fix(pu): fix test_serial_entry** by add resume_training=False in rela…

…ted config
opendilab · Nov 4, 2024 · f165c59 · f165c59
1 parent 22d5110
commit f165c59
Show file tree

Hide file tree

Showing 6 changed files with 6 additions and 0 deletions.
diff --git a/dizoo/classic_control/cartpole/config/cartpole_a2c_config.py b/dizoo/classic_control/cartpole/config/cartpole_a2c_config.py
@@ -21,6 +21,7 @@
             learning_rate=0.001,
             # (float) loss weight of the entropy regularization, the weight of policy network is set to 1
             entropy_weight=0.01,
+            resume_training=False,
         ),
         collect=dict(
             # (int) collect n_sample data, train model n_iteration times

diff --git a/dizoo/classic_control/cartpole/config/cartpole_pg_config.py b/dizoo/classic_control/cartpole/config/cartpole_pg_config.py
@@ -18,6 +18,7 @@
             batch_size=64,
             learning_rate=0.001,
             entropy_weight=0.001,
+            resume_training=False,
         ),
         collect=dict(n_episode=80, unroll_len=1, discount_factor=0.9),
         eval=dict(evaluator=dict(eval_freq=100, ), ),

diff --git a/dizoo/classic_control/cartpole/config/cartpole_ppo_stdim_config.py b/dizoo/classic_control/cartpole/config/cartpole_ppo_stdim_config.py
@@ -37,6 +37,7 @@
             entropy_weight=0.01,
             clip_ratio=0.2,
             learner=dict(hook=dict(save_ckpt_after_iter=100)),
+            resume_training=False,
         ),
         collect=dict(
             n_sample=256,

diff --git a/dizoo/classic_control/cartpole/config/cartpole_ppopg_config.py b/dizoo/classic_control/cartpole/config/cartpole_ppopg_config.py
@@ -19,6 +19,7 @@
             batch_size=64,
             learning_rate=0.001,
             entropy_weight=0.001,
+            resume_training=False,
         ),
         collect=dict(n_episode=80, unroll_len=1, discount_factor=0.9, collector=dict(get_train_sample=True)),
         eval=dict(evaluator=dict(eval_freq=100, ), ),

diff --git a/dizoo/classic_control/pendulum/config/pendulum_ppo_config.py b/dizoo/classic_control/pendulum/config/pendulum_ppo_config.py
@@ -35,6 +35,7 @@
             adv_norm=True,
             value_norm=True,
             ignore_done=True,
+            resume_training=False,
         ),
         collect=dict(
             n_sample=5000,

diff --git a/dizoo/petting_zoo/config/ptz_simple_spread_mappo_config.py b/dizoo/petting_zoo/config/ptz_simple_spread_mappo_config.py
@@ -53,6 +53,7 @@
             grad_clip_type='clip_norm',
             grad_clip_value=10,
             ignore_done=False,
+            resume_training=False,
         ),
         collect=dict(
             n_sample=3200,