Reinforce-PongPolGrad-20000 training episodes

README.md

@@ -1,6 +1,6 @@
 ---
 tags:
-- CartPole-v1
+- Pong-PLE-v0
 - reinforce
 - reinforcement-learning
 - custom-implementation
@@ -12,11 +12,11 @@ model-index:
       type: reinforcement-learning
       name: reinforcement-learning
     dataset:
-      name: CartPole-v1
-      type: CartPole-v1
+      name: Pong-PLE-v0
+      type: Pong-PLE-v0
     metrics:
     - type: mean_reward
-      value: 9.46 +/- 0.75
+      value: -16.00 +/- 0.00
       name: mean_reward
       verified: false
 ---

hyperparameters.json

@@ -1 +1 @@
-{"h_size": 64, "n_training_episodes": 2200, "n_evaluation_episodes": 13, "max_t": 5000, "gamma": 0.96, "lr": 0.1, "env_id": "CartPole-v1", "state_space": 4, "action_space": 2}
+{"h_size": 64, "n_training_episodes": 20000, "n_evaluation_episodes": 13, "max_t": 7000, "gamma": 0.96, "lr": 0.1, "env_id": "Pong-PLE-v0", "state_space": 7, "action_space": 3}

model.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2fd52b3f849488df92070f4ccc83e1fdce79fd015dc50413ca406876558efd2e
-size 3935
+oid sha256:89e608cb4679a26c46c5c82a51092abcdea5b550ac7b099f88cffa4067af6be4
+size 39511

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:938bd1c3e2735fe0683024a311a264ce01cdedf14a0759bcee16c7938b77578c
-size 3390
+oid sha256:40c078913fa4b7822d9072fe39f134dbd3b64efd1af8ba8248b8f49be7a522ba
+size 13911

results.json

@@ -1 +1 @@
-{"env_id": "CartPole-v1", "mean_reward": 9.461538461538462, "n_evaluation_episodes": 13, "eval_datetime": "2022-08-28T17:37:10.586053"}
+{"env_id": "Pong-PLE-v0", "mean_reward": -16.0, "n_evaluation_episodes": 13, "eval_datetime": "2022-09-02T16:11:18.862127"}