The correct script to load this dataset
#16
by
pbanavara
- opened
from lerobot.common.policies.pi0.modeling_pi0 import PI0Policy
import torch
# Load the pretrained policy from HF
policy = PI0Policy.from_pretrained("lerobot/pi0")
# Prepare a sample batch (obs, goal) for testing
batch = {
"image": torch.randn(1, 3, 224, 224), # Replace with real image
"goal": ["scrub the plate"]
}
# Forward pass
action = policy.select_action(batch)
print("Predicted joint actions:", action)
Note the uppercase 'I' in PIOPolicy , in the model card the case is Pi0Policy which is incorrect.
hello i am using the xArm environnement but i dont know what i have to pus in batch can you help me ?
File "C:\Users\SOUICI amir\lerobot\lerobot\common\policies\pi0\modeling_pi0.py", line 349, in prepare_images
raise ValueError(
ValueError: All image features are missing from the batch. At least one expected. (batch: dict_keys(['image', 'goal'])) (image_features:{})