Initial commit

README.md

@@ -16,7 +16,7 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 14.50 +/- 12.34
+      value: 557.00 +/- 33.03
       name: mean_reward
       verified: false
 ---
@@ -70,10 +70,10 @@ OrderedDict([('batch_size', 32),
              ('exploration_fraction', 0.1),
              ('frame_stack', 4),
              ('gradient_steps', 1),
-             ('learning_rate', 1e-05),
-             ('learning_starts', 10000),
-             ('n_timesteps', 10000.0),
-             ('optimize_memory_usage', False),
+             ('learning_rate', 0.0001),
+             ('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
-    - null
-  - - device
-    - auto
   - - env
     - SpaceInvadersNoFrameskip-v4
   - - env_kwargs
     - null
   - - eval_episodes
-    - 5
+    - 10
   - - eval_freq
-    - 25000
+    - 10000
   - - gym_packages
     - []
   - - hyperparams
     - null
   - - log_folder
-    - llogs/
+    - 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
@@ -54,30 +40,20 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 180516644
+    - 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
-    - []
-  - - yaml_file
-    - null

config.yml

@@ -14,13 +14,13 @@
   - - gradient_steps
     - 1
   - - learning_rate
-    - 1.0e-05
+    - 0.0001
   - - learning_starts
-    - 10000
+    - 100000
   - - n_timesteps
-    - 10000.0
+    - 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:b611c51d23374350d42e3756b769ce5d1dc5dae146aa42837be00d5ecce6341b
-size 13716269
+oid sha256:f235f4a0baa9671bb03fabd56019c304438f993bf4abadc6db10fb2cd587d62b
+size 27223341

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab4765f5084aeb6b763bb2ec9930ea0fffd2e90f09aba83441f807538a349154
-size 687
+oid sha256:b0dabac2feca69093b74c171665ee8b882ca78866d034898359780b03e7d3898
+size 13505419

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca47a3078d221798b51b2a36f55140db47bc9c6e3be660d7fe62aa51ca5aafd5
-size 13504937
+oid sha256:486e3bb45269c467fcf022e1a4a19565cfccd15b6c482cb8d97915f646d04b4a
+size 13504745

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -2,6 +2,6 @@
 - Python: 3.8.10
 - Stable-Baselines3: 1.8.0a2
 - PyTorch: 1.13.1+cu116
-- GPU Enabled: True
+- GPU Enabled: False
 - Numpy: 1.21.6
 - Gym: 0.21.0

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fed103c47817f3f94ef4564da11c292e2891460f38d04aed7c8ce7eca6e1ea9
-size 275554
+oid sha256:639e90f73d519d41c88669a9b84f4b7937c867efce64cf10bb303043d5dd6efe
+size 216287

results.json

@@ -1 +1 @@
-{"mean_reward": 14.5, "std_reward": 12.338962679253067, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-02-16T17:58:23.109934"}
+{"mean_reward": 557.0, "std_reward": 33.03028912982749, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-02-16T19:53:00.090983"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26706af863bf3df5dc71fd3aa1db8e2ee7fd7c536554fd64ca2b17a1e46003ba
-size 541
+oid sha256:96201f3fe6e4d104176606933043a79f3ad33c827e92b2a568d52dc505d91a3b
+size 446800