DELG: extract_features with GPU

Prerequisites

Please answer the following question for yourself before submitting an issue.

• I checked to make sure that this feature has not been requested already.

1. The entire URL of the file you are using

https://github.com/tensorflow/models/blob/master/research/delf/delf/python/delg/extract_features.py

2. Describe the feature you request

Support feature extraction using GPUs

3. Additional context

Thanks for your great work! When I use extract_features.py to extract delg on my images, it's very slow. I am wondering if we can use GPUs to make it faster.
I added the following code before the extractor_fn:

os.environ['CUDA_VISIBLE_DEVICES'] = '0'
extractor_fn = extractor.MakeExtractor(config)

However, when I use nvidia-smi command to check the GPU usage during feature extraction, it shows the GPU:0 has never been used and the speed is still very slow. Could you please help me with this?

4. Are you willing to contribute it? (Yes or No)