Initial commit

README.md

@@ -6,7 +6,7 @@ tags:
 - reinforcement-learning
 - stable-baselines3
 model-index:
-- name: PPO
+- name: DQN
   results:
   - task:
       type: reinforcement-learning
@@ -16,13 +16,13 @@ model-index:
       type: SpaceInvadersNoFrameskip-v4
     metrics:
     - type: mean_reward
-      value: 808.50 +/- 361.36
+      value: 892.50 +/- 340.74
       name: mean_reward
       verified: false
 ---
 
-# **PPO** Agent playing **SpaceInvadersNoFrameskip-v4**
-This is a trained model of a **PPO** agent playing **SpaceInvadersNoFrameskip-v4**
+# **DQN** Agent playing **SpaceInvadersNoFrameskip-v4**
+This is a trained model of a **DQN** agent playing **SpaceInvadersNoFrameskip-v4**
 using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3)
 and the [RL Zoo](https://github.com/DLR-RM/rl-baselines3-zoo).
 
@@ -38,37 +38,39 @@ SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib
 
 ```
 # Download model and save it into the logs/ folder
-python -m rl_zoo3.load_from_hub --algo ppo --env SpaceInvadersNoFrameskip-v4 -orga Roberto -f logs/
-python enjoy.py --algo ppo --env SpaceInvadersNoFrameskip-v4  -f logs/
+python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga Roberto -f logs/
+python enjoy.py --algo dqn --env SpaceInvadersNoFrameskip-v4  -f logs/
 ```
 
 If you installed the RL Zoo3 via pip (`pip install rl_zoo3`), from anywhere you can do:
 ```
-python -m rl_zoo3.load_from_hub --algo ppo --env SpaceInvadersNoFrameskip-v4 -orga Roberto -f logs/
-rl_zoo3 enjoy --algo ppo --env SpaceInvadersNoFrameskip-v4  -f logs/
+python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga Roberto -f logs/
+rl_zoo3 enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4  -f logs/
 ```
 
 ## Training (with the RL Zoo)
 ```
-python train.py --algo ppo --env SpaceInvadersNoFrameskip-v4 -f logs/
+python train.py --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
 # Upload the model and generate video (when possible)
-python -m rl_zoo3.push_to_hub --algo ppo --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga Roberto
+python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga Roberto
 ```
 
 ## Hyperparameters
 ```python
-OrderedDict([('batch_size', 256),
-             ('clip_range', 'lin_0.1'),
-             ('ent_coef', 0.01),
+OrderedDict([('batch_size', 32),
+             ('buffer_size', 100000),
              ('env_wrapper',
               ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
+             ('exploration_final_eps', 0.01),
+             ('exploration_fraction', 0.1),
              ('frame_stack', 4),
-             ('learning_rate', 'lin_2.5e-4'),
-             ('n_envs', 8),
-             ('n_epochs', 4),
-             ('n_steps', 128),
+             ('gradient_steps', 1),
+             ('learning_rate', 0.0001),
+             ('learning_starts', 100000),
              ('n_timesteps', 10000000.0),
+             ('optimize_memory_usage', False),
              ('policy', 'CnnPolicy'),
-             ('vf_coef', 0.5),
+             ('target_update_interval', 1000),
+             ('train_freq', 4),
              ('normalize', False)])
 ```

args.yml

@@ -1,6 +1,6 @@
 !!python/object/apply:collections.OrderedDict
 - - - algo
-    - ppo
+    - dqn
   - - conf_file
     - null
   - - device
@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 1410119825
+    - 1190980102
   - - storage
     - null
   - - study_name

config.yml

@@ -1,25 +1,29 @@
 !!python/object/apply:collections.OrderedDict
 - - - batch_size
-    - 256
-  - - clip_range
-    - lin_0.1
-  - - ent_coef
-    - 0.01
+    - 32
+  - - buffer_size
+    - 100000
   - - env_wrapper
     - - stable_baselines3.common.atari_wrappers.AtariWrapper
+  - - exploration_final_eps
+    - 0.01
+  - - exploration_fraction
+    - 0.1
   - - frame_stack
     - 4
+  - - gradient_steps
+    - 1
   - - learning_rate
-    - lin_2.5e-4
-  - - n_envs
-    - 8
-  - - n_epochs
-    - 4
-  - - n_steps
-    - 128
+    - 0.0001
+  - - learning_starts
+    - 100000
   - - n_timesteps
     - 10000000.0
+  - - optimize_memory_usage
+    - false
   - - policy
     - CnnPolicy
-  - - vf_coef
-    - 0.5
+  - - target_update_interval
+    - 1000
+  - - train_freq
+    - 4

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90891d61dc5e16f246814babd3eebe4c53e51ccfeed146e5b468e23b817fbc65
-size 13719727
+oid sha256:7799bb75a0957ba4816da88b2f04eaaff09bd2dab5407643b47e94229530b500
+size 27224904

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1e067afe9912f3dd1b7925918b8cbe439229f6008e572c9c7e431ae731419f1
-size 687
+oid sha256:ad841495214761f229b0f45b3914cbe83dea6bfa4a5f0f563197fcb8df95214f
+size 13505419

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f823934be8aca8ed4e0ef45a70013d70e4fe120c6c3b6c1d673577629bb07ad
-size 13504937
+oid sha256:fdeaf40f2bd414d341781ddaf39e1a4a94ce9e09dcb93d0e475f17c3d90385d3
+size 13504745

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -1,7 +1,7 @@
-OS: Linux-5.10.133+-x86_64-with-glibc2.27 #1 SMP Fri Aug 26 08:44:51 UTC 2022
-Python: 3.8.16
+OS: Linux-5.15.0-56-generic-x86_64-with-glibc2.35 #62-Ubuntu SMP Tue Nov 22 19:54:14 UTC 2022
+Python: 3.9.15
 Stable-Baselines3: 1.6.2
-PyTorch: 1.13.0+cu116
-GPU Enabled: True
-Numpy: 1.21.6
+PyTorch: 1.13.1+cu117
+GPU Enabled: False
+Numpy: 1.24.0
 Gym: 0.21.0

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85bfc6ed9433811788324a541ce8b129a920cb965bae7796fe02c1a772a65659
-size 203960
+oid sha256:40a9f1812bd15d2fc97d7f87743d9093e2f64c994cea01c5945212b96d119541
+size 195972

results.json

@@ -1 +1 @@
-{"mean_reward": 808.5, "std_reward": 361.35889362239305, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-21T09:29:07.019988"}
+{"mean_reward": 892.5, "std_reward": 340.74367198819704, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-24T17:16:22.255684"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:271a39b9269e2be53529507d07f422263036bf2d60154c534c274b85d139e3f2
-size 314162
+oid sha256:4b41fa0a29e21fab4b06e0f9a07bc5c3a2d3887a8976ec789a81a0dcf00730b2
+size 135924