_indices, mode\u001b[38;5;241m=\u001b[39mmode)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/ops/scatter.py:76\u001b[0m, in \u001b[0;36m_scatter_update\u001b[0;34m(x, idx, y, scatter_op, indices_are_sorted, unique_indices, mode, normalize_indices)\u001b[0m\n\u001b[1;32m     73\u001b[0m \u001b[38;5;66;03m# XLA gathers and scatters are very similar in structure; the scatter logic\u001b[39;00m\n\u001b[1;32m     74\u001b[0m \u001b[38;5;66;03m# is more or less a transpose of the gather equivalent.\u001b[39;00m\n\u001b[1;32m     75\u001b[0m treedef, static_idx, dynamic_idx \u001b[38;5;241m=\u001b[39m jnp\u001b[38;5;241m.\u001b[39m_split_index_for_jit(idx, x\u001b[38;5;241m.\u001b[39mshape)\n\u001b[0;32m---> 76\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m _scatter_impl(x, y, scatter_op, treedef, static_idx, dynamic_idx,\n\u001b[1;32m     77\u001b[0m                      indices_are_sorted, unique_indices, mode,\n\u001b[1;32m     78\u001b[0m                      normalize_indices)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/ops/scatter.py:111\u001b[0m, in \u001b[0;36m_scatter_impl\u001b[0;34m(x, y, scatter_op, treedef, static_idx, dynamic_idx, indices_are_sorted, unique_indices, mode, normalize_indices)\u001b[0m\n\u001b[1;32m    109\u001b[0m y \u001b[38;5;241m=\u001b[39m jnp\u001b[38;5;241m.\u001b[39mbroadcast_to(y, \u001b[38;5;28mtuple\u001b[39m(indexer\u001b[38;5;241m.\u001b[39mslice_shape))\n\u001b[1;32m    110\u001b[0m \u001b[38;5;66;03m# Collapse any `None`/`jnp.newaxis` dimensions.\u001b[39;00m\n\u001b[0;32m--> 111\u001b[0m y \u001b[38;5;241m=\u001b[39m jnp\u001b[38;5;241m.\u001b[39msqueeze(y, axis\u001b[38;5;241m=\u001b[39mindexer\u001b[38;5;241m.\u001b[39mnewaxis_dims)\n\u001b[1;32m    112\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m indexer\u001b[38;5;241m.\u001b[39mreversed_y_dims:\n\u001b[1;32m    113\u001b[0m   y \u001b[38;5;241m=\u001b[39m lax\u001b[38;5;241m.\u001b[39mrev(y, indexer\u001b[38;5;241m.\u001b[39mreversed_y_dims)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/numpy/lax_numpy.py:844\u001b[0m, in \u001b[0;36msqueeze\u001b[0;34m(a, axis)\u001b[0m\n\u001b[1;32m    841\u001b[0m \u001b[38;5;129m@util\u001b[39m\u001b[38;5;241m.\u001b[39m_wraps(np\u001b[38;5;241m.\u001b[39msqueeze, lax_description\u001b[38;5;241m=\u001b[39m_ARRAY_VIEW_DOC)\n\u001b[1;32m    842\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21msqueeze\u001b[39m(a: ArrayLike, axis: Optional[Union[\u001b[38;5;28mint\u001b[39m, Tuple[\u001b[38;5;28mint\u001b[39m, \u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m]]] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Array:\n\u001b[1;32m    843\u001b[0m   util\u001b[38;5;241m.\u001b[39mcheck_arraylike(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124msqueeze\u001b[39m\u001b[38;5;124m\"\u001b[39m, a)\n\u001b[0;32m--> 844\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m _squeeze(asarray(a), _ensure_index_tuple(axis) \u001b[38;5;28;01mif\u001b[39;00m axis \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m)\n",
      "    \u001b[0;31m[... skipping hidden 1 frame]\u001b[0m\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/pjit.py:208\u001b[0m, in \u001b[0;36m_cpp_pjit.<locals>.cache_miss\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    206\u001b[0m \u001b[38;5;129m@api_boundary\u001b[39m\n\u001b[1;32m    207\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mcache_miss\u001b[39m(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[0;32m--> 208\u001b[0m   outs, out_flat, out_tree, args_flat \u001b[38;5;241m=\u001b[39m _python_pjit_helper(\n\u001b[1;32m    209\u001b[0m       fun, infer_params_fn, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[1;32m    211\u001b[0m   executable \u001b[38;5;241m=\u001b[39m _read_most_recent_pjit_call_executable()\n\u001b[1;32m    213\u001b[0m   use_fastpath \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m    214\u001b[0m       executable \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m\n\u001b[1;32m    215\u001b[0m       \u001b[38;5;28misinstance\u001b[39m(executable, pxla\u001b[38;5;241m.\u001b[39mMeshExecutable) \u001b[38;5;129;01mand\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    221\u001b[0m       \u001b[38;5;28mall\u001b[39m(\u001b[38;5;28misinstance\u001b[39m(x, xc\u001b[38;5;241m.\u001b[39mArrayImpl) \u001b[38;5;28;01mfor\u001b[39;00m x \u001b[38;5;129;01min\u001b[39;00m out_flat)\n\u001b[1;32m    222\u001b[0m   )\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/pjit.py:155\u001b[0m, in \u001b[0;36m_python_pjit_helper\u001b[0;34m(fun, infer_params_fn, *args, **kwargs)\u001b[0m\n\u001b[1;32m    153\u001b[0m   dispatch\u001b[38;5;241m.\u001b[39mcheck_arg(arg)\n\u001b[1;32m    154\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 155\u001b[0m   out_flat \u001b[38;5;241m=\u001b[39m pjit_p\u001b[38;5;241m.\u001b[39mbind(\u001b[38;5;241m*\u001b[39margs_flat, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mparams)\n\u001b[1;32m    156\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m pxla\u001b[38;5;241m.\u001b[39mDeviceAssignmentMismatchError \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m    157\u001b[0m   fails, \u001b[38;5;241m=\u001b[39m e\u001b[38;5;241m.\u001b[39margs\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/core.py:2633\u001b[0m, in \u001b[0;36mAxisPrimitive.bind\u001b[0;34m(self, *args, **params)\u001b[0m\n\u001b[1;32m   2629\u001b[0m axis_main \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mmax\u001b[39m((axis_frame(a)\u001b[38;5;241m.\u001b[39mmain_trace \u001b[38;5;28;01mfor\u001b[39;00m a \u001b[38;5;129;01min\u001b[39;00m used_axis_names(\u001b[38;5;28mself\u001b[39m, params)),\n\u001b[1;32m   2630\u001b[0m                 default\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, key\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mlambda\u001b[39;00m t: \u001b[38;5;28mgetattr\u001b[39m(t, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mlevel\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m))\n\u001b[1;32m   2631\u001b[0m top_trace \u001b[38;5;241m=\u001b[39m (top_trace \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m axis_main \u001b[38;5;129;01mor\u001b[39;00m axis_main\u001b[38;5;241m.\u001b[39mlevel \u001b[38;5;241m<\u001b[39m top_trace\u001b[38;5;241m.\u001b[39mlevel\n\u001b[1;32m   2632\u001b[0m              \u001b[38;5;28;01melse\u001b[39;00m axis_main\u001b[38;5;241m.\u001b[39mwith_cur_sublevel())\n\u001b[0;32m-> 2633\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mbind_with_trace(top_trace, args, params)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/core.py:383\u001b[0m, in \u001b[0;36mPrimitive.bind_with_trace\u001b[0;34m(self, trace, args, params)\u001b[0m\n\u001b[1;32m    382\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mbind_with_trace\u001b[39m(\u001b[38;5;28mself\u001b[39m, trace, args, params):\n\u001b[0;32m--> 383\u001b[0m   out \u001b[38;5;241m=\u001b[39m trace\u001b[38;5;241m.\u001b[39mprocess_primitive(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;28mmap\u001b[39m(trace\u001b[38;5;241m.\u001b[39mfull_raise, args), params)\n\u001b[1;32m    384\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mmap\u001b[39m(full_lower, out) \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmultiple_results \u001b[38;5;28;01melse\u001b[39;00m full_lower(out)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/core.py:790\u001b[0m, in \u001b[0;36mEvalTrace.process_primitive\u001b[0;34m(self, primitive, tracers, params)\u001b[0m\n\u001b[1;32m    789\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mprocess_primitive\u001b[39m(\u001b[38;5;28mself\u001b[39m, primitive, tracers, params):\n\u001b[0;32m--> 790\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m primitive\u001b[38;5;241m.\u001b[39mimpl(\u001b[38;5;241m*\u001b[39mtracers, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mparams)\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/pjit.py:1081\u001b[0m, in \u001b[0;36m_pjit_call_impl\u001b[0;34m(jaxpr, in_shardings, out_shardings, resource_env, donated_invars, name, keep_unused, inline, *args)\u001b[0m\n\u001b[1;32m   1076\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_pjit_call_impl\u001b[39m(\u001b[38;5;241m*\u001b[39margs, jaxpr,\n\u001b[1;32m   1077\u001b[0m                     in_shardings, out_shardings, resource_env,\n\u001b[1;32m   1078\u001b[0m                     donated_invars, name, keep_unused, inline):\n\u001b[1;32m   1079\u001b[0m   \u001b[38;5;28;01mglobal\u001b[39;00m _most_recent_pjit_call_executable\n\u001b[0;32m-> 1081\u001b[0m   in_shardings \u001b[38;5;241m=\u001b[39m _resolve_in_shardings(\n\u001b[1;32m   1082\u001b[0m       args, in_shardings, out_shardings,\n\u001b[1;32m   1083\u001b[0m       resource_env\u001b[38;5;241m.\u001b[39mphysical_mesh \u001b[38;5;28;01mif\u001b[39;00m resource_env \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m)\n\u001b[1;32m   1085\u001b[0m   compiled \u001b[38;5;241m=\u001b[39m _pjit_lower(\n\u001b[1;32m   1086\u001b[0m       jaxpr, in_shardings, out_shardings, resource_env,\n\u001b[1;32m   1087\u001b[0m       donated_invars, name, keep_unused, inline,\n\u001b[1;32m   1088\u001b[0m       always_lower\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m, lowering_platform\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m)\u001b[38;5;241m.\u001b[39mcompile()\n\u001b[1;32m   1089\u001b[0m   _most_recent_pjit_call_executable\u001b[38;5;241m.\u001b[39mvalue \u001b[38;5;241m=\u001b[39m compiled\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/pjit.py:1018\u001b[0m, in \u001b[0;36m_resolve_in_shardings\u001b[0;34m(args, pjit_in_shardings, out_shardings, pjit_mesh)\u001b[0m\n\u001b[1;32m   1010\u001b[0m       committed_arg_shardings\u001b[38;5;241m.\u001b[39mappend((arg_s, pxla\u001b[38;5;241m.\u001b[39mMismatchType\u001b[38;5;241m.\u001b[39mARG_SHARDING, \u001b[38;5;28;01mNone\u001b[39;00m))\n\u001b[1;32m   1012\u001b[0m \u001b[38;5;66;03m# Check if the device_assignment across inputs, outputs and arguments is the\u001b[39;00m\n\u001b[1;32m   1013\u001b[0m \u001b[38;5;66;03m# same.\u001b[39;00m\n\u001b[1;32m   1014\u001b[0m pxla\u001b[38;5;241m.\u001b[39m_get_and_check_device_assignment(\n\u001b[1;32m   1015\u001b[0m     it\u001b[38;5;241m.\u001b[39mchain(\n\u001b[1;32m   1016\u001b[0m         committed_arg_shardings,\n\u001b[1;32m   1017\u001b[0m         [(i, pxla\u001b[38;5;241m.\u001b[39mMismatchType\u001b[38;5;241m.\u001b[39mIN_SHARDING, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m pjit_in_shardings],\n\u001b[0;32m-> 1018\u001b[0m         [(o, pxla\u001b[38;5;241m.\u001b[39mMismatchType\u001b[38;5;241m.\u001b[39mOUT_SHARDING, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;28;01mfor\u001b[39;00m o \u001b[38;5;129;01min\u001b[39;00m out_shardings]),\n\u001b[1;32m   1019\u001b[0m     (\u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01mif\u001b[39;00m pjit_mesh \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mor\u001b[39;00m pjit_mesh\u001b[38;5;241m.\u001b[39mempty \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mlist\u001b[39m(pjit_mesh\u001b[38;5;241m.\u001b[39mdevices\u001b[38;5;241m.\u001b[39mflat)))\n\u001b[1;32m   1021\u001b[0m resolved_in_shardings \u001b[38;5;241m=\u001b[39m []\n\u001b[1;32m   1022\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m arg, pjit_in_s \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(args, pjit_in_shardings):\n",
      "File \u001b[0;32m~/miniconda3/envs/jaxosqp/lib/python3.11/site-packages/jax/_src/pjit.py:1018\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m   1010\u001b[0m       committed_arg_shardings\u001b[38;5;241m.\u001b[39mappend((arg_s, pxla\u001b[38;5;241m.\u001b[39mMismatchType\u001b[38;5;241m.\u001b[39mARG_SHARDING, \u001b[38;5;28;01mNone\u001b[39;00m))\n\u001b[1;32m   1012\u001b[0m \u001b[38;5;66;03m# Check if the device_assignment across inputs, outputs and arguments is the\u001b[39;00m\n\u001b[1;32m   1013\u001b[0m \u001b[38;5;66;03m# same.\u001b[39;00m\n\u001b[1;32m   1014\u001b[0m pxla\u001b[38;5;241m.\u001b[39m_get_and_check_device_assignment(\n\u001b[1;32m   1015\u001b[0m     it\u001b[38;5;241m.\u001b[39mchain(\n\u001b[1;32m   1016\u001b[0m         committed_arg_shardings,\n\u001b[1;32m   1017\u001b[0m         [(i, pxla\u001b[38;5;241m.\u001b[39mMismatchType\u001b[38;5;241m.\u001b[39mIN_SHARDING, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m pjit_in_shardings],\n\u001b[0;32m-> 1018\u001b[0m         [(o, pxla\u001b[38;5;241m.\u001b[39mMismatchType\u001b[38;5;241m.\u001b[39mOUT_SHARDING, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;28;01mfor\u001b[39;00m o \u001b[38;5;129;01min\u001b[39;00m out_shardings]),\n\u001b[1;32m   1019\u001b[0m     (\u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01mif\u001b[39;00m pjit_mesh \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mor\u001b[39;00m pjit_mesh\u001b[38;5;241m.\u001b[39mempty \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mlist\u001b[39m(pjit_mesh\u001b[38;5;241m.\u001b[39mdevices\u001b[38;5;241m.\u001b[39mflat)))\n\u001b[1;32m   1021\u001b[0m resolved_in_shardings \u001b[38;5;241m=\u001b[39m []\n\u001b[1;32m   1022\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m arg, pjit_in_s \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(args, pjit_in_shardings):\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# Comparing how the solutions diverge\n",
    "prob, data, state_scaled = osqp.OSQPProblem.from_data(P, q, A, l, u, config=osqp.OSQPConfig(max_iters=1))\n",
    "\n",
    "results = []\n",
    "T = 250\n",
    "div_x = jnp.zeros((B, T))\n",
    "div_y = jnp.zeros((B, T))\n",
    "# [mm.update_settings(verbose=True) for mm in probs]\n",
    "for ii in range(T):\n",
    "    probs = []\n",
    "    setup_all(probs)\n",
    "    results = []\n",
    "    [mm.update_settings(max_iter=ii+1) for mm in probs]\n",
    "#         for jj in range(B):\n",
    "#             probs[jj].warm_start(x=np.asarray(state.x[jj]), y=np.asarray(state.y[jj]))\n",
    "        \n",
    "    state_scaled = jax.vmap(prob.step)(data, state_scaled)\n",
    "    state = jax.vmap(prob.unscale_vars)(data, state_scaled)\n",
    "    solve_all(probs, results)\n",
    "    for jj in range(B):\n",
    "        error = jnp.linalg.norm(state.x[jj] - results[jj].x)\n",
    "        y_error = jnp.linalg.norm(state.y[jj] - results[jj].y)\n",
    "        div_x = div_x.at[jj, ii].set(error)\n",
    "        div_y = div_y.at[jj, ii].set(y_error)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "69ec815b-86ee-427a-bc87-30056f7bfad5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'parameter differences jax vx. OSQP')"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "plt.close(\"all\")\n",
    "fig, ax = plt.subplots()\n",
    "divx_mean = jnp.mean(div_x, axis=0)\n",
    "divx_std = jnp.std(div_x, axis=0)\n",
    "divy_mean = jnp.mean(div_y, axis=0)\n",
    "divy_std = jnp.std(div_y, axis=0)\n",
    "ax.plot(jnp.arange(len(divx_mean)), divx_mean, label='x error')\n",
    "ax.fill_between(jnp.arange(len(divx_mean)), divx_mean - 2 * divx_std, divx_mean + 2 * divx_std, alpha=0.1)\n",
    "ax.plot(jnp.arange(len(divy_mean)), divy_mean, label='y error')\n",
    "ax.fill_between(jnp.arange(len(divy_mean)), divy_mean - 2 * divy_std, divy_mean + 2 * divy_std, alpha=0.1)\n",
    "ax.legend()\n",
    "ax.set_xlabel('iters')\n",
    "ax.set_ylabel('l2 error')\n",
    "ax.set_title('parameter differences jax vx. OSQP')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "id": "f4c9ab92-c8dc-4020-a4a1-931e4c394123",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  100,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  75,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  75,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  50,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25,\n",
       "  25,\n",
       "  50,\n",
       "  25,\n",
       "  25],\n",
       " Array([False], dtype=bool))"
      ]
     },
     "execution_count": 309,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[rr.info.iter for rr in results], state.converged"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "4890d782-0087-4b52-8a80-0f607777eaec",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Array([51, 26, 76, 26, 26], dtype=int64, weak_type=True),\n",
       " OSQPData(P=Array([[[ 1.        ,  0.04953851,  0.14390485, ..., -0.09622749,\n",
       "          -0.02397834,  0.36724911],\n",
       "         [ 0.04953851,  1.        , -0.09837497, ..., -0.26287997,\n",
       "          -0.0879546 , -0.30567721],\n",
       "         [ 0.14390485, -0.09837497,  1.        , ...,  0.1008785 ,\n",
       "           0.59088617,  0.18411601],\n",
       "         ...,\n",
       "         [-0.09622749, -0.26287997,  0.1008785 , ...,  1.        ,\n",
       "          -0.15360153,  0.20523239],\n",
       "         [-0.02397834, -0.0879546 ,  0.59088617, ..., -0.15360153,\n",
       "           1.        , -0.25783934],\n",
       "         [ 0.36724911, -0.30567721,  0.18411601, ...,  0.20523239,\n",
       "          -0.25783934,  1.        ]],\n",
       " \n",
       "        [[ 1.        , -0.25564654,  0.13749279, ..., -0.14170522,\n",
       "          -0.16413242, -0.38841411],\n",
       "         [-0.25564654,  1.        , -0.20499922, ...,  0.40492141,\n",
       "          -0.00808829, -0.21062601],\n",
       "         [ 0.13749279, -0.20499922,  1.        , ...,  0.39414183,\n",
       "           0.02563283, -0.03257141],\n",
       "         ...,\n",
       "         [-0.14170522,  0.40492141,  0.39414183, ...,  1.        ,\n",
       "          -0.00494682,  0.16319483],\n",
       "         [-0.16413242, -0.00808829,  0.02563283, ..., -0.00494682,\n",
       "           1.        ,  0.06249207],\n",
       "         [-0.38841411, -0.21062601, -0.03257141, ...,  0.16319483,\n",
       "           0.06249207,  1.        ]],\n",
       " \n",
       "        [[ 1.        ,  0.31734322, -0.29633194, ..., -0.13019276,\n",
       "           0.05334407,  0.24510536],\n",
       "         [ 0.31734322,  1.        ,  0.33657715, ...,  0.26357683,\n",
       "           0.07397066,  0.31023319],\n",
       "         [-0.29633194,  0.33657715,  1.        , ...,  0.42667685,\n",
       "           0.11165038,  0.17020207],\n",
       "         ...,\n",
       "         [-0.13019276,  0.26357683,  0.42667685, ...,  1.        ,\n",
       "          -0.19261803,  0.10421181],\n",
       "         [ 0.05334407,  0.07397066,  0.11165038, ..., -0.19261803,\n",
       "           1.        ,  0.00380238],\n",
       "         [ 0.24510536,  0.31023319,  0.17020207, ...,  0.10421181,\n",
       "           0.00380238,  1.        ]],\n",
       " \n",
       "        [[ 1.        ,  0.12596388, -0.1163087 , ...,  0.06417903,\n",
       "          -0.3528719 , -0.42261649],\n",
       "         [ 0.12596388,  1.        , -0.04117697, ..., -0.19325742,\n",
       "           0.17516661, -0.28162416],\n",
       "         [-0.1163087 , -0.04117697,  1.        , ..., -0.33185398,\n",
       "          -0.20133654, -0.36589102],\n",
       "         ...,\n",
       "         [ 0.06417903, -0.19325742, -0.33185398, ...,  1.        ,\n",
       "           0.05365754,  0.20900683],\n",
       "         [-0.3528719 ,  0.17516661, -0.20133654, ...,  0.05365754,\n",
       "           1.        ,  0.22349076],\n",
       "         [-0.42261649, -0.28162416, -0.36589102, ...,  0.20900683,\n",
       "           0.22349076,  1.        ]],\n",
       " \n",
       "        [[ 1.        ,  0.12563016, -0.01844979, ..., -0.00758674,\n",
       "           0.01103505, -0.03704892],\n",
       "         [ 0.12563016,  1.        , -0.23084397, ...,  0.1521826 ,\n",
       "           0.01715459, -0.29035873],\n",
       "         [-0.01844979, -0.23084397,  1.        , ...,  0.26360626,\n",
       "           0.10921537, -0.05892267],\n",
       "         ...,\n",
       "         [-0.00758674,  0.1521826 ,  0.26360626, ...,  1.        ,\n",
       "           0.11207065, -0.17067657],\n",
       "         [ 0.01103505,  0.01715459,  0.10921537, ...,  0.11207065,\n",
       "           1.        ,  0.02241457],\n",
       "         [-0.03704892, -0.29035873, -0.05892267, ..., -0.17067657,\n",
       "           0.02241457,  1.        ]]], dtype=float64), q=Array([[-0.00999712, -0.06237303,  0.46090481, -0.02704644, -0.09776588,\n",
       "          0.10878925, -0.1358791 ,  0.06885418, -0.07373864, -0.00327073,\n",
       "         -0.22518864, -0.40960507,  0.11492542,  0.24800123, -0.06066784,\n",
       "          0.49825806,  0.03493108, -0.26621934,  0.16796181, -0.66776625],\n",
       "        [-0.04413102,  0.2641059 , -0.1513174 , -0.29025892, -0.09048927,\n",
       "         -0.04821666, -0.26169809, -0.18437177,  0.37085535, -0.00548435,\n",
       "         -0.11791084,  0.01451673,  0.12289089,  0.06424743, -0.49909932,\n",
       "          0.15264358, -0.3139832 , -0.20359037,  0.03211793,  0.11452213],\n",
       "        [-0.20347391,  0.27310179, -0.06056221,  0.06562776,  0.25199257,\n",
       "         -0.22626004, -0.08297851,  0.37501188,  0.11787947, -0.2263122 ,\n",
       "         -0.66446975, -0.18737472,  0.12014142, -0.34972181, -0.16767536,\n",
       "          0.51464728, -0.35234811, -0.04157922, -0.01176228, -0.17113946],\n",
       "        [-0.00304147, -0.24140192,  0.33294678,  0.41080247, -0.0320878 ,\n",
       "         -0.23987077,  0.33385701, -0.09682153, -0.45765387,  0.23591895,\n",
       "          0.61885598, -0.07107581,  0.19040377, -0.24833233, -0.0401516 ,\n",
       "          0.11646527,  0.01117732, -0.16805338,  0.15824826, -0.02144344],\n",
       "        [-0.41925787,  0.02474373,  0.28755977,  0.03172297, -0.35719991,\n",
       "          0.23876021, -0.21058449,  0.23047672,  0.34906865, -0.13750204,\n",
       "          0.15311109,  0.18821594, -0.10627047,  0.28340234, -0.5708202 ,\n",
       "          0.24683457,  0.21769987,  0.1451698 , -0.16709591,  0.31724457]],      dtype=float64), A=Array([[[-3.52892713e-01, -1.27339953e-01,  3.04707237e-01, ...,\n",
       "           4.87279734e-01,  3.52905568e-01, -1.03282607e-01],\n",
       "         [-6.83232325e-01,  0.00000000e+00,  4.67616806e-01, ...,\n",
       "           8.88252484e-02, -4.24860187e-01,  2.40008906e-01],\n",
       "         [ 3.14859288e-01, -2.23322431e-01,  4.65457138e-01, ...,\n",
       "           1.68737925e-01,  6.36377679e-03, -2.30883009e-01],\n",
       "         ...,\n",
       "         [-3.89777807e-01, -8.48216699e-01,  6.90905001e-01, ...,\n",
       "          -5.50007502e-01,  3.08854375e-01,  9.97710090e-01],\n",
       "         [ 2.44187474e-01, -9.52320631e-02, -3.91430375e-01, ...,\n",
       "          -3.63659307e-02,  1.82010690e-01,  0.00000000e+00],\n",
       "         [ 4.24202845e-01, -3.22918048e-01, -4.79424924e-01, ...,\n",
       "           1.81921428e-01,  4.64983554e-01,  1.58748755e-01]],\n",
       " \n",
       "        [[ 3.91525333e-01,  4.51883754e-01, -7.17443881e-02, ...,\n",
       "           1.34909140e-01,  8.72857883e-02,  0.00000000e+00],\n",
       "         [ 4.17943867e-01, -3.45528910e-01,  1.10733796e-01, ...,\n",
       "           8.14007133e-01,  1.61624101e-01,  2.35280470e-01],\n",
       "         [ 8.31437794e-04,  5.91225999e-01, -1.90473071e-01, ...,\n",
       "           1.61962731e-01,  2.45218989e-02,  6.68617295e-02],\n",
       "         ...,\n",
       "         [-3.95472799e-01, -7.67938023e-02,  1.29652817e-01, ...,\n",
       "          -4.28261357e-01, -2.28831418e-02,  2.23467479e-01],\n",
       "         [-3.30643483e-01,  3.66823565e-02, -5.27200375e-01, ...,\n",
       "          -3.20786299e-01,  2.96360395e-01, -1.27697388e-01],\n",
       "         [-3.23894175e-01,  1.01437836e-01,  6.97175795e-01, ...,\n",
       "           5.17780208e-01, -9.78564557e-01, -2.41802046e-01]],\n",
       " \n",
       "        [[ 1.22695514e-02, -4.95301823e-01,  1.07722612e-01, ...,\n",
       "           3.36663307e-01,  3.27241982e-01,  3.07766016e-01],\n",
       "         [-7.97569028e-01, -4.49317438e-01, -5.34963719e-01, ...,\n",
       "          -6.25513136e-01, -5.41652909e-01, -9.98182187e-01],\n",
       "         [ 1.01564565e-01,  5.15494143e-01,  1.04830621e-01, ...,\n",
       "           7.60502714e-02,  5.32061756e-01, -7.74220024e-01],\n",
       "         ...,\n",
       "         [ 2.30760397e-01,  5.49094017e-01,  2.45397243e-02, ...,\n",
       "           3.56435204e-01,  5.26688092e-02, -3.51249541e-01],\n",
       "         [ 4.63124590e-02,  4.79058883e-01, -3.57273121e-01, ...,\n",
       "           2.80686219e-01,  5.86380635e-01, -2.25587865e-01],\n",
       "         [ 0.00000000e+00,  2.33904960e-01, -3.47614759e-01, ...,\n",
       "          -4.15225311e-01, -1.06517814e-02, -1.26641938e-01]],\n",
       " \n",
       "        [[-1.79860317e-01, -1.08642116e-01,  4.37166757e-01, ...,\n",
       "           9.57344117e-02,  7.12251756e-02,  2.63538695e-01],\n",
       "         [ 4.44278644e-01,  6.22288000e-01,  3.00419712e-01, ...,\n",
       "           2.16448847e-02, -3.43412778e-01, -9.26289543e-01],\n",
       "         [ 3.95473799e-01, -5.37364718e-02,  1.40838806e-01, ...,\n",
       "          -1.36445744e-01, -1.09349718e-01,  2.05512316e-01],\n",
       "         ...,\n",
       "         [ 8.71922947e-01,  7.37887146e-02,  3.00767248e-01, ...,\n",
       "           2.38436230e-01,  9.94711710e-02, -5.56067707e-01],\n",
       "         [ 8.03282141e-01,  5.48740267e-01, -4.09066549e-02, ...,\n",
       "           2.47849152e-01,  0.00000000e+00,  6.34193927e-02],\n",
       "         [ 3.22885928e-01, -7.08922008e-02,  0.00000000e+00, ...,\n",
       "           1.25846158e-01, -1.42706768e-01, -8.11130872e-01]],\n",
       " \n",
       "        [[ 0.00000000e+00,  1.25975472e-01,  4.71377243e-01, ...,\n",
       "          -1.09674371e-01, -6.23711025e-01,  9.63111704e-02],\n",
       "         [-3.23350492e-01, -9.97646523e-01, -1.76882563e-01, ...,\n",
       "           2.58748141e-01, -1.78281175e-01, -6.78506202e-02],\n",
       "         [ 1.28325380e-01, -1.49671030e-01,  1.17593705e-01, ...,\n",
       "          -4.58090308e-01, -9.97811386e-01, -6.26992261e-01],\n",
       "         ...,\n",
       "         [-5.28482213e-02,  1.53248930e-01, -3.34686579e-01, ...,\n",
       "          -2.61798265e-01,  5.97793578e-02,  5.40555456e-01],\n",
       "         [-2.07858588e-01,  1.12920116e-01,  4.91159185e-01, ...,\n",
       "           8.71019020e-03, -4.75399146e-01,  0.00000000e+00],\n",
       "         [ 6.09751418e-01,  1.08215135e-01,  0.00000000e+00, ...,\n",
       "           1.84613616e-01,  2.87399335e-01, -2.24706987e-01]]],      dtype=float64), l=Array([[-1.45461445, -1.75722213, -0.10634527, -0.46308247, -0.82953826,\n",
       "         -0.75946181, -0.62167168, -1.66578586, -1.6447167 , -0.16123679,\n",
       "         -0.52584045, -1.35511245, -1.36542642, -1.39809096, -0.30285512,\n",
       "         -2.7926459 , -1.1941947 , -0.78194825, -1.14854538, -0.80079461,\n",
       "         -1.16837899, -0.00369529, -0.09848964, -1.1176137 , -0.83557478,\n",
       "         -1.37365169, -0.93398786, -1.96400225, -0.07727236, -1.37228212],\n",
       "        [-2.04349382, -0.32387326, -0.99291847, -1.73242939, -0.28032049,\n",
       "         -0.7392608 , -0.12123767, -1.27161029, -1.94977518, -0.54949498,\n",
       "         -0.61055438, -1.91702651, -0.84930006, -1.11076211, -1.3614034 ,\n",
       "         -1.18631126, -1.59617081, -0.63608889, -0.83578644, -0.97975056,\n",
       "         -0.40441653, -2.26782114, -1.15738467, -0.09601576, -0.5916854 ,\n",
       "         -0.23660637, -0.85747174, -1.34871383, -1.54297901, -1.50089816],\n",
       "        [-0.66397915, -1.13652498, -0.09734754, -1.31255369, -0.40864066,\n",
       "         -0.93920197, -0.78034592, -0.8270752 , -1.16737153, -0.31468576,\n",
       "         -0.68267251, -1.12304981, -1.77755466, -0.12366476, -0.52819168,\n",
       "         -1.92935332, -0.66276295, -0.1545602 , -0.85414634, -1.33088357,\n",
       "         -1.9550898 , -1.40717666, -0.22399203, -1.42025685, -0.78909073,\n",
       "         -0.85593658, -1.14316211, -1.10270839, -0.85250916, -0.04888551],\n",
       "        [-1.64050655, -0.7108954 , -0.19457717, -0.16285537, -0.64435167,\n",
       "         -1.6042298 , -1.12858194, -0.76239622, -0.26593055, -2.18754503,\n",
       "         -0.45351858, -1.39541989, -1.01708896, -1.83397712, -1.04047605,\n",
       "         -2.36549238, -0.93997885, -0.70969817, -0.99794301, -1.39678324,\n",
       "         -1.19736999, -1.21880587, -1.67932458, -0.5739831 , -1.43657079,\n",
       "         -0.59634281, -0.47905596, -1.51261288, -1.16220243, -1.79235562],\n",
       "        [-1.62669243, -2.01746844, -1.86429417, -1.75581601, -1.28819674,\n",
       "         -1.47393758, -1.78667166, -1.39689411, -0.96253095, -0.40977285,\n",
       "         -0.42079518, -1.7502037 , -0.65641348, -0.04937464, -1.41734101,\n",
       "         -1.97737682, -1.31741273, -1.68735151, -0.24119839, -0.84687541,\n",
       "         -1.95851495, -1.63469657, -0.53606051, -1.30236273, -0.45554005,\n",
       "         -0.91563771, -0.42552188, -1.24317653, -1.13421089, -0.45678369]],      dtype=float64), u=Array([[0.73455094, 0.63716117, 0.04596918, 1.55815106, 1.05931708,\n",
       "         1.11395425, 1.67016792, 2.21040639, 1.0508755 , 0.15635299,\n",
       "         0.84864237, 0.13298601, 1.63322507, 2.47460504, 1.75263276,\n",
       "         0.12466848, 0.036233  , 0.52445204, 1.5319904 , 1.5608631 ,\n",
       "         1.30424943, 0.94553013, 0.72668991, 3.08402974, 1.16968541,\n",
       "         0.08898989, 0.34085389, 1.01049111, 1.26545719, 0.39708368],\n",
       "        [0.79837672, 1.18017688, 0.43883951, 2.60801502, 0.27360284,\n",
       "         0.18403407, 0.84844907, 1.54580371, 0.89150378, 0.99272909,\n",
       "         1.74935764, 2.07373852, 1.60064998, 0.17402182, 1.43812728,\n",
       "         1.65410257, 0.91180745, 0.53118128, 0.66063714, 0.4483506 ,\n",
       "         0.67130745, 0.30297416, 0.68117548, 1.18478644, 2.00695083,\n",
       "         1.00953979, 1.52203162, 0.77702134, 0.0171055 , 0.39046971],\n",
       "        [0.69641544, 0.54453126, 1.1919225 , 1.15052296, 1.16903858,\n",
       "         0.14726401, 0.19629297, 1.73823553, 0.94189914, 0.38445395,\n",
       "         0.65616698, 0.44609039, 1.45859722, 1.06638917, 0.65216742,\n",
       "         0.83183846, 1.63777569, 1.14924197, 0.77111788, 0.5946154 ,\n",
       "         0.52945232, 1.22502226, 0.82172699, 0.50360587, 1.44623243,\n",
       "         0.3990591 , 0.62829941, 0.29177674, 1.5394964 , 1.32832157],\n",
       "        [0.34134627, 2.3564841 , 1.44319306, 1.11815238, 1.13320216,\n",
       "         1.62418411, 0.7695469 , 0.43475429, 0.23040089, 0.44161484,\n",
       "         1.06245073, 0.75525904, 1.56767201, 0.5848292 , 1.15857219,\n",
       "         2.02186553, 1.10614312, 2.13732997, 0.02623413, 1.58148897,\n",
       "         1.14336766, 0.02494314, 1.75103578, 1.21572021, 0.82250189,\n",
       "         1.17468465, 1.4639664 , 1.43927342, 0.98945281, 0.15467769],\n",
       "        [0.10127169, 1.87920982, 0.4541108 , 1.85590721, 1.86267892,\n",
       "         0.42512449, 1.00420492, 1.26343088, 0.35146641, 0.79310818,\n",
       "         0.85845432, 1.9834382 , 2.04402518, 0.57883313, 0.15161696,\n",
       "         0.27508173, 0.60769408, 0.35843865, 0.45701307, 0.94459244,\n",
       "         0.79932891, 1.20664206, 1.53396214, 0.93802323, 1.01971262,\n",
       "         0.37943053, 0.76132145, 0.88143945, 0.8515312 , 1.50732427]],      dtype=float64), D=Array([[0.24961839, 0.20193924, 0.19655799, 0.17553662, 0.27368494,\n",
       "         0.3138588 , 0.22706732, 0.32469619, 0.28676974, 0.20334046,\n",
       "         0.23340104, 0.21074825, 0.19529472, 0.24710573, 0.20968837,\n",
       "         0.25945495, 0.22111458, 0.21789746, 0.19697935, 0.2234612 ],\n",
       "        [0.19746396, 0.1955523 , 0.21747665, 0.23944212, 0.21257516,\n",
       "         0.24692268, 0.34174849, 0.2185549 , 0.22984846, 0.24926452,\n",
       "         0.20843788, 0.31102266, 0.22253941, 0.19281012, 0.20204004,\n",
       "         0.22876899, 0.29117169, 0.26193006, 0.25431882, 0.23811402],\n",
       "        [0.25776226, 0.1967661 , 0.21340371, 0.27317207, 0.27746808,\n",
       "         0.19990949, 0.17977551, 0.26361342, 0.26776023, 0.26449693,\n",
       "         0.27959111, 0.31527806, 0.18408163, 0.23718622, 0.22829851,\n",
       "         0.24001466, 0.19677106, 0.1783755 , 0.34107842, 0.24119679],\n",
       "        [0.27238644, 0.1662219 , 0.21496441, 0.29552663, 0.2092681 ,\n",
       "         0.23013129, 0.24464381, 0.24405077, 0.28731117, 0.23326708,\n",
       "         0.19266126, 0.23565808, 0.22977775, 0.1856598 , 0.2303304 ,\n",
       "         0.27098437, 0.25143287, 0.18433928, 0.21337527, 0.25969507],\n",
       "        [0.22480543, 0.20989367, 0.19482871, 0.16850771, 0.34011536,\n",
       "         0.28761535, 0.18895937, 0.17621415, 0.25027396, 0.21677206,\n",
       "         0.19755388, 0.25336123, 0.21222468, 0.23994196, 0.25411326,\n",
       "         0.20775684, 0.21041247, 0.24737934, 0.2320274 , 0.23897462]],      dtype=float64), E=Array([[2.17551576, 2.47434137, 1.78033913, 1.92323565, 2.21714741,\n",
       "         2.30572055, 2.00390859, 2.48304382, 2.36615986, 1.0667829 ,\n",
       "         1.79497776, 1.6616753 , 2.06614876, 2.69784723, 1.89386792,\n",
       "         2.83021393, 1.8609862 , 1.38708324, 2.87648872, 2.47177108,\n",
       "         1.53895648, 1.76059946, 2.36210891, 3.4317128 , 3.26322636,\n",
       "         1.74896231, 2.6459978 , 2.01404884, 1.47918743, 2.30817392],\n",
       "        [2.49818358, 2.41833566, 1.35031126, 2.64618546, 1.83981877,\n",
       "         2.47949319, 1.78583679, 1.97042899, 2.27254602, 2.11175633,\n",
       "         2.33005789, 2.42484656, 1.85958681, 2.3980201 , 1.66851568,\n",
       "         3.47907561, 1.75707306, 1.39261   , 1.15466592, 1.23241136,\n",
       "         2.13925424, 3.29449088, 1.91492614, 2.34410349, 2.11351853,\n",
       "         2.65082878, 2.10561292, 1.35161412, 2.01545761, 1.68123913],\n",
       "        [1.31422413, 1.55139169, 1.50108645, 1.76245242, 1.7846289 ,\n",
       "         1.54227494, 1.97408688, 2.20254768, 1.50486441, 2.15400441,\n",
       "         1.81922795, 1.76491401, 2.04215511, 1.0754875 , 1.54913577,\n",
       "         2.20011667, 1.74886968, 1.17094551, 1.6570519 , 1.65146369,\n",
       "         2.38974423, 2.07282666, 1.58102851, 1.73282456, 2.0153637 ,\n",
       "         2.45973308, 1.88502291, 2.17364001, 1.91413845, 1.74387256],\n",
       "        [1.9766043 , 2.48490434, 1.48581926, 1.51244802, 2.3132474 ,\n",
       "         1.98309906, 1.78286688, 2.55282108, 1.93100003, 2.3099671 ,\n",
       "         2.20142064, 1.62941485, 1.7303302 , 2.18366378, 1.6648658 ,\n",
       "         2.45077674, 2.51680341, 2.19216963, 1.61863465, 2.06776636,\n",
       "         1.77315139, 1.74849339, 2.4168412 , 1.81037196, 1.84625309,\n",
       "         1.87678944, 1.95672415, 1.93104082, 1.93506508, 2.33181342],\n",
       "        [1.91610897, 2.33794007, 2.1825322 , 2.32145167, 2.17801737,\n",
       "         1.71582526, 2.17262005, 1.48332681, 2.15395699, 2.37687133,\n",
       "         2.23987018, 2.33552569, 2.64413534, 1.95492042, 2.79383161,\n",
       "         2.03404448, 1.82604401, 1.88472231, 2.73499855, 1.41212932,\n",
       "         2.64350885, 2.22432008, 2.17887638, 1.92899235, 1.88983411,\n",
       "         1.93459976, 2.30614076, 1.40702781, 1.59539447, 1.70325891]],      dtype=float64), c=Array([1.00000001, 1.00000001, 1.00000001, 1.00000001, 1.00000001],      dtype=float64)),\n",
       " OSQPState(x=Array([[-0.0352939 ,  0.06248869, -0.05517382,  0.04735412,  0.05421201,\n",
       "         -0.0415437 , -0.00686292,  0.07208576,  0.01872605, -0.04952747,\n",
       "         -0.04076229,  0.10233497, -0.06487156,  0.0027061 , -0.00974372,\n",
       "         -0.12404089,  0.103119  ,  0.08136788, -0.04971991,  0.15418859],\n",
       "        [ 0.06399716, -0.00728134,  0.05094905, -0.07003817,  0.13102202,\n",
       "          0.08736411,  0.1370147 ,  0.06521858, -0.16084254,  0.03815422,\n",
       "         -0.09157512,  0.07277431, -0.06482267,  0.06338497,  0.12930019,\n",
       "         -0.00645575,  0.20949747, -0.06616892, -0.13951698,  0.05090572],\n",
       "        [ 0.05659277, -0.02171091,  0.03651395, -0.10692127, -0.11660798,\n",
       "          0.07458688, -0.03439189, -0.14301678, -0.11509949, -0.00055135,\n",
       "          0.02809937,  0.01218996, -0.00591331,  0.05511936,  0.11420147,\n",
       "         -0.11087455,  0.07403721,  0.02600094,  0.12830472,  0.12381121],\n",
       "        [ 0.01637336,  0.10659779,  0.05345199, -0.24861138, -0.00458567,\n",
       "          0.09992116, -0.04245465,  0.21283999,  0.10495371, -0.1511072 ,\n",
       "         -0.21890699, -0.07358785,  0.03944359,  0.11342976,  0.10661966,\n",
       "         -0.09526848, -0.0784894 , -0.03169591, -0.20183778, -0.04642128],\n",
       "        [ 0.18003903,  0.04157011, -0.25408468,  0.06813072,  0.11652027,\n",
       "         -0.16985611,  0.20996455, -0.05427246, -0.18347346, -0.01327981,\n",
       "          0.16466729,  0.13363893,  0.04532168, -0.14456826,  0.2718973 ,\n",
       "         -0.24340735, -0.07697091,  0.08086271,  0.06435071,  0.03003517]],      dtype=float64), z=Array([[ 0.64262521,  0.48745753, -0.10634527, -0.18090033,  0.51361399,\n",
       "          0.12144921, -0.40447063, -0.02977096,  0.53839502, -0.16123679,\n",
       "          0.49779985,  0.13298601,  0.34935175,  0.68334632,  0.52933391,\n",
       "         -0.63896373,  0.036233  , -0.25945364, -0.8365716 , -0.80079461,\n",
       "          0.63900979,  0.11593762, -0.09848964,  1.07972983, -0.83557478,\n",
       "          0.08898989,  0.34085389, -0.21094411, -0.07727236, -0.66848725],\n",
       "        [-0.43655112, -0.32387326,  0.39667184,  0.2897702 , -0.28032049,\n",
       "          0.18403407,  0.84844907, -0.22295341, -0.12351944, -0.38979558,\n",
       "         -0.53739283, -0.14947889, -0.50271039, -1.11076211, -0.64381039,\n",
       "          0.45521862, -0.04909177, -0.39425069,  0.66063714, -0.27819587,\n",
       "         -0.2028658 ,  0.30297416, -0.2787185 ,  1.07773556,  1.2710991 ,\n",
       "          0.74843927,  0.56965442, -0.85086813,  0.0171055 ,  0.39046971],\n",
       "        [ 0.4908538 , -0.89855795, -0.09734754,  1.06690216, -0.40864066,\n",
       "          0.14726401, -0.66908783,  0.6758799 , -0.61468232,  0.15279953,\n",
       "          0.35971149,  0.44609039, -1.47894985, -0.12366476,  0.65216742,\n",
       "         -0.64838707, -0.13308272, -0.03950224, -0.85414634, -0.059376  ,\n",
       "          0.12409871, -0.91600154, -0.22399203, -1.08943025,  1.44623243,\n",
       "          0.3990591 , -0.67442265,  0.29177674,  0.66790031,  1.32832157],\n",
       "        [-0.45870951,  2.04289748, -0.19457717,  0.98391954, -0.64435167,\n",
       "         -0.85358922, -0.05485166, -0.59877155,  0.23040089,  0.44161484,\n",
       "         -0.35800185,  0.44605924, -1.01708896,  0.23862327,  0.35671655,\n",
       "         -0.79010056, -0.83740823,  0.21363393,  0.02623413,  1.58148897,\n",
       "          0.30640741, -0.58385625,  0.51347531,  1.21572021,  0.82250189,\n",
       "          0.52861993,  1.29942983, -0.81467876, -0.79817883,  0.15467769],\n",
       "        [-1.62669243, -0.51754497, -0.3886388 , -0.76768904,  0.52603867,\n",
       "         -0.35141433, -0.08163586,  0.17748434,  0.35146641, -0.40977285,\n",
       "          0.01861355,  0.09199639,  0.88610078, -0.04937464, -1.22086443,\n",
       "         -1.45711591, -0.63548991, -0.52551143, -0.24119839, -0.40723935,\n",
       "         -0.78969564, -1.5221806 ,  0.36265466, -0.82157053,  0.58843392,\n",
       "         -0.91563771, -0.42552188, -1.24317653, -0.34489029,  0.15623586]],      dtype=float64), chi=Array([[-0.14139071,  0.30944795, -0.28069989,  0.26977034,  0.19807661,\n",
       "         -0.13236908, -0.03022791,  0.22200979,  0.06529717, -0.2435712 ,\n",
       "         -0.17465404,  0.48557631, -0.33217718,  0.0109444 , -0.04647362,\n",
       "         -0.47808785,  0.46636477,  0.37342442, -0.25240643,  0.69000599],\n",
       "        [ 0.32419066, -0.03730226,  0.23432866, -0.2926833 ,  0.61674198,\n",
       "          0.35375028,  0.40102493,  0.2983742 , -0.6998767 ,  0.15317118,\n",
       "         -0.43930219,  0.2341391 , -0.29147044,  0.32882359,  0.64011989,\n",
       "         -0.02825331,  0.71961351, -0.25275178, -0.54869483,  0.21392565],\n",
       "        [ 0.21956168, -0.11034046,  0.17112513, -0.3914239 , -0.42025939,\n",
       "          0.3730971 , -0.19129936, -0.54254363, -0.42986865, -0.00209348,\n",
       "          0.10049811,  0.03866151, -0.03211295,  0.2324348 ,  0.50024904,\n",
       "         -0.46193916,  0.37624723,  0.14576496,  0.37617645,  0.51334524],\n",
       "        [ 0.06027036,  0.64141185,  0.24863619, -0.84133293, -0.02197062,\n",
       "          0.43406092, -0.17363056,  0.87227957,  0.36532359, -0.64802174,\n",
       "         -1.13623846, -0.31225324,  0.17161278,  0.6109875 ,  0.46285235,\n",
       "         -0.35148932, -0.31219755, -0.17195323, -0.94586739, -0.1788714 ],\n",
       "        [ 0.80086161,  0.19798859, -1.30416153,  0.40423629,  0.34254748,\n",
       "         -0.59050462,  1.11120208, -0.30788427, -0.73318669, -0.06125091,\n",
       "          0.83358627,  0.52745654,  0.21356785, -0.60251421,  1.06995588,\n",
       "         -1.171634  , -0.36578755,  0.32691378,  0.277436  ,  0.12559205]],      dtype=float64), zeta=Array([[ 0.64262987,  0.48745328, -0.10624622, -0.18090397,  0.51360447,\n",
       "          0.12145679, -0.40447167, -0.02974316,  0.53840725, -0.16128866,\n",
       "          0.49780878,  0.13310524,  0.3493473 ,  0.68334751,  0.52933297,\n",
       "         -0.63898651,  0.03623941, -0.25945576, -0.83657415, -0.80072172,\n",
       "          0.63901799,  0.11593575, -0.09858414,  1.0797278 , -0.83558196,\n",
       "          0.0890538 ,  0.34094275, -0.21095095, -0.07712643, -0.66849304],\n",
       "        [-0.43633662, -0.32315959,  0.39674747,  0.28980683, -0.27943848,\n",
       "          0.18486955,  0.84956491, -0.22328675, -0.12343297, -0.38972809,\n",
       "         -0.53784762, -0.14961795, -0.50313466, -1.11109742, -0.64401132,\n",
       "          0.45528627, -0.0488271 , -0.39409327,  0.66089178, -0.27806063,\n",
       "         -0.20296796,  0.30248554, -0.27888066,  1.07818906,  1.27144531,\n",
       "          0.748666  ,  0.57005173, -0.85110213,  0.01678839,  0.39064136],\n",
       "        [ 0.49084084, -0.89860118, -0.09770346,  1.06688216, -0.40849667,\n",
       "          0.14701932, -0.66912119,  0.67587753, -0.61473815,  0.15282064,\n",
       "          0.35966848,  0.44622936, -1.4789826 , -0.1251169 ,  0.65128667,\n",
       "         -0.64836341, -0.1331073 , -0.03945635, -0.85426567, -0.05936416,\n",
       "          0.12413924, -0.91601557, -0.22437633, -1.08946527,  1.44622399,\n",
       "          0.39961278, -0.67447349,  0.29174339,  0.66795246,  1.32845073],\n",
       "        [-0.45860664,  2.04344284, -0.19522768,  0.98392659, -0.64487993,\n",
       "         -0.85350578, -0.05476695, -0.59881518,  0.23005279,  0.4419351 ,\n",
       "         -0.35839017,  0.44611837, -1.01639005,  0.23883786,  0.35687553,\n",
       "         -0.79054566, -0.83715382,  0.21351474,  0.02614044,  1.58190946,\n",
       "          0.30671892, -0.5838242 ,  0.51379034,  1.21563238,  0.82274516,\n",
       "          0.52869463,  1.29953979, -0.81494073, -0.79804033,  0.15512946],\n",
       "        [-1.62702082, -0.51758829, -0.38878583, -0.76765394,  0.52626391,\n",
       "         -0.35142598, -0.08175389,  0.17758239,  0.35147261, -0.40994249,\n",
       "          0.01840329,  0.09194495,  0.88618362, -0.04945636, -1.22088349,\n",
       "         -1.45723659, -0.63525739, -0.52573788, -0.24064756, -0.40732378,\n",
       "         -0.78974813, -1.52214398,  0.36274441, -0.82148937,  0.58852888,\n",
       "         -0.91519172, -0.42589808, -1.24280472, -0.344981  ,  0.1564046 ]],      dtype=float64), y=Array([[ 0.00000000e+00,  5.15075762e-18, -2.07936105e-01,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         -3.66677040e-01,  0.00000000e+00,  7.89396494e-01,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  4.02757926e-02,  0.00000000e+00,\n",
       "          0.00000000e+00, -2.62461961e-01,  0.00000000e+00,\n",
       "          0.00000000e+00, -3.60404400e-01,  0.00000000e+00,\n",
       "         -3.15321720e-02,  1.83775095e-01,  2.46084376e-01,\n",
       "          0.00000000e+00, -1.99926329e-01,  0.00000000e+00],\n",
       "        [ 0.00000000e+00, -1.37057463e-01,  8.78406234e-19,\n",
       "          0.00000000e+00, -2.48224097e-01,  5.84978181e-01,\n",
       "          2.40769276e-01,  0.00000000e+00,  0.00000000e+00,\n",
       "          1.46532530e-18,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00, -8.50917460e-03,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          2.02597500e-01,  0.00000000e+00, -1.85550726e-18,\n",
       "          1.01390989e-01,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  4.59845487e-18,  0.00000000e+00,\n",
       "          0.00000000e+00,  6.81846416e-02,  4.35726500e-01],\n",
       "        [ 2.73577852e-18,  0.00000000e+00, -5.90169056e-01,\n",
       "          0.00000000e+00, -1.54304260e-01,  4.74673828e-01,\n",
       "          4.28061852e-18,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  3.99528162e-02,\n",
       "          0.00000000e+00, -1.95890158e-01,  8.11448879e-03,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         -2.48740394e-02,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00, -3.04968536e-01,  0.00000000e+00,\n",
       "          1.32388792e-01,  3.34421050e-01,  0.00000000e+00,\n",
       "          2.71368966e-01,  0.00000000e+00,  4.00576953e-01],\n",
       "        [ 0.00000000e+00,  0.00000000e+00, -6.53045414e-01,\n",
       "         -6.55919764e-19, -1.40894153e-01,  0.00000000e+00,\n",
       "          0.00000000e+00,  8.85687723e-18,  1.34567480e-01,\n",
       "          4.43918945e-01,  3.81885602e-18,  0.00000000e+00,\n",
       "         -3.11811024e-01,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  8.73191584e-18,  0.00000000e+00,\n",
       "          1.58344761e-02,  5.98648978e-01,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  5.83978458e-02,\n",
       "          7.30409536e-03,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  5.40075782e-01],\n",
       "        [-4.50424265e-02,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  3.64022213e-02,\n",
       "         -7.62036837e-01,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00, -4.15275763e-01,  0.00000000e+00,\n",
       "          7.05700936e-18,  0.00000000e+00,  0.00000000e+00,\n",
       "         -5.05370439e-01,  0.00000000e+00,  0.00000000e+00,\n",
       "          1.92929011e-17,  0.00000000e+00,  8.03104388e-18,\n",
       "          0.00000000e+00, -1.65249033e-01, -2.04709730e-01,\n",
       "         -6.58069473e-01,  0.00000000e+00,  0.00000000e+00]],      dtype=float64), dx=Array([[-1.86649131e-06, -1.31206944e-05, -1.27259824e-07,\n",
       "         -6.87367660e-06,  1.39039110e-05,  1.27972863e-05,\n",
       "          9.97203662e-06,  2.47605581e-07,  7.43680933e-06,\n",
       "          5.37752418e-06,  2.44426274e-05,  7.73670507e-06,\n",
       "          1.22076528e-05,  1.80480571e-05,  1.60472578e-05,\n",
       "          1.40772984e-05, -1.26756529e-05, -4.34633051e-06,\n",
       "         -1.42619351e-05, -1.16148294e-05],\n",
       "        [-2.54066731e-04,  1.80000872e-04, -1.46651199e-04,\n",
       "          4.73743607e-04, -1.02875205e-03,  1.63536505e-04,\n",
       "         -2.72816699e-04,  9.06891832e-05,  2.67355332e-04,\n",
       "         -2.77327298e-04, -1.01154656e-04, -4.13704441e-04,\n",
       "          4.91168645e-04, -2.15046092e-04, -3.91473030e-04,\n",
       "          9.01293103e-05, -3.07801232e-04,  3.49915684e-04,\n",
       "          2.77241067e-04, -3.69282090e-04],\n",
       "        [-2.01062055e-05,  4.78431027e-06, -5.98820870e-05,\n",
       "          4.69378830e-05,  5.59693584e-06,  1.64571775e-05,\n",
       "         -1.40500565e-05,  5.05608817e-05,  2.22882430e-05,\n",
       "          2.38362720e-05,  9.43487865e-06,  7.03504042e-06,\n",
       "         -2.75619161e-05, -1.23367740e-04, -5.42443697e-05,\n",
       "         -2.64787366e-05,  3.57845037e-05,  6.37194458e-07,\n",
       "         -7.44857292e-06, -6.64354231e-05],\n",
       "        [-4.25549649e-04, -3.03233790e-04,  5.03696874e-05,\n",
       "          2.24706312e-04,  1.53975629e-04,  3.49440162e-04,\n",
       "          2.50572227e-04, -4.42623720e-04, -7.27073537e-05,\n",
       "          6.27986898e-04,  2.97257231e-05, -3.22124112e-05,\n",
       "          1.25199871e-04, -8.69733365e-05,  1.23773173e-04,\n",
       "         -2.00472708e-04,  7.76932097e-05,  2.64255223e-05,\n",
       "         -1.63324093e-04,  3.15667561e-04],\n",
       "        [ 1.17819665e-05,  1.72226431e-04,  4.69519161e-05,\n",
       "          2.18183187e-04,  1.14603850e-04, -1.66125259e-04,\n",
       "         -1.05922999e-04, -2.86003194e-04,  2.56492282e-04,\n",
       "         -2.86079095e-05, -1.47309362e-04,  1.99052218e-05,\n",
       "         -3.37368998e-05,  2.02141955e-06,  7.68188419e-05,\n",
       "          9.79128290e-05, -5.89587254e-05, -9.70589818e-05,\n",
       "         -2.53383771e-04,  2.43916208e-04]], dtype=float64), dy=Array([[ 0.00000000e+00,  0.00000000e+00,  1.58480821e-05,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         -8.29902280e-06,  0.00000000e+00,  1.90764929e-05,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  1.02569082e-06,  0.00000000e+00,\n",
       "          0.00000000e+00,  1.16612822e-05,  0.00000000e+00,\n",
       "          0.00000000e+00, -1.51197570e-05,  0.00000000e+00,\n",
       "         -1.14883800e-06,  1.02257834e-05,  1.42163430e-05,\n",
       "          0.00000000e+00,  2.33489743e-05,  0.00000000e+00],\n",
       "        [ 0.00000000e+00,  1.14187259e-04,  0.00000000e+00,\n",
       "          0.00000000e+00,  1.41121170e-04,  1.33677039e-04,\n",
       "          1.78533716e-04,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00, -5.36490559e-05,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          4.07423737e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "         -7.81793814e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00, -5.07375358e-05,  2.74647148e-05],\n",
       "        [ 0.00000000e+00,  0.00000000e+00, -5.69466304e-05,\n",
       "          0.00000000e+00,  2.30385189e-05, -3.91503958e-05,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  2.22354941e-05,\n",
       "          0.00000000e+00, -2.32341241e-04, -1.40920537e-04,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         -1.90924523e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00, -6.14882601e-05,  0.00000000e+00,\n",
       "         -1.35103353e-06,  8.85887670e-05,  0.00000000e+00,\n",
       "         -5.33630871e-06,  0.00000000e+00,  2.06657618e-05],\n",
       "        [ 0.00000000e+00,  0.00000000e+00, -1.04080352e-04,\n",
       "          0.00000000e+00, -8.45212385e-05,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00, -5.56964248e-05,\n",
       "          5.12419791e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "          1.11825137e-04,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "         -1.49900195e-05,  6.72794141e-05,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00, -1.40529664e-05,\n",
       "          3.89231777e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  7.22824886e-05],\n",
       "        [-5.25418120e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  9.92515446e-07,\n",
       "         -2.71418905e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00, -1.30748747e-05,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          8.81328336e-05,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "          0.00000000e+00,  7.13591935e-05, -6.01918177e-05,\n",
       "          5.94885446e-05,  0.00000000e+00,  0.00000000e+00]],      dtype=float64), converged=Array([ True,  True,  True,  True,  True], dtype=bool), primal_infeas=Array([False, False, False, False, False], dtype=bool), dual_infeas=Array([False, False, False, False, False], dtype=bool), kkt_mat=Array([[[  1.000001  ,   0.04953851,   0.14390485, ...,  -0.38977781,\n",
       "            0.24418747,   0.42420285],\n",
       "         [  0.04953851,   1.000001  ,  -0.09837497, ...,  -0.8482167 ,\n",
       "           -0.09523206,  -0.32291805],\n",
       "         [  0.14390485,  -0.09837497,   1.000001  , ...,   0.690905  ,\n",
       "           -0.39143038,  -0.47942492],\n",
       "         ...,\n",
       "         [ -0.38977781,  -0.8482167 ,   0.690905  , ..., -10.        ,\n",
       "           -0.        ,  -0.        ],\n",
       "         [  0.24418747,  -0.09523206,  -0.39143038, ...,  -0.        ,\n",
       "          -10.        ,  -0.        ],\n",
       "         [  0.42420285,  -0.32291805,  -0.47942492, ...,  -0.        ,\n",
       "           -0.        , -10.        ]],\n",
       " \n",
       "        [[  1.000001  ,  -0.25564654,   0.13749279, ...,  -0.3954728 ,\n",
       "           -0.33064348,  -0.32389417],\n",
       "         [ -0.25564654,   1.000001  ,  -0.20499922, ...,  -0.0767938 ,\n",
       "            0.03668236,   0.10143784],\n",
       "         [  0.13749279,  -0.20499922,   1.000001  , ...,   0.12965282,\n",
       "           -0.52720037,   0.6971758 ],\n",
       "         ...,\n",
       "         [ -0.3954728 ,  -0.0767938 ,   0.12965282, ..., -10.        ,\n",
       "           -0.        ,  -0.        ],\n",
       "         [ -0.33064348,   0.03668236,  -0.52720037, ...,  -0.        ,\n",
       "          -10.        ,  -0.        ],\n",
       "         [ -0.32389417,   0.10143784,   0.6971758 , ...,  -0.        ,\n",
       "           -0.        , -10.        ]],\n",
       " \n",
       "        [[  1.000001  ,   0.31734322,  -0.29633194, ...,   0.2307604 ,\n",
       "            0.04631246,   0.        ],\n",
       "         [  0.31734322,   1.000001  ,   0.33657715, ...,   0.54909402,\n",
       "            0.47905888,   0.23390496],\n",
       "         [ -0.29633194,   0.33657715,   1.000001  , ...,   0.02453972,\n",
       "           -0.35727312,  -0.34761476],\n",
       "         ...,\n",
       "         [  0.2307604 ,   0.54909402,   0.02453972, ..., -10.        ,\n",
       "           -0.        ,  -0.        ],\n",
       "         [  0.04631246,   0.47905888,  -0.35727312, ...,  -0.        ,\n",
       "          -10.        ,  -0.        ],\n",
       "         [  0.        ,   0.23390496,  -0.34761476, ...,  -0.        ,\n",
       "           -0.        , -10.        ]],\n",
       " \n",
       "        [[  1.000001  ,   0.12596388,  -0.1163087 , ...,   0.87192295,\n",
       "            0.80328214,   0.32288593],\n",
       "         [  0.12596388,   1.000001  ,  -0.04117697, ...,   0.07378871,\n",
       "            0.54874027,  -0.0708922 ],\n",
       "         [ -0.1163087 ,  -0.04117697,   1.000001  , ...,   0.30076725,\n",
       "           -0.04090665,   0.        ],\n",
       "         ...,\n",
       "         [  0.87192295,   0.07378871,   0.30076725, ..., -10.        ,\n",
       "           -0.        ,  -0.        ],\n",
       "         [  0.80328214,   0.54874027,  -0.04090665, ...,  -0.        ,\n",
       "          -10.        ,  -0.        ],\n",
       "         [  0.32288593,  -0.0708922 ,   0.        , ...,  -0.        ,\n",
       "           -0.        , -10.        ]],\n",
       " \n",
       "        [[  1.000001  ,   0.12563016,  -0.01844979, ...,  -0.05284822,\n",
       "           -0.20785859,   0.60975142],\n",
       "         [  0.12563016,   1.000001  ,  -0.23084397, ...,   0.15324893,\n",
       "            0.11292012,   0.10821513],\n",
       "         [ -0.01844979,  -0.23084397,   1.000001  , ...,  -0.33468658,\n",
       "            0.49115918,   0.        ],\n",
       "         ...,\n",
       "         [ -0.05284822,   0.15324893,  -0.33468658, ..., -10.        ,\n",
       "           -0.        ,  -0.        ],\n",
       "         [ -0.20785859,   0.11292012,   0.49115918, ...,  -0.        ,\n",
       "          -10.        ,  -0.        ],\n",
       "         [  0.60975142,   0.10821513,   0.        , ...,  -0.        ,\n",
       "           -0.        , -10.        ]]], dtype=float64), kkt_lu=(Array([[[ 1.00000100e+00,  4.95385077e-02,  1.43904850e-01, ...,\n",
       "          -3.89777807e-01,  2.44187474e-01,  4.24202845e-01],\n",
       "         [ 4.95384582e-02,  9.97546939e-01, -1.05503792e-01, ...,\n",
       "          -8.28907707e-01, -1.07328734e-01, -3.43932403e-01],\n",
       "         [ 1.43904706e-01, -1.05763236e-01,  9.68133992e-01, ...,\n",
       "           6.59327900e-01, -4.37921536e-01, -5.76845114e-01],\n",
       "         ...,\n",
       "         [ 3.67248741e-01, -3.24666589e-01,  1.00206764e-01, ...,\n",
       "           1.66557315e+01, -6.50231168e+00, -2.75100578e+00],\n",
       "         [ 2.56535814e-01, -1.44848615e-01,  1.78799100e-01, ...,\n",
       "           4.94243746e-01,  1.69820511e+01,  6.86391937e+00],\n",
       "         [ 1.40823856e-01,  2.97827070e-01,  2.47219736e-01, ...,\n",
       "           6.61377451e-01, -1.25753066e-02,  8.84908303e+00]],\n",
       " \n",
       "        [[ 1.00000100e+00, -2.55646542e-01,  1.37492788e-01, ...,\n",
       "          -3.95472799e-01, -3.30643483e-01, -3.23894175e-01],\n",
       "         [-2.55646287e-01,  9.34645911e-01, -1.69849699e-01, ...,\n",
       "          -1.77894955e-01, -4.78454221e-02,  1.86354927e-02],\n",
       "         [-6.09148239e-01, -2.23140862e-01,  1.04349862e+00, ...,\n",
       "          -2.80597193e-01, -2.12087164e-01, -1.93141227e-01],\n",
       "         ...,\n",
       "         [ 2.12787473e-02,  1.03697877e-02,  2.37591672e-01, ...,\n",
       "          -1.35459078e+01,  2.04411524e-01, -7.03550704e+00],\n",
       "         [-1.25249076e-01,  1.19467553e-01,  3.79070337e-02, ...,\n",
       "           7.03772437e-01, -7.19904277e+00,  8.74746455e+00],\n",
       "         [ 2.40206270e-01,  2.68273112e-02,  1.07256645e-01, ...,\n",
       "          -4.16450048e-01,  6.16404252e-01, -1.25842889e+01]],\n",
       " \n",
       "        [[ 1.00000100e+00,  3.17343218e-01, -2.96331940e-01, ...,\n",
       "           2.30760397e-01,  4.63124590e-02,  0.00000000e+00],\n",
       "         [ 3.17342901e-01,  8.99294382e-01,  4.30615986e-01, ...,\n",
       "           4.75863844e-01,  4.64361953e-01,  2.33904960e-01],\n",
       "         [-5.91453015e-02,  2.87503074e-01, -8.44373466e-01, ...,\n",
       "          -1.23163924e-01, -1.30766324e-01, -6.72483950e-02],\n",
       "         ...,\n",
       "         [ 4.50576179e-01, -2.01132719e-01, -2.93397896e-01, ...,\n",
       "           1.61751439e+01, -2.84830665e+00, -5.43699257e+00],\n",
       "         [-3.53510416e-01, -1.37632339e-02,  4.44926932e-01, ...,\n",
       "          -2.58864020e-01, -1.47550099e+01,  2.09425802e+00],\n",
       "         [-1.64932789e-01,  7.23125229e-02,  2.85759340e-01, ...,\n",
       "          -5.74259809e-01,  3.72851169e-01, -5.31324204e+00]],\n",
       " \n",
       "        [[ 1.00000100e+00,  1.25963877e-01, -1.16308698e-01, ...,\n",
       "           8.71922947e-01,  8.03282141e-01,  3.22885928e-01],\n",
       "         [ 9.97334365e-01, -1.12088230e+00,  2.49320990e-01, ...,\n",
       "          -8.69598719e-01, -8.01140885e-01, -3.22025232e-01],\n",
       "         [-1.16308582e-01,  2.36655491e-02,  9.80572982e-01, ...,\n",
       "           4.22758901e-01,  7.14813910e-02,  4.51753084e-02],\n",
       "         ...,\n",
       "         [-4.22616070e-01,  2.03758954e-01, -4.75075629e-01, ...,\n",
       "           2.23758687e+01,  1.39376407e+00, -9.58459069e-01],\n",
       "         [ 8.03281338e-01, -3.99288877e-01,  1.55085907e-01, ...,\n",
       "          -5.60337738e-02, -1.35558231e+01,  6.34833356e+00],\n",
       "         [ 5.95628240e-02,  4.20082532e-02, -3.69042811e-01, ...,\n",
       "          -2.92832346e-01,  2.75691777e-02, -1.04723462e+01]],\n",
       " \n",
       "        [[ 1.00000100e+00,  1.25630163e-01, -1.84497899e-02, ...,\n",
       "          -5.28482213e-02, -2.07858588e-01,  6.09751418e-01],\n",
       "         [ 0.00000000e+00,  9.97821024e-01, -6.69654857e-01, ...,\n",
       "           0.00000000e+00,  0.00000000e+00, -0.00000000e+00],\n",
       "         [ 3.36502324e-01,  2.10776197e-01,  1.14511427e+00, ...,\n",
       "           1.77835493e-02,  6.99448980e-02, -2.05182769e-01],\n",
       "         ...,\n",
       "         [ 1.97440057e-01, -1.31085777e-01, -8.87989400e-02, ...,\n",
       "          -6.93473183e+00,  1.09730558e-01, -5.26817721e+00],\n",
       "         [-2.07858380e-01,  1.39337010e-01,  5.07051533e-01, ...,\n",
       "           7.90960609e-01, -1.18068204e+01,  4.55464113e+00],\n",
       "         [ 1.29925090e-01, -1.63518450e-01, -9.99851672e-02, ...,\n",
       "          -7.75966535e-01, -1.06365465e-02, -1.38913496e+01]]],      dtype=float64), Array([[ 0,  1,  2, 47, 38, 27, 43, 20, 22, 25, 49, 33, 37, 29, 44, 45,\n",
       "         42, 34, 34, 35, 44, 21, 42, 23, 24, 40, 26, 33, 28, 29, 30, 31,\n",
       "         32, 47, 41, 39, 36, 38, 40, 41, 47, 48, 43, 45, 47, 45, 46, 48,\n",
       "         48, 49],\n",
       "        [ 0,  1, 31, 44,  4, 25, 28,  7, 48, 29, 21, 27, 43, 37, 22, 30,\n",
       "         36, 35, 20, 33, 31, 39, 49, 23, 24, 45, 26, 40, 29, 41, 33, 42,\n",
       "         32, 45, 34, 39, 37, 44, 38, 40, 45, 49, 45, 46, 47, 47, 47, 49,\n",
       "         49, 49],\n",
       "        [ 0,  1, 35, 44, 49, 41, 23, 48, 26,  9, 39, 25, 29, 31, 45, 40,\n",
       "         46, 36, 27, 47, 20, 21, 22, 31, 24, 25, 43, 42, 28, 38, 30, 47,\n",
       "         32, 33, 34, 41, 36, 37, 48, 48, 40, 48, 45, 49, 45, 49, 48, 49,\n",
       "         48, 49],\n",
       "        [ 0, 27,  2, 33,  4, 44, 21, 35, 45, 24, 38, 29, 34, 13, 20, 43,\n",
       "         28, 30, 39, 36, 25, 40, 22, 23, 46, 41, 26, 27, 32, 37, 37, 31,\n",
       "         45, 42, 49, 43, 37, 40, 45, 47, 43, 49, 45, 46, 44, 45, 47, 47,\n",
       "         48, 49],\n",
       "        [ 0, 33, 45, 20,  4, 40,  6, 31, 42, 41, 36, 43, 12, 26, 37, 39,\n",
       "         28, 32, 21, 47, 27, 46, 22, 23, 24, 25, 44, 40, 28, 29, 30, 46,\n",
       "         39, 33, 34, 35, 39, 45, 38, 49, 46, 49, 46, 49, 47, 47, 49, 47,\n",
       "         48, 49]], dtype=int32)), rho=Array([[0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1],\n",
       "        [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1],\n",
       "        [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1],\n",
       "        [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1],\n",
       "        [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1,\n",
       "         0.1, 0.1, 0.1, 0.1]], dtype=float64, weak_type=True)))"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sol"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "id": "e732f3e4-d36d-427b-869f-89e93e0cf157",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array(True, dtype=bool)"
      ]
     },
     "execution_count": 311,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "jnp.all(sol[-1].converged)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "jaxosqp",
   "language": "python",
   "name": "jaxosqp"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}