search

yu4u / noise2noise

An unofficial and partial Keras implementation of "Noise2Noise: Learning Image Restoration without Clean Data"
MIT License
1.07k stars 237 forks source link

I want to use our own noisy data to train the model #45

Open kidd-f opened 4 years ago

kidd-f commented 4 years ago

Thanks for your sharing! I have a question about which part of the code I should modify if I have my own different noisy picture.Because the training code is adding noise to the clean image.I want to train the model using my own nosiy image including real nosie.Hope to get your answer, I will be very grateful.

yu4u commented 4 years ago

Sorry, the purpose of this repo is to reproduce "the concept" of Noise2Noise thus it requires non trivial modifications to train real data.

zhushuyun commented 4 years ago

Can you tell me how to modify ?

ghost commented 3 years ago

`!python3 "/content/noise2noise/train.py" \ --image_dir "/content/noise/target_noise" \ --image_noised_dir "/content/noise/source_noise" \ --test_dir "/content/noise/target_noise" \ --image_size 512 --batch_size 2 --lr 0.001 --model unet --target_noise_model clean --val_noise_model clean --output_path "/content/drive/MyDrive/noise2noise/clean/" \

--weight "/content/drive/MyDrive/noise2noise/clean/weights.006-22.673-36.41483.hdf5"`

noise2noise_real_data.zip

This is a quick change in the code, there may be bugs

Training result with 800 images image_26767 out