Robotics
Transformers
Safetensors
Inference Endpoints
alexandersoare commited on
Commit
f7872bf
1 Parent(s): e6d30a8

Upload folder using huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +5 -3
README.md CHANGED
@@ -9,6 +9,8 @@ pipeline_tag: robotics
9
  Diffusion Policy (as per [Diffusion Policy: Visuomotor Policy
10
  Learning via Action Diffusion](https://arxiv.org/abs/2303.04137)) trained for the `PushT` environment from [gym-pusht](https://github.com/huggingface/gym-pusht) with keypoint-only observations.
11
 
 
 
12
  ## How to Get Started with the Model
13
 
14
  Use `python lerobot/scripts/eval.py -p lerobot/diffusion_pusht` to evaluate for 50 episodes with the outputs sent to `outputs/eval`.
@@ -17,9 +19,9 @@ For further information, please see the [LeRobot library](https://github.com/hug
17
 
18
  ## Training Details
19
 
20
- Trained with TODO commit hash
21
 
22
- The model was trained using [LeRobot's training script](https://github.com/huggingface/lerobot/blob/d747195c5733c4f68d4bfbe62632d6fc1b605712/lerobot/scripts/train.py) and with the [pusht_keypoints](https://huggingface.co/datasets/lerobot/pusht_keypoints/tree/v1.5) dataset, using this command:
23
 
24
  ```bash
25
  python lerobot/scripts/train.py \
@@ -55,7 +57,7 @@ The model was evaluated on the `PushT` environment from [gym-pusht](https://gith
55
  Here are the metrics for 500 episodes worth of evaluation.
56
 
57
  Metric|Average over 500 episodes
58
- -|-|-
59
  Average max. overlap ratio | 0.97
60
  Success rate (%) | 71.0
61
 
 
9
  Diffusion Policy (as per [Diffusion Policy: Visuomotor Policy
10
  Learning via Action Diffusion](https://arxiv.org/abs/2303.04137)) trained for the `PushT` environment from [gym-pusht](https://github.com/huggingface/gym-pusht) with keypoint-only observations.
11
 
12
+ Note: The original work trains keypoints-only with conditioning via inpainting. Here, we encode the observation along with the agent position and use the encoding as global conditioning for the denoising U-Net.
13
+
14
  ## How to Get Started with the Model
15
 
16
  Use `python lerobot/scripts/eval.py -p lerobot/diffusion_pusht` to evaluate for 50 episodes with the outputs sent to `outputs/eval`.
 
19
 
20
  ## Training Details
21
 
22
+ Trained with TODO commit hash one the PR is merged
23
 
24
+ The model was trained using [LeRobot's training script](TODO link with commit hash one the PR is merged) and with the [pusht_keypoints](https://huggingface.co/datasets/lerobot/pusht_keypoints/tree/v1.5) dataset, using this command:
25
 
26
  ```bash
27
  python lerobot/scripts/train.py \
 
57
  Here are the metrics for 500 episodes worth of evaluation.
58
 
59
  Metric|Average over 500 episodes
60
+ -|-
61
  Average max. overlap ratio | 0.97
62
  Success rate (%) | 71.0
63