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: 317.50 +/- 95.66
+      value: 862.50 +/- 470.60
       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).
 
@@ -43,37 +43,39 @@ pip install rl_zoo3
 
 ```
 # Download model and save it into the logs/ folder
-python -m rl_zoo3.load_from_hub --algo ppo --env SpaceInvadersNoFrameskip-v4 -orga arkadyark -f logs/
-python -m rl_zoo3.enjoy --algo ppo --env SpaceInvadersNoFrameskip-v4  -f logs/
+python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga arkadyark -f logs/
+python -m rl_zoo3.enjoy --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 arkadyark -f logs/
-python -m rl_zoo3.enjoy --algo ppo --env SpaceInvadersNoFrameskip-v4  -f logs/
+python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga arkadyark -f logs/
+python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4  -f logs/
 ```
 
 ## Training (with the RL Zoo)
 ```
-python -m rl_zoo3.train --algo ppo --env SpaceInvadersNoFrameskip-v4 -f logs/
+python -m rl_zoo3.train --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 arkadyark
+python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga arkadyark
 ```
 
 ## 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),
-             ('n_timesteps', 1000000.0),
+             ('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
@@ -18,7 +18,7 @@
   - - hyperparams
     - null
   - - log_folder
-    - logs/ppo
+    - logs/default-params
   - - log_interval
     - -1
   - - max_total_trials
@@ -54,7 +54,7 @@
   - - save_replay_buffer
     - false
   - - seed
-    - 2105563713
+    - 1747788574
   - - 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
-    - 1000000.0
+    - 10000000.0
+  - - optimize_memory_usage
+    - false
   - - policy
     - CnnPolicy
-  - - vf_coef
-    - 0.5
+  - - target_update_interval
+    - 1000
+  - - train_freq
+    - 4

dqn-SpaceInvadersNoFrameskip-v4.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95ff457396e98ae304d554d2fba40fa9a3bc2f3ee5f2e25d37db98537503adec
+size 27225173

dqn-SpaceInvadersNoFrameskip-v4/_stable_baselines3_version

@@ -0,0 +1 @@
+1.8.0

dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f21656725dd262199a7069a812cde4fa22a7cfe581fe6f27c23ae88f9ee1cd6
+size 13505739

dqn-SpaceInvadersNoFrameskip-v4/policy.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:479056d2a916843d996ef877d29a35c7682b87072e1f69c024d3ba0b0358e02b
+size 13504937

dqn-SpaceInvadersNoFrameskip-v4/pytorch_variables.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d030ad8db708280fcae77d87e973102039acd23a11bdecc3db8eb6c0ac940ee1
+size 431

dqn-SpaceInvadersNoFrameskip-v4/system_info.txt

@@ -0,0 +1,7 @@
+- OS: Linux-5.19.0-38-generic-x86_64-with-glibc2.35 # 39~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Fri Mar 17 21:16:15 UTC 2
+- Python: 3.9.16
+- Stable-Baselines3: 1.8.0
+- PyTorch: 2.0.0+cu117
+- GPU Enabled: True
+- Numpy: 1.24.2
+- Gym: 0.21.0

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89559496e67928cc4a33b6064356e46b847843893492586495e482716ef52c70
-size 253512
+oid sha256:a8d32bc5208bb2a67c0d27c497e76f0fc82ba8c6551cc30059911ab7fd336d96
+size 182038

results.json

@@ -1 +1 @@
-{"mean_reward": 317.5, "std_reward": 95.66216598007804, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-04-08T19:56:33.399274"}
+{"mean_reward": 862.5, "std_reward": 470.59669569600675, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-04-09T21:52:11.322268"}

train_eval_metrics.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c588d7177559deafbe98349c748cefb63721112c6b195060997d5634b17123a
-size 39717
+oid sha256:aaaff6d10618715c72c73c32d8c0b427437fde16470b12d14d8719b34b001b4b
+size 268212