diff --git a/example-notebooks/Test new_gamma_survival_model with simulated data.ipynb b/example-notebooks/Test new_gamma_survival_model with simulated data.ipynb index e0335a4..fcfc4de 100644 --- a/example-notebooks/Test new_gamma_survival_model with simulated data.ipynb +++ b/example-notebooks/Test new_gamma_survival_model with simulated data.ipynb @@ -53,7 +53,13 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:stancache.stancache:sim_data_exp_correlated: cache_filename set to sim_data_exp_correlated.cached.N_100.censor_time_20.rate_coefs_54462717316.rate_form_1 + sex.pkl\n", + "INFO:stancache.stancache:sim_data_exp_correlated: cache_filename set to sim_data_exp_correlated.cached.N_100.censor_time_20.rate_coefs_54462717316.rate_form_1 + sex.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ "INFO:stancache.stancache:sim_data_exp_correlated: Loading result from cache\n" ] } @@ -226,7 +232,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 4, @@ -347,31 +353,42 @@ "name": "stderr", "output_type": "stream", "text": [ - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "/home/jacki/projects/survivalstan/survivalstan/survivalstan.py:368: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n", - " 'x': self.x_df.as_matrix(),\n", - "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n", - "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_14429915565770599621.pystan_2_18_1_0.stanmodel.pkl\n", - "INFO:stancache.stancache:StanModel: Starting execution\n", - "INFO:pystan:COMPILING THE C++ CODE FOR MODEL anon_model_f418131a20ef101bf99f9c02185578f1 NOW.\n", - "INFO:stancache.stancache:StanModel: Execution completed (0:01:43.433295 elapsed)\n", - "INFO:stancache.stancache:StanModel: Saving results to cache\n", - "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n", - "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_14429915565770599621.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n", - "INFO:stancache.stancache:sampling: Starting execution\n", - "INFO:stancache.stancache:sampling: Execution completed (0:01:33.270480 elapsed)\n", - "INFO:stancache.stancache:sampling: Saving results to cache\n", - "/srv/conda/lib/python3.6/site-packages/stancache/stancache.py:285: UserWarning: Pickling fit objects is an experimental feature!\n", - "The relevant StanModel instance must be pickled along with this fit object.\n", - "When unpickling the StanModel must be unpickled first.\n", - " pickle.dump(res, open(cache_filepath, 'wb'), pickle.HIGHEST_PROTOCOL)\n" + "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_14429915565770599621.pystan_2_18_1_0.stanmodel.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: Loading result from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_14429915565770599621.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: Loading result from cache\n" ] } ], @@ -486,31 +503,42 @@ "name": "stderr", "output_type": "stream", "text": [ - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "/home/jacki/projects/survivalstan/survivalstan/survivalstan.py:368: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n", - " 'x': self.x_df.as_matrix(),\n", - "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n", - "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_9177012762674257483.pystan_2_18_1_0.stanmodel.pkl\n", - "INFO:stancache.stancache:StanModel: Starting execution\n", - "INFO:pystan:COMPILING THE C++ CODE FOR MODEL anon_model_8758b4bc02931efdf69b3e349afdba01 NOW.\n", - "INFO:stancache.stancache:StanModel: Execution completed (0:01:29.291961 elapsed)\n", - "INFO:stancache.stancache:StanModel: Saving results to cache\n", - "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n", - "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_9177012762674257483.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n", - "INFO:stancache.stancache:sampling: Starting execution\n", - "INFO:stancache.stancache:sampling: Execution completed (0:01:20.742172 elapsed)\n", - "INFO:stancache.stancache:sampling: Saving results to cache\n", - "/srv/conda/lib/python3.6/site-packages/stancache/stancache.py:285: UserWarning: Pickling fit objects is an experimental feature!\n", - "The relevant StanModel instance must be pickled along with this fit object.\n", - "When unpickling the StanModel must be unpickled first.\n", - " pickle.dump(res, open(cache_filepath, 'wb'), pickle.HIGHEST_PROTOCOL)\n" + "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_9177012762674257483.pystan_2_18_1_0.stanmodel.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: Loading result from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_9177012762674257483.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: Loading result from cache\n" ] } ], @@ -532,7 +560,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 11, "metadata": { "collapsed": true, "jupyter": { @@ -627,29 +655,42 @@ "name": "stderr", "output_type": "stream", "text": [ - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n", - "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_1293841621968646714.pystan_2_18_1_0.stanmodel.pkl\n", - "INFO:stancache.stancache:StanModel: Starting execution\n", - "INFO:pystan:COMPILING THE C++ CODE FOR MODEL anon_model_050903aecc782294461f27307e7596e0 NOW.\n", - "INFO:stancache.stancache:StanModel: Execution completed (0:01:42.204878 elapsed)\n", - "INFO:stancache.stancache:StanModel: Saving results to cache\n", - "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n", - "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_1293841621968646714.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n", - "INFO:stancache.stancache:sampling: Starting execution\n", - "INFO:stancache.stancache:sampling: Execution completed (0:00:42.036498 elapsed)\n", - "INFO:stancache.stancache:sampling: Saving results to cache\n", - "/srv/conda/lib/python3.6/site-packages/stancache/stancache.py:285: UserWarning: Pickling fit objects is an experimental feature!\n", - "The relevant StanModel instance must be pickled along with this fit object.\n", - "When unpickling the StanModel must be unpickled first.\n", - " pickle.dump(res, open(cache_filepath, 'wb'), pickle.HIGHEST_PROTOCOL)\n" + "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_1293841621968646714.pystan_2_18_1_0.stanmodel.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: Loading result from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_1293841621968646714.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: Loading result from cache\n" ] } ], @@ -791,29 +832,42 @@ "name": "stderr", "output_type": "stream", "text": [ - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n", - "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_16881928540873162731.pystan_2_18_1_0.stanmodel.pkl\n", - "INFO:stancache.stancache:StanModel: Starting execution\n", - "INFO:pystan:COMPILING THE C++ CODE FOR MODEL anon_model_b18a6495e568fcff90662e16a3d2aa85 NOW.\n", - "INFO:stancache.stancache:StanModel: Execution completed (0:01:34.346981 elapsed)\n", - "INFO:stancache.stancache:StanModel: Saving results to cache\n", - "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n", - "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_16881928540873162731.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n", - "INFO:stancache.stancache:sampling: Starting execution\n", - "INFO:stancache.stancache:sampling: Execution completed (0:00:16.703755 elapsed)\n", - "INFO:stancache.stancache:sampling: Saving results to cache\n", - "/srv/conda/lib/python3.6/site-packages/stancache/stancache.py:285: UserWarning: Pickling fit objects is an experimental feature!\n", - "The relevant StanModel instance must be pickled along with this fit object.\n", - "When unpickling the StanModel must be unpickled first.\n", - " pickle.dump(res, open(cache_filepath, 'wb'), pickle.HIGHEST_PROTOCOL)\n" + "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_16881928540873162731.pystan_2_18_1_0.stanmodel.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:StanModel: Loading result from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_16881928540873162731.pystan_2_18_1_0.stanfit.chains_4.data_36753546383.iter_5000.seed_9001.pkl\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:stancache.stancache:sampling: Loading result from cache\n" ] } ], diff --git a/example-notebooks/Test pem_survival_model_randomwalk with simulated data.ipynb b/example-notebooks/Test pem_survival_model_randomwalk with simulated data.ipynb index 4c402a7..50e92ba 100644 --- a/example-notebooks/Test pem_survival_model_randomwalk with simulated data.ipynb +++ b/example-notebooks/Test pem_survival_model_randomwalk with simulated data.ipynb @@ -356,7 +356,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 5, @@ -557,22 +557,6 @@ "name": "stderr", "output_type": "stream", "text": [ - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:96: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.stan_data = stan_data\n", - "/home/jacki/projects/survivalstan/survivalstan/formulas.py:97: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access\n", - " self.meta_data = meta_data\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/home/jacki/projects/survivalstan/survivalstan/survivalstan.py:368: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n", - " 'x': self.x_df.as_matrix(),\n", "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n" ] }, @@ -824,7 +808,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 14, diff --git a/example-notebooks/Test pem_survival_model_timevarying with simulated data.ipynb b/example-notebooks/Test pem_survival_model_timevarying with simulated data.ipynb index 17a8eb1..1afdf8d 100644 --- a/example-notebooks/Test pem_survival_model_timevarying with simulated data.ipynb +++ b/example-notebooks/Test pem_survival_model_timevarying with simulated data.ipynb @@ -270,13 +270,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:stancache.stancache:sim_data_exp_correlated: cache_filename set to sim_data_exp_correlated.cached.N_100.censor_time_20.rate_coefs_54462717316.rate_form_1 + sex.pkl\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ + "INFO:stancache.stancache:sim_data_exp_correlated: cache_filename set to sim_data_exp_correlated.cached.N_100.censor_time_20.rate_coefs_54462717316.rate_form_1 + sex.pkl\n", "INFO:stancache.stancache:sim_data_exp_correlated: Loading result from cache\n" ] }, @@ -407,7 +401,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 4, @@ -444,13 +438,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:stancache.stancache:prep_data_long_surv: cache_filename set to prep_data_long_surv.cached.df_14209590808.event_col_event.time_col_t.pkl\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ + "INFO:stancache.stancache:prep_data_long_surv: cache_filename set to prep_data_long_surv.cached.df_14209590808.event_col_event.time_col_t.pkl\n", "INFO:stancache.stancache:prep_data_long_surv: Loading result from cache\n" ] } @@ -609,129 +597,12 @@ "name": "stderr", "output_type": "stream", "text": [ - "/home/jacki/projects/survivalstan/survivalstan/survivalstan.py:368: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n", - " 'x': self.x_df.as_matrix(),\n", - "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_10216236489136838232.pystan_2_18_1_0.stanmodel.pkl\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:stancache.stancache:StanModel: Loading result from cache\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_10216236489136838232.pystan_2_18_1_0.stanfit.chains_4.data_98562805320.iter_10000.seed_9001.pkl\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:stancache.stancache:sampling: Starting execution\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:Maximum (flat) parameter count (1000) exceeded: skipping diagnostic tests for n_eff and Rhat.\n", - "To run all diagnostics call pystan.check_hmc_diagnostics(fit)\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:457 of 20000 iterations ended with a divergence (2.29 %).\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:Try running with adapt_delta larger than 0.8 to remove the divergences.\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:16349 of 20000 iterations saturated the maximum tree depth of 10 (81.7 %)\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:Run again with max_treedepth larger than 10 to avoid saturation\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:Chain 1: E-BFMI = 0.024\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:Chain 2: E-BFMI = 0.00708\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:Chain 3: E-BFMI = 0.0191\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:Chain 4: E-BFMI = 0.0211\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "WARNING:pystan:E-BFMI below 0.2 indicates you may need to reparameterize your model\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:stancache.stancache:sampling: Execution completed (3:12:36.548684 elapsed)\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:stancache.stancache:sampling: Saving results to cache\n" + "INFO:stancache.stancache:Step 1: Get compiled model code, possibly from cache\n", + "INFO:stancache.stancache:StanModel: cache_filename set to anon_model.cython_0_29_2.model_code_10216236489136838232.pystan_2_18_1_0.stanmodel.pkl\n", + "INFO:stancache.stancache:StanModel: Loading result from cache\n", + "INFO:stancache.stancache:Step 2: Get posterior draws from model, possibly from cache\n", + "INFO:stancache.stancache:sampling: cache_filename set to anon_model.cython_0_29_2.model_code_10216236489136838232.pystan_2_18_1_0.stanfit.chains_4.data_98562805320.iter_10000.seed_9001.pkl\n", + "INFO:stancache.stancache:sampling: Loading result from cache\n" ] } ], @@ -874,7 +745,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 12, @@ -950,9 +821,7 @@ }, { "cell_type": "markdown", - "metadata": { - "collapsed": true - }, + "metadata": {}, "source": [ "## summarize time-varying effect of sex on survival" ] @@ -972,23 +841,16 @@ }, "outputs": [ { - "ename": "ValueError", - "evalue": "You are trying to merge on object and int64 columns. If you wish to proceed you should use pd.concat", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0msurvivalstan\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot_coefs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mtestfit\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0melement\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'beta_time'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mylim\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2.5\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36mplot_coefs\u001b[0;34m(models, element, force_direction, trans, by, **kwargs)\u001b[0m\n\u001b[1;32m 944\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0melement\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'beta_time'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 945\u001b[0m return plot_time_betas(models=models, element=element,\n\u001b[0;32m--> 946\u001b[0;31m trans=trans, **kwargs)\n\u001b[0m\u001b[1;32m 947\u001b[0m \u001b[0;31m# prep data from models given\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 948\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0melement\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'baseline'\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0melement\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'baseline_raw'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36mplot_time_betas\u001b[0;34m(models, df, element, y, trans, coefs, x, by, timepoint_id_col, timepoint_end_col, subplot, ticks_at, ylabel, xlabel, num_ticks, step_size, fill, alpha, pal, value_name, **kwargs)\u001b[0m\n\u001b[1;32m 339\u001b[0m \u001b[0mvalue_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mvalue_name\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 340\u001b[0m \u001b[0mtimepoint_id_col\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimepoint_id_col\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 341\u001b[0;31m timepoint_end_col=timepoint_end_col)\n\u001b[0m\u001b[1;32m 342\u001b[0m timepoint_id_col, timepoint_end_col = _get_timepoint_cols(\n\u001b[1;32m 343\u001b[0m \u001b[0mmodels\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmodels\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36mextract_time_betas\u001b[0;34m(models, element, value_name, **kwargs)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0melement\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0melement\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m value_name=value_name, **kwargs)\n\u001b[0;32m---> 92\u001b[0;31m for model in models]\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconcat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0melement\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0melement\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m value_name=value_name, **kwargs)\n\u001b[0;32m---> 92\u001b[0;31m for model in models]\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconcat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36m_extract_time_betas_single_model\u001b[0;34m(stanmodel, element, coefs, value_name, timepoint_id_col, timepoint_end_col)\u001b[0m\n\u001b[1;32m 143\u001b[0m \u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mtimepoint_id_col\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtimepoint_end_col\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 144\u001b[0m .drop_duplicates())\n\u001b[0;32m--> 145\u001b[0;31m \u001b[0mtime_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmerge\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtimepoint_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mon\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimepoint_id_col\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 146\u001b[0m \u001b[0mtime_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'exp({})'\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvalue_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexp\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mvalue_name\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 147\u001b[0m \u001b[0mtime_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'model_cohort'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstanmodel\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'model_cohort'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/srv/conda/lib/python3.6/site-packages/pandas/core/reshape/merge.py\u001b[0m in \u001b[0;36mmerge\u001b[0;34m(left, right, how, on, left_on, right_on, left_index, right_index, sort, suffixes, copy, indicator, validate)\u001b[0m\n\u001b[1;32m 59\u001b[0m \u001b[0mright_index\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mright_index\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msort\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msort\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msuffixes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msuffixes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 60\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindicator\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mindicator\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 61\u001b[0;31m validate=validate)\n\u001b[0m\u001b[1;32m 62\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mop\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_result\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 63\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/srv/conda/lib/python3.6/site-packages/pandas/core/reshape/merge.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, left, right, how, on, left_on, right_on, axis, left_index, right_index, sort, suffixes, copy, indicator, validate)\u001b[0m\n\u001b[1;32m 553\u001b[0m \u001b[0;31m# validate the merge keys dtypes. We may need to coerce\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 554\u001b[0m \u001b[0;31m# to avoid incompat dtypes\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 555\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_coerce_merge_keys\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 556\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 557\u001b[0m \u001b[0;31m# If argument passed to validate,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/srv/conda/lib/python3.6/site-packages/pandas/core/reshape/merge.py\u001b[0m in \u001b[0;36m_maybe_coerce_merge_keys\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 984\u001b[0m elif (not is_numeric_dtype(lk)\n\u001b[1;32m 985\u001b[0m and (is_numeric_dtype(rk) and not is_bool_dtype(rk))):\n\u001b[0;32m--> 986\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 987\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0mis_datetimelike\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlk\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mis_datetimelike\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrk\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 988\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mValueError\u001b[0m: You are trying to merge on object and int64 columns. If you wish to proceed you should use pd.concat" - ] + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" } ], "source": [ @@ -1004,19 +866,21 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 16, "metadata": { "collapsed": false }, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiUAAAF4CAYAAACLlYlfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRJQIgiAIgtAXiCgRBEEQBKEv+P8gocGO6FgohgAAAABJRU5ErkJggg==\n", + "image/png": "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\n", "text/plain": [ - "" + "
" ] }, - "metadata": {}, + "metadata": { + "needs_background": "light" + }, "output_type": "display_data" } ], @@ -1033,7 +897,7 @@ }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 17, "metadata": { "collapsed": false }, @@ -1042,11 +906,25 @@ "data": { "text/html": [ "
\n", + "\n", "\n", " \n", " \n", " \n", - " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -1057,47 +935,52 @@ " \n", " \n", " \n", + " \n", " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", + " \n", " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", + " \n", " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", + " \n", " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -1105,15 +988,15 @@ "" ], "text/plain": [ - " timepoint_id beta coef end_time exp(beta) model_cohort\n", - "0 1 0.281855 sex[T.male] 0.118611 1.325587 test model\n", - "1 1 1.263018 sex[T.male] 0.118611 3.536076 test model\n", - "2 1 0.875265 sex[T.male] 0.118611 2.399512 test model\n", - "3 1 0.128097 sex[T.male] 0.118611 1.136664 test model\n", - "4 1 0.185878 sex[T.male] 0.118611 1.204276 test model" + " iter _timepoint_id beta coef end_time exp(beta) model_cohort\n", + "0 0 1 1.303891 sex 0.009787 3.683601 test model\n", + "1 1 1 0.349997 sex 0.009787 1.419063 test model\n", + "2 2 1 0.348042 sex 0.009787 1.416292 test model\n", + "3 3 1 0.200147 sex 0.009787 1.221583 test model\n", + "4 4 1 0.189725 sex 0.009787 1.208917 test model" ] }, - "execution_count": 27, + "execution_count": 17, "metadata": {}, "output_type": "execute_result" } @@ -1132,89 +1015,26 @@ }, { "cell_type": "code", - "execution_count": 28, + "execution_count": 18, "metadata": { "collapsed": false }, "outputs": [ { - "data": { - "text/html": [ - "
\n", - "
timepoint_iditer_timepoint_idbetacoefend_time
0010.281855sex[T.male]0.1186111.3255871.303891sex0.0097873.683601test model
111.263018sex[T.male]0.1186113.53607610.349997sex0.0097871.419063test model
2210.875265sex[T.male]0.1186112.3995120.348042sex0.0097871.416292test model
3310.128097sex[T.male]0.1186111.1366640.200147sex0.0097871.221583test model
4410.185878sex[T.male]0.1186111.2042760.189725sex0.0097871.208917test model
\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
timepoint_idbetacoefend_timeexp(beta)model_cohort
010.281855sex[T.male]0.1186111.325587test model
111.263018sex[T.male]0.1186113.536076test model
210.875265sex[T.male]0.1186112.399512test model
310.128097sex[T.male]0.1186111.136664test model
410.185878sex[T.male]0.1186111.204276test model
\n", - "
" - ], - "text/plain": [ - " timepoint_id beta coef end_time exp(beta) model_cohort\n", - "0 1 0.281855 sex[T.male] 0.118611 1.325587 test model\n", - "1 1 1.263018 sex[T.male] 0.118611 3.536076 test model\n", - "2 1 0.875265 sex[T.male] 0.118611 2.399512 test model\n", - "3 1 0.128097 sex[T.male] 0.118611 1.136664 test model\n", - "4 1 0.185878 sex[T.male] 0.118611 1.204276 test model" - ] - }, - "execution_count": 28, - "metadata": {}, - "output_type": "execute_result" + "ename": "ValueError", + "evalue": "No objects to concatenate", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mfirst_beta\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msurvivalstan\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mextract_time_betas\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mtestfit\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcoefs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'sex[T.male]'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0mfirst_beta\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhead\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36mextract_time_betas\u001b[0;34m(models, element, value_name, **kwargs)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0melement\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0melement\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m value_name=value_name, **kwargs)\n\u001b[0;32m---> 92\u001b[0;31m for model in models]\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconcat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0melement\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0melement\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 91\u001b[0m value_name=value_name, **kwargs)\n\u001b[0;32m---> 92\u001b[0;31m for model in models]\n\u001b[0m\u001b[1;32m 93\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconcat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/projects/survivalstan/survivalstan/utils.py\u001b[0m in \u001b[0;36m_extract_time_betas_single_model\u001b[0;34m(stanmodel, element, coefs, value_name, timepoint_id_col, timepoint_end_col)\u001b[0m\n\u001b[1;32m 139\u001b[0m \u001b[0mtb_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'coef'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcoef_names\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 140\u001b[0m \u001b[0mtime_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtb_df\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 141\u001b[0;31m \u001b[0mtime_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconcat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime_data\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 142\u001b[0m timepoint_data = (stanmodel['df']\n\u001b[1;32m 143\u001b[0m \u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mtimepoint_id_col\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtimepoint_end_col\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/srv/conda/lib/python3.6/site-packages/pandas/core/reshape/concat.py\u001b[0m in \u001b[0;36mconcat\u001b[0;34m(objs, axis, join, join_axes, ignore_index, keys, levels, names, verify_integrity, sort, copy)\u001b[0m\n\u001b[1;32m 223\u001b[0m \u001b[0mkeys\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mkeys\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlevels\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlevels\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnames\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnames\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 224\u001b[0m \u001b[0mverify_integrity\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mverify_integrity\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 225\u001b[0;31m copy=copy, sort=sort)\n\u001b[0m\u001b[1;32m 226\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mop\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_result\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 227\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/srv/conda/lib/python3.6/site-packages/pandas/core/reshape/concat.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, objs, axis, join, join_axes, keys, levels, names, ignore_index, verify_integrity, copy, sort)\u001b[0m\n\u001b[1;32m 257\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 258\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mobjs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 259\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'No objects to concatenate'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 260\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 261\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mkeys\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mValueError\u001b[0m: No objects to concatenate" + ] } ], "source": [ @@ -1335,9 +1155,9 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.5.2" + "version": "3.6.7" } }, "nbformat": 4, - "nbformat_minor": 0 + "nbformat_minor": 2 }