upload via upload_folder 2025-08-05T09:49:11.989523+00:00

README.md

@@ -5,9 +5,9 @@ tags:
 - td3
 - reinforcement-learning
 - custom-implementation
-- TD3
-- DDPG
-- Walker2d
+- policy-gradient
+- pytorch
+- ddpg
 model-index:
 - name: TD3-Walker2dV5
   results:
@@ -19,7 +19,7 @@ model-index:
       type: Walker2d-v5
     metrics:
     - type: mean_reward
-      value: 4348.90 +/- 73.31
+      value: 4348.91 +/- 73.32
       name: mean_reward
       verified: false
 ---
@@ -28,9 +28,22 @@ model-index:
 This is a trained model of a **TD3** agent playing **Walker2d-v5**.
 
 ## Usage
+### create the conda env in https://github.com/GeneHit/drl_practice
+```bash
+conda create -n drl python=3.10
+conda activate drl
+python -m pip install -r requirements.txt
+```
 
-model = load_from_hub(repo_id="winkin119/TD3-Walker2dV5", filename="td3_walker2d.pth")
-
+### play with full model
+```python
+# load the full model
+model = load_from_hub(repo_id="winkin119/TD3-Walker2dV5", filename="full_model.pt")
 
+# Create the environment. 
 env = gym.make("Walker2d-v5")
+state, _ = env.reset()
+action = model.action(state)
 ...
+```
+There is also a state dict version of the model, you can check the corresponding definition in the repo.

eval_result.json

@@ -1,6 +1,6 @@
 {
-    "mean_reward": 4348.90,
-    "std_reward": 73.31,
+    "mean_reward": 4348.906197770564,
+    "std_reward": 73.3169869523695,
     "datetime": "2025-07-25 20:06:04",
     "train_duration_min": "148.37"
 }

full_model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59e8b5b2f3226b3711abde1c3f1a525575f43414a7ab2639d745b68a6f325864
+size 522233

params.json

@@ -39,5 +39,6 @@
     },
     "max_action": 1.0,
     "tau": 0.05,
-    "max_grad_norm": 0.5
+    "max_grad_norm": 0.5,
+    "smooth_l1_loss_beta": null
 }

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f43c0349fdfbc426f0b2135f1589cd6680547390fd35246969efd7a909a6f443
-size 1261247
+oid sha256:e7078510567605d99b2179aeb8340b514dc3df0f6bb2620a3289b5488daeac3f
+size 931191

state_dict.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59118fa73c1658960e1406c927dfd1de9a200e6a4c34995623dfb0359688338a
+size 520377

tensorboard/events.out.tfevents.1753436180.winkindeMacBook-Air.local.66308.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8071dea26196af78543fc8aeee2338bfd189961194c2e99f4565c4d2770e0b7c
+size 159485697