Save requesting image to local and return a resultant image with bounding box?

🚀 The feature

By requesting curl http://192.168.1.100:8080/predictions/alexnet -T ./serve/examples/image_classifier/kitten.jpg,

Wondering if there is an option to:

• save the uploaded kitten.jpg to a local folder on http://192.168.1.100 ??
• return an image with bounding box to the requesting web browser ?

Motivation, pitch

As described in The feature, the motivation is quite obvious.

Alternatives

N/A

Additional context

N/A

[agunapal]

@TillmannRheude The following code worked for me with the AlexNet example. I modified the inference code to return a tuple of the inference result and input image We can unpack the data in the postprocess method and do the necessary post processing operation on the image and save it

` def inference(self, data, *args, **kwargs): marshalled_data = data.to(self.device) with torch.no_grad(): results = self.model(marshalled_data, *args, **kwargs) return (results, data)

def postprocess(self, data):
    data, inp_img = data[0], data[1][0]
    
    ps = F.softmax(data, dim=1)
    probs, classes = torch.topk(ps, self.topk, dim=1)
    probs = probs.tolist()
    classes = classes.tolist()

    # post process the image
    op_image = inp_img

    # save image to file
    save_image(op_image, "/home/ubuntu/op_img1.png")
    
    return map_class_to_label(probs, self.mapping, classes)

`

Please let us know if this answers your question

[TillmannRheude]

Thank ya @agunapal, it worked like a charm - My fault was not to define “/home/…” in front of the path for the image.