Passo 8: Inferenza su NVIDIA DGX Spark

Installare l'ambiente

• PyTorch

Installa PyTorch separatamente dal sito ufficiale, usando la versione con CUDA 13.0

pip3 install torch torchvision --index-url https://download.pytorch.org/whl/cu130

• Poi commenta torch da solo nel file pyproject.toml

Poi esegui pip install -e .

• Nota: nella riga di comando di inferenza, cambia il percorso policy.path con il percorso effettivo del modello all'interno dello Spark

Elimina il dataset esistente con prefisso eval (se presente)

sudo chmod 666 /dev/ttyACM*

sudo rm -rf /home/apx103/.cache/huggingface/lerobot/Tommymy/eval_lerobot_my_dataset_shake_hands

ACT

lerobot-record  \
  --robot.type=so101_follower \
  --robot.port=/dev/ttyACM0 \
  --robot.cameras="{ front: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 60, fourcc: "MJPG"}}" \
  --robot.id=my_follower_arm \
  --display_data=false \
  --dataset.repo_id=Tommymy/eval_lerobot_my_dataset_shake_hands \
  --dataset.single_task="Shake Hands" \
  --dataset.episode_time_s=1000 \
  --policy.path=/home/apx103/Downloads/lerobot_output/shake/ACT/5K/pretrained_model

SmolVLA

• Installare l'ambiente

conda activate base
conda create -y -n lerobot-smolvla python=3.10 -y
conda activate lerobot-smolvla
conda install ffmpeg=7.1.1 -c conda-forge -y

pip3 install torch torchvision --index-url https://download.pytorch.org/whl/cu130

cd lerobot
pip install -e ".[feetech,smolvla]"

• Inferenza

lerobot-record  \
  --robot.type=so101_follower \
  --robot.port=/dev/ttyACM0 \
  --robot.cameras="{ front: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 60, fourcc: "MJPG"}}" \
  --robot.id=my_follower_arm \
  --display_data=false \
  --dataset.repo_id=Tommymy/eval_lerobot_my_dataset_shake_hands \
  --dataset.single_task="Shake Hands" \
  --dataset.episode_time_s=1000 \
  --policy.path=/home/apx103/Downloads/lerobot_output/shake/smolvla/40K/pretrained_model

WALL-OSS

• Installare l'ambiente

cd lerobot
pip install -e ".[feetech,wallx]"

• Aggiungi il codice

lerobot-record  \
  --dataset.repo_id=Tommymy/eval_lerobot_my_dataset_shake_hands \
  --dataset.single_task="Shake Hands" \
  --policy.path=/home/apx103/Downloads/lerobot_output/shake/wallx/30K/pretrained_model \
  --dataset.push_to_hub=false \
  --robot.type=so101_follower \
  --robot.id=my_follower_arm \
  --robot.port=/dev/ttyACM0 \
  --robot.cameras="{front: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 60, fourcc: "MJPG"}}" \
  --display_data=false \
  --dataset.episode_time_s=1000

pi0

• Installare l'ambiente

conda create -y -n lerobot-pi python=3.10 -y
conda activate lerobot-pi
conda install ffmpeg=7.1.1 -c conda-forge -y

cd lerobot
pip3 install torch torchvision --index-url https://download.pytorch.org/whl/cu130
pip install -e ".[feetech,pi]"

• Inferenza

lerobot-record  \
  --robot.type=so101_follower \
  --robot.port=/dev/ttyACM0 \
  --robot.cameras="{ front: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 60, fourcc: "MJPG"}}" \
  --robot.id=my_follower_arm \
  --display_data=false \
  --dataset.repo_id=Tommymy/eval_lerobot_my_dataset_shake_hands \
  --dataset.single_task="Shake Hands" \
  --dataset.episode_time_s=1000 \
  --dataset.push_to_hub=false \
  --policy.path=/home/apx103/Downloads/lerobot_output/shake/pi0/50K/pretrained_model

Prodotti correlati

Kit sviluppatore SO-ARM101→