Oleg Botnar

I've checked pytorch package in local/official package and figured out that this function download_url_to_file [exists](https://github.com/pytorch/pytorch/blob/cb90e36684b638fd4cb33b1cdf23c22a96fd9806/torch/hub.py#L370). This issue was fixed by modifying mmcv. Just add undersore symbol at the begging of...

Get it from first [release](https://github.com/Div99/Image-Captioning/releases/download/v0.1/model_weight.h5)

This issue can be fixed by changing [line at eval_model.py](https://github.com/Div99/Image-Captioning/blob/d7465d54e02ace69abc395e250176a1936f705b5/CapGenerator/eval_model.py#L22) to model = Model(inputs=model.inputs, outputs=model.layers[-2].output) We need feature vector from fully connected layer that goes as input to the last...

It looks like RKNN overrides ros logging. Try to launch detection in a separate process may help you.

@airockchip your repo is awesome. Can you publish sample from Baidu disk to Google Drive?

@zhaoqiangshuju I find links in the issue below(https://github.com/airockchip/rknn_model_zoo/issues/29#issuecomment-1480547015). There you can find zip archive.

> Okay after extracting the models/samples (and renaming `Models/` to `models/`) I get a couple steps further. Still I'm having errors. > > First I got some CUDA errors which...

Just update tensorflow via command ```pip install -U tensoflow-gpu ```

It's possible, but this task will be little bit harder than solved one.