KichangKim / DeepDanbooru

AI based multi-label girl image classification system, implemented by using TensorFlow.
MIT License
2.63k stars 259 forks source link

run the py in spyder #39

Closed lofangmeng closed 3 years ago

lofangmeng commented 3 years ago

can i know that if i run this programe in spyder ,which py file is given us to run with come out those tag result

KichangKim commented 3 years ago

I don't know about spyder. But the entry file of DeepDanbooru is __main__.py https://github.com/KichangKim/DeepDanbooru/blob/ea22b374d1bf00b61c6c1438b58ff8da99569e54/deepdanbooru/__main__.py

lofangmeng commented 3 years ago

image how can i run this py ,it is error? please help me

KichangKim commented 3 years ago

If you want evaluate in your python code, see https://github.com/KichangKim/DeepDanbooru/blob/ea22b374d1bf00b61c6c1438b58ff8da99569e54/deepdanbooru/commands/evaluate_project.py

lofangmeng commented 3 years ago

THANKS I WILL TRY IT

lofangmeng commented 3 years ago

can i ask that what is the thresold,is this any files image

KichangKim commented 3 years ago

It is score threshold. DD console application shows tags that its score is higher than threshold. If you want to get all tags with score, set 0.

lofangmeng commented 3 years ago

image here is my path directory after run the py,yt goes wrong why image

lofangmeng commented 3 years ago

It is score threshold. DD console application shows tags that its score is higher than threshold. If you want to get all tags with score, set 0.

thank you

KichangKim commented 3 years ago

Did you download pre-trained model and extract to project folder? If not, you should make project data and train model yourself.

lofangmeng commented 3 years ago

image how can i solve this problem haha i always stuck

KichangKim commented 3 years ago

Sorry, evaluate_project.py is old code.

See https://github.com/KichangKim/DeepDanbooru/blob/master/deepdanbooru/commands/evaluate.py instead.

lofangmeng commented 3 years ago

ok

lofangmeng commented 3 years ago

image CAN I know that the (allow_gpu, compile_model, allow_folder, folder_filters, verbose),what is its parameter,how to call this function thank you

KichangKim commented 3 years ago

allow_gpu means that allow TensorFlow uses GPU for its training/evaluating. If you have CUDA compatible GPU, set it to True. compile_model is flag for https://www.tensorflow.org/api_docs/python/tf/keras/Model#compile

Also this method is simple python method, you can freely call this method from your python code. I highly recommend that read https://www.tensorflow.org/tutorials first before using DeepDanbooru with your custom code.

lofangmeng commented 3 years ago

thanks image i have a new problem now it seems ready to work but still stuck at here