from grid.model.perception.vla.openvla import OpenVLA
car = AirGenCar()

# We will be capturing an image from the AirGen simulator 
# and run model inference on it.

img =  car.getImage("front_center", "rgb").data

model = OpenVLA(use_local = True)
result = model.run(rgbimage=img, prompt = "Close the drawer")
print(result)

The OpenVLA class provides core functionality for this module.

​
class OpenVLA()
​
use_local
boolean
default:
true

If True, inference call is run on the local VM, else offloaded onto GRID-Cortex. Defaults to True.

​
def run()
​
rgbimage
np.ndarray
required

The input RGB image of shape (M,N,3)(M,N,3).

​
prompt
str
required

Task instruction.

​
Returns
List[float]

Predicted action based on the query and image, represented as a 7-DoF vector.

from grid.model.perception.vla.openvla import OpenVLA
car = AirGenCar()

# We will be capturing an image from the AirGen simulator 
# and run model inference on it.

img =  car.getImage("front_center", "rgb").data

model = OpenVLA(use_local = True)
result = model.run(rgbimage=img, prompt = "Close the drawer")
print(result)

This code is licensed under the MIT License.

Was this page helpful?

Previous
LLaVANeXT
Next