Using Pre-trained Deep Learning Models on BigEarthNet-19
I could not find how to use pre-trained models in the bigearthnet-models-tf repo and I tried to do myself as following some web pages.
I downloaded K-BranchCNN pre-trained models and write the following codes.
with tf.Session() as sess:
saver = tf.train.import_meta_graph('K-BranchCNN/K-BranchCNN_BigEarthNet-19_labels.meta')
Although import meta_graph works, Can't load save_path when it is None error returns because of the checkpoint line.
Probably index and data files are not loaded as a checkpoint.
What should I do?