polarolouis/REE-RL-Lola
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Adding plots and sigmoid for relevant params

Browse source
This commit is contained in:
Louis 2025-12-10 22:19:03 +01:00
parent 28efe2e515
commit 8aa1d27675
1 changed files with 26 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

33
pyvbmc_optimization.py
View file

@ -348,7 +348,7 @@ def qlearning_generic(
def fit_participant_pyvbmc( def fit_participant_pyvbmc(
participant_data: pd.DataFrame, model_config: Dict, verbose: bool = True participant_data: pd.DataFrame, model_config: Dict, verbose: bool = True, plot : bool = True
) -> Dict: ) -> Dict:
""" """
Optimise les paramètres du modèle pour un participant utilisant PyVBMC. Optimise les paramètres du modèle pour un participant utilisant PyVBMC.
@ -399,6 +399,8 @@ def fit_participant_pyvbmc(
options={ options={
# "verbose": 0 if not verbose else 1, # "verbose": 0 if not verbose else 1,
"display": "off", "display": "off",
"plot": plot,
"log_file_name": None,
}, },
prior=UniformBox( prior=UniformBox(
a=model_config["lower"], b=model_config["upper"], D=model_config["n_params"] a=model_config["lower"], b=model_config["upper"], D=model_config["n_params"]
@ -446,6 +448,11 @@ def fit_participant_pyvbmc(
# Ajout des paramètres estimés # Ajout des paramètres estimés
for i, param_name in enumerate(model_config["param_names"]): for i, param_name in enumerate(model_config["param_names"]):
# Here we use expit for parameters that were originally bounded between 0 and 1
if param_name.startswith("alpha") or param_name.startswith("forget"):
result[param_name] = expit(posterior_mean[i])
result[f"sd_{param_name}"] = expit(posterior_sd[i])
else:
result[param_name] = posterior_mean[i] result[param_name] = posterior_mean[i]
result[f"sd_{param_name}"] = posterior_sd[i] result[f"sd_{param_name}"] = posterior_sd[i]
@ -530,8 +537,12 @@ def fit_participant_deoptim(
"posterior_mean": posterior_mean, "posterior_mean": posterior_mean,
} }
# Ajout des paramètres estimés # Ajout des paramètres estimés après les avoir renvoyés dans par logis
for i, param_name in enumerate(model_config["param_names"]): for i, param_name in enumerate(model_config["param_names"]):
# Here we expit alpha, forget
if param_name.startswith("alpha") or param_name.startswith("forget"):
result_dict[param_name] = expit(posterior_mean[i])
else:
result_dict[param_name] = posterior_mean[i] result_dict[param_name] = posterior_mean[i]
return result_dict return result_dict
@ -548,6 +559,7 @@ def fit_all_participants(
method: str = "VBMC", method: str = "VBMC",
n_participants: Optional[int] = None, n_participants: Optional[int] = None,
verbose: bool = True, verbose: bool = True,
plot: bool = True,
) -> Dict[str, List[Dict]]: ) -> Dict[str, List[Dict]]:
""" """
Ajuste tous les modèles pour tous les participants. Ajuste tous les modèles pour tous les participants.
@ -588,7 +600,7 @@ def fit_all_participants(
try: try:
if method == "VBMC": if method == "VBMC":
result = fit_participant_pyvbmc( result = fit_participant_pyvbmc(
participant_data, model_config, verbose=False participant_data, model_config, verbose=False, plot=plot
) )
else: else:
result = fit_participant_deoptim( result = fit_participant_deoptim(
@ -849,14 +861,14 @@ if __name__ == "__main__":
# Ajustement de quelques modèles pour test # Ajustement de quelques modèles pour test
models_to_fit = [ models_to_fit = [
"HOMOGENEOUS", # "HOMOGENEOUS",
"GAIN_LOSS", # "GAIN_LOSS",
"REE_BIASED_SIMPLE", # "REE_BIASED_SIMPLE",
"REE_BIASED_COMPLEX", # "REE_BIASED_COMPLEX",
"REE_LEARNING_SIMPLE", # "REE_LEARNING_SIMPLE",
"REE_LEARNING_COMPLEX", # "REE_LEARNING_COMPLEX",
"REE_LEARNING_BIASED_SIMPLE", "REE_LEARNING_BIASED_SIMPLE",
"REE_LEARNING_BIASED_COMPLEX", # "REE_LEARNING_BIASED_COMPLEX",
] ]
# all_results = fit_all_participants_both_methods( # all_results = fit_all_participants_both_methods(
@ -874,6 +886,7 @@ if __name__ == "__main__":
method=method, method=method,
n_participants=1, # Set to a number to limit for testing n_participants=1, # Set to a number to limit for testing
verbose=True, verbose=True,
plot=True
) )
# Comparaison des modèles # Comparaison des modèles