Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -21,7 +21,7 @@ model-index:
|
|
| 21 |
type: PongNoFrameskip-v4
|
| 22 |
metrics:
|
| 23 |
- type: mean_reward
|
| 24 |
-
value:
|
| 25 |
name: mean_reward
|
| 26 |
---
|
| 27 |
|
|
@@ -129,7 +129,7 @@ from huggingface_ding import push_model_to_hub
|
|
| 129 |
# Instantiate the agent
|
| 130 |
agent = EfficientZeroAgent(env_id="PongNoFrameskip-v4", exp_name="PongNoFrameskip-v4-EfficientZero")
|
| 131 |
# Train the agent
|
| 132 |
-
return_ = agent.train(step=int(
|
| 133 |
# Push model to huggingface hub
|
| 134 |
push_model_to_hub(
|
| 135 |
agent=agent.best,
|
|
@@ -289,7 +289,7 @@ exp_config = {
|
|
| 289 |
- **Demo:** [video](https://huggingface.co/OpenDILabCommunity/PongNoFrameskip-v4-EfficientZero/blob/main/replay.mp4)
|
| 290 |
<!-- Provide the size information for the model. -->
|
| 291 |
- **Parameters total size:** 33023.14 KB
|
| 292 |
-
- **Last Update Date:** 2024-01-
|
| 293 |
|
| 294 |
## Environments
|
| 295 |
<!-- Address questions around what environment the model is intended to be trained and deployed at, including the necessary information needed to be provided for future users. -->
|
|
|
|
| 21 |
type: PongNoFrameskip-v4
|
| 22 |
metrics:
|
| 23 |
- type: mean_reward
|
| 24 |
+
value: 20.4 +/- 0.66
|
| 25 |
name: mean_reward
|
| 26 |
---
|
| 27 |
|
|
|
|
| 129 |
# Instantiate the agent
|
| 130 |
agent = EfficientZeroAgent(env_id="PongNoFrameskip-v4", exp_name="PongNoFrameskip-v4-EfficientZero")
|
| 131 |
# Train the agent
|
| 132 |
+
return_ = agent.train(step=int(2000000))
|
| 133 |
# Push model to huggingface hub
|
| 134 |
push_model_to_hub(
|
| 135 |
agent=agent.best,
|
|
|
|
| 289 |
- **Demo:** [video](https://huggingface.co/OpenDILabCommunity/PongNoFrameskip-v4-EfficientZero/blob/main/replay.mp4)
|
| 290 |
<!-- Provide the size information for the model. -->
|
| 291 |
- **Parameters total size:** 33023.14 KB
|
| 292 |
+
- **Last Update Date:** 2024-01-16
|
| 293 |
|
| 294 |
## Environments
|
| 295 |
<!-- Address questions around what environment the model is intended to be trained and deployed at, including the necessary information needed to be provided for future users. -->
|