Upload policy weights, train config and readme
Browse files- README.md +2 -2
- config.json +1 -9
- model.safetensors +2 -2
- train_config.json +3 -11
README.md
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
---
|
| 2 |
-
datasets: peyschen/
|
| 3 |
library_name: lerobot
|
| 4 |
license: apache-2.0
|
| 5 |
model_name: reward_classifier
|
| 6 |
pipeline_tag: robotics
|
| 7 |
tags:
|
| 8 |
-
- reward_classifier
|
| 9 |
- robotics
|
|
|
|
| 10 |
- lerobot
|
| 11 |
---
|
| 12 |
|
|
|
|
| 1 |
---
|
| 2 |
+
datasets: peyschen/hil_reward__no_wrist
|
| 3 |
library_name: lerobot
|
| 4 |
license: apache-2.0
|
| 5 |
model_name: reward_classifier
|
| 6 |
pipeline_tag: robotics
|
| 7 |
tags:
|
|
|
|
| 8 |
- robotics
|
| 9 |
+
- reward_classifier
|
| 10 |
- lerobot
|
| 11 |
---
|
| 12 |
|
config.json
CHANGED
|
@@ -2,14 +2,6 @@
|
|
| 2 |
"type": "reward_classifier",
|
| 3 |
"n_obs_steps": 1,
|
| 4 |
"input_features": {
|
| 5 |
-
"observation.images.wrist": {
|
| 6 |
-
"type": "VISUAL",
|
| 7 |
-
"shape": [
|
| 8 |
-
3,
|
| 9 |
-
128,
|
| 10 |
-
128
|
| 11 |
-
]
|
| 12 |
-
},
|
| 13 |
"observation.images.front": {
|
| 14 |
"type": "VISUAL",
|
| 15 |
"shape": [
|
|
@@ -44,7 +36,7 @@
|
|
| 44 |
"dropout_rate": 0.1,
|
| 45 |
"model_name": "helper2424/resnet10",
|
| 46 |
"model_type": "cnn",
|
| 47 |
-
"num_cameras":
|
| 48 |
"learning_rate": 0.0001,
|
| 49 |
"weight_decay": 0.01,
|
| 50 |
"grad_clip_norm": 1.0,
|
|
|
|
| 2 |
"type": "reward_classifier",
|
| 3 |
"n_obs_steps": 1,
|
| 4 |
"input_features": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"observation.images.front": {
|
| 6 |
"type": "VISUAL",
|
| 7 |
"shape": [
|
|
|
|
| 36 |
"dropout_rate": 0.1,
|
| 37 |
"model_name": "helper2424/resnet10",
|
| 38 |
"model_type": "cnn",
|
| 39 |
+
"num_cameras": 1,
|
| 40 |
"learning_rate": 0.0001,
|
| 41 |
"weight_decay": 0.01,
|
| 42 |
"grad_clip_norm": 1.0,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8347757c5b21eb43ebcb552c7ef08760a3c060e647fdaa12e0bae715cabc9143
|
| 3 |
+
size 24357828
|
train_config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"dataset": {
|
| 3 |
-
"repo_id": "peyschen/
|
| 4 |
"root": "/home/peyschen/rl_playground/datasets/PandaPickCubeGamepad-v0_reward",
|
| 5 |
"episodes": null,
|
| 6 |
"image_transforms": {
|
|
@@ -84,14 +84,6 @@
|
|
| 84 |
"type": "reward_classifier",
|
| 85 |
"n_obs_steps": 1,
|
| 86 |
"input_features": {
|
| 87 |
-
"observation.images.wrist": {
|
| 88 |
-
"type": "VISUAL",
|
| 89 |
-
"shape": [
|
| 90 |
-
3,
|
| 91 |
-
128,
|
| 92 |
-
128
|
| 93 |
-
]
|
| 94 |
-
},
|
| 95 |
"observation.images.front": {
|
| 96 |
"type": "VISUAL",
|
| 97 |
"shape": [
|
|
@@ -126,7 +118,7 @@
|
|
| 126 |
"dropout_rate": 0.1,
|
| 127 |
"model_name": "helper2424/resnet10",
|
| 128 |
"model_type": "cnn",
|
| 129 |
-
"num_cameras":
|
| 130 |
"learning_rate": 0.0001,
|
| 131 |
"weight_decay": 0.01,
|
| 132 |
"grad_clip_norm": 1.0,
|
|
@@ -134,7 +126,7 @@
|
|
| 134 |
"VISUAL": "MEAN_STD"
|
| 135 |
}
|
| 136 |
},
|
| 137 |
-
"output_dir": "outputs/train/2026-01-13/
|
| 138 |
"job_name": "reward-classifier",
|
| 139 |
"resume": false,
|
| 140 |
"seed": 2,
|
|
|
|
| 1 |
{
|
| 2 |
"dataset": {
|
| 3 |
+
"repo_id": "peyschen/hil_reward__no_wrist",
|
| 4 |
"root": "/home/peyschen/rl_playground/datasets/PandaPickCubeGamepad-v0_reward",
|
| 5 |
"episodes": null,
|
| 6 |
"image_transforms": {
|
|
|
|
| 84 |
"type": "reward_classifier",
|
| 85 |
"n_obs_steps": 1,
|
| 86 |
"input_features": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
"observation.images.front": {
|
| 88 |
"type": "VISUAL",
|
| 89 |
"shape": [
|
|
|
|
| 118 |
"dropout_rate": 0.1,
|
| 119 |
"model_name": "helper2424/resnet10",
|
| 120 |
"model_type": "cnn",
|
| 121 |
+
"num_cameras": 1,
|
| 122 |
"learning_rate": 0.0001,
|
| 123 |
"weight_decay": 0.01,
|
| 124 |
"grad_clip_norm": 1.0,
|
|
|
|
| 126 |
"VISUAL": "MEAN_STD"
|
| 127 |
}
|
| 128 |
},
|
| 129 |
+
"output_dir": "outputs/train/2026-01-13/17-55-12_reward-classifier",
|
| 130 |
"job_name": "reward-classifier",
|
| 131 |
"resume": false,
|
| 132 |
"seed": 2,
|