Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 91.00 +/- 71.02
+      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),
@@ -71,9 +71,9 @@ OrderedDict([('batch_size', 32),
              ('frame_stack', 4),
              ('gradient_steps', 1),
              ('learning_rate', 0.0001),
-             ('learning_starts', 10000),
-             ('n_timesteps', 200000.0),
-             ('optimize_memory_usage', False),
+             ('learning_starts', 100000),
+             ('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/
+    - 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
-    - 2546880517
+    - 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
@@ -16,11 +16,11 @@
   - - learning_rate
     - 0.0001
   - - learning_starts
-    - 10000
+    - 100000
   - - n_timesteps
-    - 200000.0
+    - 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 sha256:2dd8390fc453f8fefdcc5fc25e37d5d20f88bea65fc94cbfb09ef0d52af4a030
-size 27218487
+oid sha256:7e09b0a8b7b8a926e4521903da2723171a7d4842a1cd2893a30f1318e52417ba
+size 27217426

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9729dc7b46177b0ef496d770a708dc120d11666fd4aadc13f43fc3a0ddbd7d7
-size 13505739
+oid sha256:3777b09439e069151f9c8acf7e9ddda005d6bc55d44a2b61a0dd6212801df87a
+size 13505675

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3288158ad3cdece57f31ff57af91195fd1fb05a0e9d665e3b770fdfb31675037
+oid sha256:39c032b7cba764cdc8fc52598b46c5ca54e68b97ee6cc511259e5638e6d2e268
 size 13504937

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52ac6794b55e65d0104d43803a0e5b39b9a9431bff929b113ca161b65233df63
-size 228570
+oid sha256:4a32e739dbfba2a3972bd27c8f8227ef199c2e1e10e76867c94e12170b36eef5
+size 234854

results.json

@@ -1 +1 @@
-{"mean_reward": 91.0, "std_reward": 71.02112361825881, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-09-17T07:34:21.468175"}
+{"mean_reward": 587.0, "std_reward": 118.36807001890332, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-09-17T07:36:56.976385"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a7db3a57f8cfda8675b590f00e2d5b97ed7806513648c5e0305cc75a66e9207
-size 8964
+oid sha256:3939e7260c41e1bb533d712da224a20c40346c31c1ebc8399173e5821373e7b3
+size 446800