04_RNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random\n",
    "\n",
    "from sklearn.metrics import roc_auc_score, precision_score, accuracy_score, recall_score, f1_score\n",
    "\n",
    "import torch\n",
    "from torch import nn, optim\n",
    "import torch.nn.functional as F\n",
    "\n",
    "from vocabulary import Vocabulary\n",
    "from dataset import Dataset, collate_fn\n",
    "from models import Classifier\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['font.sans-serif'] = \"Menlo\"\n",
    "plt.rcParams['font.family'] = \"sans-serif\"\n",
    "plt.rcParams.update({'font.size': 22})\n",
    "\n",
    "from tqdm import tqdm\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Import data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "activity = pd.read_csv('clean/activity_clean.csv')\n",
    "hemolysis = pd.read_csv('clean/hemolysis_clean.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get vocabulary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "vocabulary = Vocabulary.get_vocabulary_from_sequences(activity['sequence'].values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Define RNN functions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Helper functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def randomChoice(l):\n",
    "    return l[random.randint(0, len(l) - 1)]\n",
    "\n",
    "def categoryFromOutput(output):\n",
    "    top_n, top_i = output.topk(1)\n",
    "    category_i = top_i[0].item()\n",
    "    return category_i\n",
    "\n",
    "def nan_equal(a,b):\n",
    "    try:\n",
    "        np.testing.assert_equal(a,b)\n",
    "    except AssertionError:\n",
    "        return False\n",
    "    return True\n",
    "\n",
    "def models_are_equal(model1, model2):\n",
    "    model1.vocabulary == model2.vocabulary\n",
    "    model1.hidden_size == model2.hidden_size\n",
    "    for a,b in zip(model1.model.parameters(), model2.model.parameters()):\n",
    "        if nan_equal(a.detach().numpy(), b.detach().numpy()) == True:\n",
    "            print(\"true\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train RNNs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define training functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def training(model, test_dataloader, training_dataloader, n_epoch, optimizer, filename, logname):\n",
    "    \n",
    "    roc_training = []\n",
    "    roc_test = []\n",
    "    \n",
    "    for e in tqdm(range(1, n_epoch + 1)):\n",
    "\n",
    "        for i_batch, sample_batched in enumerate(training_dataloader):\n",
    "            seq_batched = sample_batched[0][0].to(model.device, non_blocking=True)\n",
    "            seq_lengths = sample_batched[0][1].to(model.device, non_blocking=True)\n",
    "            cat_batched = sample_batched[1].to(model.device, non_blocking=True)\n",
    "\n",
    "            output = model.evaluate(seq_batched, seq_lengths)\n",
    "            \n",
    "            log_probs = F.log_softmax(output, dim=1)\n",
    "            loss = nn.NLLLoss()(log_probs, cat_batched)\n",
    "\n",
    "            optimizer.zero_grad()\n",
    "            loss.backward()  \n",
    "            torch.nn.utils.clip_grad_value_(model.model.parameters(), 2)\n",
    "            optimizer.step()\n",
    "\n",
    "        model.save(filename.format(e))\n",
    "        \n",
    "        def _evaluate_ROC(data_loader):\n",
    "            cat_list = []\n",
    "            out_list = []\n",
    "\n",
    "            with torch.no_grad():\n",
    "                for i_batch, sample_batched in enumerate(data_loader):    \n",
    "                    seq_batched = sample_batched[0][0].to(model.device, non_blocking=True)\n",
    "                    seq_lengths = sample_batched[0][1].to(model.device, non_blocking=True)\n",
    "                    \n",
    "                    cat_list += sample_batched[1].to(\"cpu\", non_blocking=True)\n",
    "                    out_list += torch.exp(model.evaluate(seq_batched, seq_lengths))[: ,1].to(\"cpu\", non_blocking=True)\n",
    "\n",
    "                cat_list = torch.stack(cat_list)\n",
    "                out_list = torch.stack(out_list)\n",
    "\n",
    "                roc = roc_auc_score(cat_list.cpu().numpy().astype(int), out_list.cpu().numpy())\n",
    "            return roc\n",
    "        \n",
    "        roc_tr = _evaluate_ROC(training_dataloader)\n",
    "        roc_te = _evaluate_ROC(test_dataloader)\n",
    "        roc_training.append(roc_tr)\n",
    "        roc_test.append(roc_te)\n",
    "\n",
    "        with open(logname, 'a') as f:\n",
    "            f.write('{},{}\\n'.format(roc_tr, roc_te))\n",
    "\n",
    "        if roc_tr == 1.0:\n",
    "            break\n",
    "        \n",
    "    return model, optimizer, roc_training, roc_test\n",
    "\n",
    "def train_splits(df, purpose, n_embedding, n_hidden, n_layers, learning_rate, momentum, batch_size, n_epoch):\n",
    "\n",
    "    for i in range(1, 6):\n",
    "\n",
    "        # Create model directory\n",
    "        if not os.path.exists('models/RNN_{}/split_{}'.format(purpose, i)):\n",
    "            os.makedirs('models/RNN_{}/split_{}'.format(purpose, i))\n",
    "        \n",
    "        # Write log file \n",
    "        with open('models/RNN_{}/split_{}/log.txt'.format(purpose, i), 'w') as f:\n",
    "            f.write('n_embedding: {}\\n'.format(n_embedding))\n",
    "            f.write('n_hidden: {}\\n'.format(n_hidden))\n",
    "            f.write('n_layers: {}\\n'.format(n_layers))\n",
    "            f.write('learning_rate: {}\\n'.format(learning_rate))\n",
    "            f.write('momentum: {}\\n'.format(momentum))\n",
    "            f.write('batch_size: {}\\n'.format(batch_size))\n",
    "            f.write('n_epoch: {}\\n'.format(n_epoch))\n",
    "            f.write('\\n')\n",
    "            f.write('Training ROC-AUC,Test ROC-AUC\\n')\n",
    "\n",
    "        # Initialize data splits\n",
    "        train = Dataset(df[df[f'split_{i}'] == 'training'], vocabulary)\n",
    "        test = Dataset(df[df[f'split_{i}'] == 'test'], vocabulary)\n",
    "\n",
    "        # Initialize the model\n",
    "        model = Classifier(n_embedding, n_hidden, n_layers, vocabulary)\n",
    "        model.to(device)\n",
    "\n",
    "        # Initialize the optimizer\n",
    "        optimizer = optim.SGD(model.model.parameters(), lr = learning_rate, momentum=momentum)\n",
    "\n",
    "        # Load the data\n",
    "        training_dataloader = torch.utils.data.DataLoader(train, batch_size=batch_size, shuffle=True, collate_fn = collate_fn, drop_last=True, pin_memory=True, num_workers=8)\n",
    "        test_dataloader = torch.utils.data.DataLoader(test, batch_size=batch_size, shuffle=True, collate_fn = collate_fn, drop_last=True, pin_memory=True, num_workers=8)\n",
    "\n",
    "        # Define model name\n",
    "        filename = 'models/RNN_{}/split_{}/em{}_hi{}_la{}_ep{{}}'.format(purpose, i, n_embedding, n_hidden, n_layers)\n",
    "        logname = 'models/RNN_{}/split_{}/log.txt'.format(purpose, i)\n",
    "\n",
    "        # Train the model\n",
    "        model, optimizer, roc_training, roc_test = training(model, test_dataloader, training_dataloader, n_epoch, optimizer, filename, logname)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Train model: Activity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_embedding  = 100\n",
    "n_hidden = 400\n",
    "n_layers = 2\n",
    "n_epoch = 150\n",
    "learning_rate = 0.01\n",
    "momentum = 0.9\n",
    "batch_size = 20\n",
    "\n",
    "train_splits(activity, 'activity', n_embedding, n_hidden, n_layers, learning_rate, momentum, batch_size, n_epoch)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Train model: Hemolysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_embedding  = 100\n",
    "n_hidden = 400\n",
    "n_layers = 1\n",
    "n_epoch = 150\n",
    "learning_rate = 0.01\n",
    "momentum = 0.9\n",
    "batch_size = 20\n",
    "\n",
    "train_splits(hemolysis, 'hemolysis', n_embedding, n_hidden, n_layers, learning_rate, momentum, batch_size, n_epoch)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Evaluate RNNs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Determine best model for each split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_roc_curves(path, n_splits):\n",
    "    \n",
    "    roc_training = []\n",
    "    roc_test = []\n",
    "    colors = ['#648FFF', '#785EF0', '#DC267F', '#FE6100', '#FFB000']\n",
    "    \n",
    "    for i in range(1, n_splits + 1):\n",
    "        log = pd.read_csv(f'models/RNN_{path}/split_{i}/log.txt', skiprows=[0, 1, 2, 3, 4, 5, 6, 7])\n",
    "        roc_training.append(log['Training ROC-AUC'])\n",
    "        roc_test.append(log['Test ROC-AUC'])\n",
    "\n",
    "    fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(15, 6.5), sharey=True)\n",
    "    for i in range(n_splits):\n",
    "        ax1.plot(roc_training[i], label=f'Split {i+1}', color=colors[i])\n",
    "        ax2.plot(roc_test[i], label=f'Split {i+1}', color=colors[i]) \n",
    "    \n",
    "    ax1.set_xlim(0, 150)\n",
    "    ax2.set_xlim(0, 150)\n",
    "    ax1.set_ylim(0, 1.05)\n",
    "    ax2.set_ylim(0, 1.05)\n",
    "\n",
    "    ax1.set_xlabel('Epoch')\n",
    "    ax2.set_xlabel('Epoch')\n",
    "    ax1.set_ylabel('ROC-AUC')\n",
    "    \n",
    "    ax1.set_title('Training ROC')\n",
    "    ax2.set_title('Test ROC')\n",
    "    ax2.legend(loc='lower center', bbox_to_anchor=(1.275, 0.225))\n",
    "    \n",
    "    plt.savefig(f'figures/RNN_{path}_ROC.png', dpi=1200, bbox_inches='tight')\n",
    "\n",
    "def determine_best_models(path, n_splits):\n",
    "    \n",
    "    best_epochs = []\n",
    "    for i in range(1, n_splits + 1):\n",
    "        log = pd.read_csv(f'models/RNN_{path}/split_{i}/log.txt', skiprows=[0, 1, 2, 3, 4, 5, 6, 7])\n",
    "        best_epochs.append(log['Test ROC-AUC'].idxmax() + 1)\n",
    "    \n",
    "    best_models = []\n",
    "    for i in range(1, n_splits + 1):\n",
    "        if path == 'activity':\n",
    "            best_models.append(f'models/RNN_{path}/split_{i}/em100_hi400_la2_ep{best_epochs[i-1]}')\n",
    "        elif path == 'hemolysis':\n",
    "            best_models.append(f'models/RNN_{path}/split_{i}/em100_hi400_la1_ep{best_epochs[i-1]}')\n",
    "    return best_models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x650 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "activity_rocs = plot_roc_curves('activity', 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x650 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "hemolysis_rocs = plot_roc_curves('hemolysis', 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculate metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict(model, data_loader):\n",
    "    \n",
    "    cat_list = []\n",
    "    out_list = []\n",
    "\n",
    "    with torch.no_grad():\n",
    "        for i_batch, sample_batched in enumerate(data_loader):    \n",
    "            seq_batched = sample_batched[0][0].to(model.device, non_blocking=True)\n",
    "            seq_lengths = sample_batched[0][1].to(model.device, non_blocking=True)\n",
    "\n",
    "            cat_list += sample_batched[1].to(\"cpu\", non_blocking=True)\n",
    "            out_list += torch.exp(model.evaluate(seq_batched, seq_lengths))[: ,1].to(\"cpu\", non_blocking=True)\n",
    "\n",
    "        cat_list = torch.stack(cat_list)\n",
    "        out_list = torch.stack(out_list)\n",
    "    return cat_list.cpu().numpy().astype(int), out_list.cpu().numpy()\n",
    "\n",
    "def calculate_metrics(y_true, y_pred, threshold = 0.99205756):\n",
    "    \n",
    "    y_pred_converted = y_pred > threshold\n",
    "    y_pred_converted = y_pred_converted.astype(int)\n",
    "    \n",
    "    roc_auc = roc_auc_score(y_true, y_pred)\n",
    "    accuracy = accuracy_score(y_true, y_pred_converted)\n",
    "    precision = precision_score(y_true, y_pred_converted)\n",
    "    recall = recall_score(y_true, y_pred_converted)\n",
    "    f1 = f1_score(y_true, y_pred_converted)\n",
    "    \n",
    "    return pd.DataFrame({'roc_auc': [roc_auc], 'accuracy': [accuracy], 'precision': [precision], 'recall': [recall], 'f1': [f1]})\n",
    "\n",
    "def evaluate_models(df, path, n_splits, vocabulary, threshold):\n",
    "\n",
    "    df_out = pd.DataFrame()\n",
    "    filenames = determine_best_models(path, n_splits)\n",
    "\n",
    "    for i in range(1, n_splits+1):\n",
    "\n",
    "        test = Dataset(df[df[f'split_{i}'] == 'test'], vocabulary)\n",
    "        dataloader = torch.utils.data.DataLoader(test, batch_size=1, shuffle=False, collate_fn = collate_fn, drop_last=True, pin_memory=True, num_workers=4)\n",
    "\n",
    "        model = Classifier.load_from_file(filenames[i-1])\n",
    "        model.to(device)\n",
    "        \n",
    "        y_true, y_pred = predict(model, dataloader)\n",
    "        row = calculate_metrics(y_true, y_pred, threshold)\n",
    "\n",
    "        df_out = pd.concat([df_out, row], ignore_index=True)\n",
    "    \n",
    "    df_out = pd.concat([df_out, pd.DataFrame(df_out.mean()).T], ignore_index=True)\n",
    "    df_out = pd.concat([df_out, pd.DataFrame(df_out.std()).T], ignore_index=True)\n",
    "    df_out.index = [f'fold_{i}' for i in range(1, 6)] + ['mean', 'std']\n",
    "\n",
    "    return df_out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "activity_eval = evaluate_models(activity, 'activity', 5, vocabulary, 0.5)\n",
    "hemolysis_eval = evaluate_models(hemolysis, 'hemolysis', 5, vocabulary, 0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Export data frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "activity_eval.to_csv('results/rnn_activity_5cv.csv')\n",
    "hemolysis_eval.to_csv('results/rnn_hemolysis_5cv.csv')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "openaienv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}