run my model with the object-detector-python

[MenahemOwlytics]

I am trying to run the object-detector-python example with my model to detect specific objects

There is a major difference between the models in terms of input size. The example MPDEL received input of 300x3003 while my model received input of 640x6403

I got the following error when I tried to run it:

Could not run job: Could only read 1228800 out of 4915200 bytes from file descriptor

How can I change the maximum number of bytes that can be read?

Thanks

[Corallo]

Hi @MenahemOwlytics
You are probably using a wrong data type, make sure to set the input as np.uint8 or np.int8 depending on the model you are using

[MenahemOwlytics]

@Corallo, thanks for the quick response

My model input is:
[{'name': 'inputs_0', 'index': 0, 'shape': array([ 1, 640, 640, 3], dtype=int32), 'shape_signature': array([ 1, 640, 640, 3], dtype=int32), 'dtype': <class 'numpy.float32'>, 'quantization': (0.0, 0), 'quantization_parameters': {'scales': array([], dtype=float32), 'zero_points': array([], dtype=int32), 'quantized_dimension': 0}, 'sparsity_parameters': {}}]

So i think the input type is not the problem

Nevertheless, I added:
image = np.array(image, dtype=np.unit32)
or image = np.array(image, dtype=np.unit32)

but it not success

this is the relevant code:

  def detect(self, image):
        image = np.expand_dims(image, axis=0)
        image = np.array(image, dtype=np.unit32)
        image = image.astype(np.uint8)
        success, result = self.inference_client.infer({'data': image}, self.model_path)

Once again, thank you

[Corallo]

I made a mistake before, the error is actually telling you that you are only providing one quarter of the amount of bytes needed by the model.
Try to remove the line:
image = image.astype(np.uint8)
However, I am not sure if this model can run on DLPU, it should work only with int8 models, but maybe it is still worth to try.