search

xinghaochen / SqueezeTime

Official PyTorch implementation of "No Time to Waste: Squeeze Time into Channel for Mobile Video Understanding"
Apache License 2.0
29 stars 1 forks source link

reproduction #4

Open Andyfever123 opened 1 month ago

Andyfever123 commented 1 month ago

Sorry, when I was testing the code with your trained weights, I found that there was something wrong with the model indication in the.py file in your project config. I wonder if you found it? Also, what if I want to visualize the results, like showing the recognition on a video? 1 2

Are there any differences between the models in the two places? If it's just for testing, do the two models have to be the same?

zyjwuyan commented 1 month ago

Thanks.The line 151 should be "load_from = None". We have corrected it in the config file. If you want to visualize the results, please see mmation2 offical demo for reference.