'GroundingDINO' object has no attribute 'predict_with_caption'

[andysingal]

while working on GroundingDINO with SAM here is the notebook: https://colab.research.google.com/drive/1uObAs_PI6caKHhwZRN3mQSzkc3zxrkeR?usp=sharing

detections, phrases = model.predict_with_caption(
    image=image_bgr,
    caption=TEXT_PROMPT,
    box_threshold=BOX_THRESHOLD,
    text_threshold=TEXT_THRESHOLD,
)

giving error:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
[<ipython-input-52-a4301172c7a0>](https://localhost:8080/#) in <cell line: 1>()
----> 1 detections, phrases = model.predict_with_caption(
      2     image=image_bgr,
      3     caption=TEXT_PROMPT,
      4     box_threshold=BOX_THRESHOLD,
      5     text_threshold=TEXT_THRESHOLD,

[/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py](https://localhost:8080/#) in __getattr__(self, name)
   1612             if name in modules:
   1613                 return modules[name]
-> 1614         raise AttributeError("'{}' object has no attribute '{}'".format(
   1615             type(self).__name__, name))
   1616 

AttributeError: 'GroundingDINO' object has no attribute 'predict_with_caption'

Looking forward to hearing from you. Thanks

[Rivoks]

I thinks it's because you didn't initialized the model Class. This should be working:

from groundingdino.util.inference import Model

GROUNDING_DINO_CONFIG_PATH = "../groundingdino/config/GroundingDINO_SwinT_OGC.py"
GROUNDING_DINO_CHECKPOINT_PATH = "../weights/groundingdino_swint_ogc.pth"

grounding_dino_model = Model(model_config_path=GROUNDING_DINO_CONFIG_PATH, model_checkpoint_path=GROUNDING_DINO_CHECKPOINT_PATH)

detections, phrases = grounding_dino_model.predict_with_caption(
        image=image_np,
        caption=CAPTION,
        box_threshold=BOX_THRESHOLD,
        text_threshold=TEXT_THRESHOLD
    )