Instructions to use NotALynx/smolvla-debug with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use NotALynx/smolvla-debug with LeRobot:
# See https://github.com/huggingface/lerobot?tab=readme-ov-file#installation for more details git clone https://github.com/huggingface/lerobot.git cd lerobot pip install -e .[smolvla]
# Launch finetuning on your dataset python lerobot/scripts/train.py \ --policy.path=NotALynx/smolvla-debug \ --dataset.repo_id=lerobot/svla_so101_pickplace \ --batch_size=64 \ --steps=20000 \ --output_dir=outputs/train/my_smolvla \ --job_name=my_smolvla_training \ --policy.device=cuda \ --wandb.enable=true
# Run the policy using the record function python -m lerobot.record \ --robot.type=so101_follower \ --robot.port=/dev/ttyACM0 \ # <- Use your port --robot.id=my_blue_follower_arm \ # <- Use your robot id --robot.cameras="{ front: {type: opencv, index_or_path: 8, width: 640, height: 480, fps: 30}}" \ # <- Use your cameras --dataset.single_task="Grasp a lego block and put it in the bin." \ # <- Use the same task description you used in your dataset recording --dataset.repo_id=HF_USER/dataset_name \ # <- This will be the dataset name on HF Hub --dataset.episode_time_s=50 \ --dataset.num_episodes=10 \ --policy.path=NotALynx/smolvla-debug - Notebooks
- Google Colab
- Kaggle
Upload policy weights, train config and readme
Browse files- README.md +1 -1
- config.json +1 -1
- train_config.json +2 -2
README.md
CHANGED
|
@@ -6,8 +6,8 @@ license: apache-2.0
|
|
| 6 |
model_name: smolvla
|
| 7 |
pipeline_tag: robotics
|
| 8 |
tags:
|
| 9 |
-
- lerobot
|
| 10 |
- robotics
|
|
|
|
| 11 |
- smolvla
|
| 12 |
---
|
| 13 |
|
|
|
|
| 6 |
model_name: smolvla
|
| 7 |
pipeline_tag: robotics
|
| 8 |
tags:
|
|
|
|
| 9 |
- robotics
|
| 10 |
+
- lerobot
|
| 11 |
- smolvla
|
| 12 |
---
|
| 13 |
|
config.json
CHANGED
|
@@ -60,7 +60,7 @@
|
|
| 60 |
"tokenizer_max_length": 48,
|
| 61 |
"num_steps": 10,
|
| 62 |
"use_cache": true,
|
| 63 |
-
"freeze_vision_encoder":
|
| 64 |
"train_expert_only": true,
|
| 65 |
"train_state_proj": true,
|
| 66 |
"optimizer_lr": 0.0001,
|
|
|
|
| 60 |
"tokenizer_max_length": 48,
|
| 61 |
"num_steps": 10,
|
| 62 |
"use_cache": true,
|
| 63 |
+
"freeze_vision_encoder": false,
|
| 64 |
"train_expert_only": true,
|
| 65 |
"train_state_proj": true,
|
| 66 |
"optimizer_lr": 0.0001,
|
train_config.json
CHANGED
|
@@ -142,7 +142,7 @@
|
|
| 142 |
"tokenizer_max_length": 48,
|
| 143 |
"num_steps": 10,
|
| 144 |
"use_cache": true,
|
| 145 |
-
"freeze_vision_encoder":
|
| 146 |
"train_expert_only": true,
|
| 147 |
"train_state_proj": true,
|
| 148 |
"optimizer_lr": 0.0001,
|
|
@@ -170,7 +170,7 @@
|
|
| 170 |
"max_period": 4.0,
|
| 171 |
"rtc_config": null
|
| 172 |
},
|
| 173 |
-
"output_dir": "outputs/train/2025-12-
|
| 174 |
"job_name": "smolvla",
|
| 175 |
"resume": false,
|
| 176 |
"seed": 1000,
|
|
|
|
| 142 |
"tokenizer_max_length": 48,
|
| 143 |
"num_steps": 10,
|
| 144 |
"use_cache": true,
|
| 145 |
+
"freeze_vision_encoder": false,
|
| 146 |
"train_expert_only": true,
|
| 147 |
"train_state_proj": true,
|
| 148 |
"optimizer_lr": 0.0001,
|
|
|
|
| 170 |
"max_period": 4.0,
|
| 171 |
"rtc_config": null
|
| 172 |
},
|
| 173 |
+
"output_dir": "outputs/train/2025-12-24/01-17-20_smolvla",
|
| 174 |
"job_name": "smolvla",
|
| 175 |
"resume": false,
|
| 176 |
"seed": 1000,
|