Running on your own data

hoangthang1607 / nfnets-Tensorflow-2

Pre-trained NFNets with 99% of the accuracy of the official paper "High-Performance Large-Scale Image Recognition Without Normalization".

https://arxiv.org/abs/2102.06171

MIT License

30 stars 12 forks source link

Running on your own data #2

Open Sicily-F opened 3 years ago

Sicily-F commented 3 years ago

Hi there,

I haven't tested this repo yet, but I presume it would be something along the lines of just respecifying the number of classes and images, as well as your directory to images? I've not seen the model up on Tensorflow Hub yet, any ideas would be great!

abhay-7 commented 3 years ago

I tried running this on my own data, but somehow it is just predicting a single class as the output. Any tips to debug this issue? (I have already tried playing around with the activation functions in the final and intermediate layers, and I am using one other dense layer over and above the nfnet model defined in nfnet.py)