Replace deprecated optuna.suggest_loguniform(...) by optuna.suggest_float(..., log=True) (#362)

Author: qgallouedec

CHANGELOG.md

@@ -17,6 +17,7 @@
 ### Other
 - Added support for `ruff` (fast alternative to flake8) in the Makefile
 - Removed Gitlab CI file
+- Replaced deprecated `optuna.suggest_loguniform(...)` by `optuna.suggest_float(..., log=True)`
 
 ## Release 1.7.0 (2023-01-10)
 

rl_zoo3/hyperparams_opt.py

@@ -18,11 +18,11 @@ def sample_ppo_params(trial: optuna.Trial) -> Dict[str, Any]:
     batch_size = trial.suggest_categorical("batch_size", [8, 16, 32, 64, 128, 256, 512])
     n_steps = trial.suggest_categorical("n_steps", [8, 16, 32, 64, 128, 256, 512, 1024, 2048])
     gamma = trial.suggest_categorical("gamma", [0.9, 0.95, 0.98, 0.99, 0.995, 0.999, 0.9999])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
     lr_schedule = "constant"
     # Uncomment to enable learning rate schedule
     # lr_schedule = trial.suggest_categorical('lr_schedule', ['linear', 'constant'])
-    ent_coef = trial.suggest_loguniform("ent_coef", 0.00000001, 0.1)
+    ent_coef = trial.suggest_float("ent_coef", 0.00000001, 0.1, log=True)
     clip_range = trial.suggest_categorical("clip_range", [0.1, 0.2, 0.3, 0.4])
     n_epochs = trial.suggest_categorical("n_epochs", [1, 5, 10, 20])
     gae_lambda = trial.suggest_categorical("gae_lambda", [0.8, 0.9, 0.92, 0.95, 0.98, 0.99, 1.0])
@@ -86,7 +86,7 @@ def sample_trpo_params(trial: optuna.Trial) -> Dict[str, Any]:
     batch_size = trial.suggest_categorical("batch_size", [8, 16, 32, 64, 128, 256, 512])
     n_steps = trial.suggest_categorical("n_steps", [8, 16, 32, 64, 128, 256, 512, 1024, 2048])
     gamma = trial.suggest_categorical("gamma", [0.9, 0.95, 0.98, 0.99, 0.995, 0.999, 0.9999])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
     lr_schedule = "constant"
     # Uncomment to enable learning rate schedule
     # lr_schedule = trial.suggest_categorical('lr_schedule', ['linear', 'constant'])
@@ -159,8 +159,8 @@ def sample_a2c_params(trial: optuna.Trial) -> Dict[str, Any]:
     gae_lambda = trial.suggest_categorical("gae_lambda", [0.8, 0.9, 0.92, 0.95, 0.98, 0.99, 1.0])
     n_steps = trial.suggest_categorical("n_steps", [8, 16, 32, 64, 128, 256, 512, 1024, 2048])
     lr_schedule = trial.suggest_categorical("lr_schedule", ["linear", "constant"])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
-    ent_coef = trial.suggest_loguniform("ent_coef", 0.00000001, 0.1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
+    ent_coef = trial.suggest_float("ent_coef", 0.00000001, 0.1, log=True)
     vf_coef = trial.suggest_uniform("vf_coef", 0, 1)
     # Uncomment for gSDE (continuous actions)
     # log_std_init = trial.suggest_uniform("log_std_init", -4, 1)
@@ -216,7 +216,7 @@ def sample_sac_params(trial: optuna.Trial) -> Dict[str, Any]:
     :return:
     """
     gamma = trial.suggest_categorical("gamma", [0.9, 0.95, 0.98, 0.99, 0.995, 0.999, 0.9999])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
     batch_size = trial.suggest_categorical("batch_size", [16, 32, 64, 128, 256, 512, 1024, 2048])
     buffer_size = trial.suggest_categorical("buffer_size", [int(1e4), int(1e5), int(1e6)])
     learning_starts = trial.suggest_categorical("learning_starts", [0, 1000, 10000, 20000])
@@ -277,7 +277,7 @@ def sample_td3_params(trial: optuna.Trial) -> Dict[str, Any]:
     :return:
     """
     gamma = trial.suggest_categorical("gamma", [0.9, 0.95, 0.98, 0.99, 0.995, 0.999, 0.9999])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
     batch_size = trial.suggest_categorical("batch_size", [16, 32, 64, 100, 128, 256, 512, 1024, 2048])
     buffer_size = trial.suggest_categorical("buffer_size", [int(1e4), int(1e5), int(1e6)])
     # Polyak coeff
@@ -335,7 +335,7 @@ def sample_ddpg_params(trial: optuna.Trial) -> Dict[str, Any]:
     :return:
     """
     gamma = trial.suggest_categorical("gamma", [0.9, 0.95, 0.98, 0.99, 0.995, 0.999, 0.9999])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
     batch_size = trial.suggest_categorical("batch_size", [16, 32, 64, 100, 128, 256, 512, 1024, 2048])
     buffer_size = trial.suggest_categorical("buffer_size", [int(1e4), int(1e5), int(1e6)])
     # Polyak coeff
@@ -391,7 +391,7 @@ def sample_dqn_params(trial: optuna.Trial) -> Dict[str, Any]:
     :return:
     """
     gamma = trial.suggest_categorical("gamma", [0.9, 0.95, 0.98, 0.99, 0.995, 0.999, 0.9999])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
     batch_size = trial.suggest_categorical("batch_size", [16, 32, 64, 100, 128, 256, 512])
     buffer_size = trial.suggest_categorical("buffer_size", [int(1e4), int(5e4), int(1e5), int(1e6)])
     exploration_final_eps = trial.suggest_uniform("exploration_final_eps", 0, 0.2)
@@ -489,7 +489,7 @@ def sample_ars_params(trial: optuna.Trial) -> Dict[str, Any]:
     # n_eval_episodes = trial.suggest_categorical("n_eval_episodes", [1, 2])
     n_delta = trial.suggest_categorical("n_delta", [4, 8, 6, 32, 64])
     # learning_rate = trial.suggest_categorical("learning_rate", [0.01, 0.02, 0.025, 0.03])
-    learning_rate = trial.suggest_loguniform("learning_rate", 1e-5, 1)
+    learning_rate = trial.suggest_float("learning_rate", 1e-5, 1, log=True)
     delta_std = trial.suggest_categorical("delta_std", [0.01, 0.02, 0.025, 0.03, 0.05, 0.1, 0.2, 0.3])
     top_frac_size = trial.suggest_categorical("top_frac_size", [0.1, 0.2, 0.3, 0.5, 0.8, 0.9, 1.0])
     zero_policy = trial.suggest_categorical("zero_policy", [True, False])