To use Tensorflow Unet in a project:

from tf_unet import unet, util, image_util

#preparing data loading
data_provider = image_util.ImageDataProvider("fishes/train/*.tif")

#setup & training
net = unet.Unet(layers=3, features_root=64, channels=1, n_class=2)
trainer = unet.Trainer(net)
path = trainer.train(data_provider, output_path, training_iters=32, epochs=100)


prediction = net.predict(path, data)

unet.error_rate(prediction, util.crop_to_shape(label, prediction.shape))

img = util.combine_img_prediction(data, label, prediction)
util.save_image(img, "prediction.jpg")

Keep track of the learning progress using Tensorboard. tf_unet automatically outputs relevant summaries.

Segmentation of a toy problem.

More examples can be found in the Jupyter notebooks for a toy problem or for a RFI problem. Further code is stored in the scripts folder.