Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 6.50 +/- 10.74
+      value: 587.00 +/- 118.37
       name: mean_reward
       verified: false
 ---
@@ -63,7 +63,7 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
 ## Hyperparameters
 ```python
 OrderedDict([('batch_size', 32),
-             ('buffer_size', 100000),
+             ('buffer_size', 10000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
              ('exploration_final_eps', 0.01),
@@ -72,8 +72,8 @@ OrderedDict([('batch_size', 32),
              ('gradient_steps', 1),
              ('learning_rate', 0.0001),
              ('learning_starts', 100000),
-             ('n_timesteps', 100.0),
-             ('optimize_memory_usage', False),
+             ('n_timesteps', 10000000.0),
+             ('optimize_memory_usage', True),
              ('policy', 'CnnPolicy'),
              ('target_update_interval', 1000),
              ('train_freq', 4),

args.yml

@@ -1,32 +1,24 @@
 !!python/object/apply:collections.OrderedDict
 - - - algo
     - dqn
-  - - conf_file
-    - dqn.yml
-  - - device
-    - auto
   - - env
     - SpaceInvadersNoFrameskip-v4
   - - env_kwargs
     - null
   - - eval_episodes
-    - 5
+    - 10
   - - eval_freq
-    - 25000
+    - 10000
   - - gym_packages
     - []
   - - hyperparams
     - null
   - - log_folder
-    - logs_3/
+    - rl-trained-agents/
   - - log_interval
     - -1
-  - - max_total_trials
-    - null
-  - - n_eval_envs
-    - 1
   - - n_evaluations
-    - null
+    - 20
   - - n_jobs
     - 1
   - - n_startup_trials
@@ -34,17 +26,11 @@
   - - n_timesteps
     - -1
   - - n_trials
-    - 500
-  - - no_optim_plots
-    - false
+    - 10
   - - num_threads
     - -1
-  - - optimization_log_path
-    - null
   - - optimize_hyperparameters
     - false
-  - - progress
-    - false
   - - pruner
     - median
   - - sampler
@@ -54,28 +40,20 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 1614370982
+    - 234163638
   - - storage
     - null
   - - study_name
     - null
   - - tensorboard_log
     - ''
-  - - track
-    - false
   - - trained_agent
     - ''
   - - truncate_last_trajectory
     - true
   - - uuid
-    - false
+    - true
   - - vec_env
     - dummy
   - - verbose
     - 1
-  - - wandb_entity
-    - null
-  - - wandb_project_name
-    - sb3
-  - - wandb_tags
-    - []

config.yml

@@ -2,7 +2,7 @@
 - - - batch_size
     - 32
   - - buffer_size
-    - 100000
+    - 10000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
   - - exploration_final_eps
@@ -18,9 +18,9 @@
   - - learning_starts
     - 100000
   - - n_timesteps
-    - 100.0
+    - 10000000.0
   - - optimize_memory_usage
-    - false
+    - true
   - - policy
     - CnnPolicy
   - - target_update_interval

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:843656b37ab9ce1d7784caa0b2170008996fba37ae6cd80b0125b928ea9d027d
-size 13710852
+oid sha256:e6f8722190b60112a040528eeb2b0416d65421aac371f7010397fe8c33343481
+size 27219001

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3fc84919e676e2bdcaa04814c7dfa2b7625d8d53a818418c6db25771364db93
-size 1120
+oid sha256:9101509075a968aff1597385716230e43838a5ee033f2283d56db6d478a843f1
+size 13506108

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6aedc054df526305a1dac1a3394b2f610bfc0947854ce90ded367a3314011e29
+oid sha256:49caa05efa450005dbf9dfb9798c8cbf0e826a937dc985ccbe496e339a219598
 size 13505370

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd6185268051f9d0226ba48767be781ec895e0fac98304a8831b0b2855bfcae2
-size 292667
+oid sha256:0fdf2840313331c8be0643895de920edf3f2f670e72b731e2e246bb708ca8cc3
+size 234832

results.json

@@ -1 +1 @@
-{"mean_reward": 6.5, "std_reward": 10.735455276791944, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-12-26T09:01:11.048285"}
+{"mean_reward": 587.0, "std_reward": 118.36807001890332, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-12-26T09:02:52.625286"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb10c0520a2541a2b37135797d573298eb0558d019aa342195c0a3a5a338a624
-size 205
+oid sha256:132f6cd4fee8951b0ff7929ffe54c7f641a5219311b6b6fd7ae5d7d0439e4eff
+size 446800