haebeom-lee / metadrop Goto Github PK

Hi, Lee! Gladly to read your inspiring paper. I met a little problem in understanding the sentence in your paper that
"Lastly, to compute the gradient of Eq. 4 w.r.t. φ, we must compute second-order derivative, otherwise the
gradient w.r.t. φ will always be zero. " in 3.1 section, the paragraph under equation (4).
May I ask why the gradient will be zero? I think the Eq.4 describes a function of φ and I find the algorithm in appendix doesn't use second order derivative to update φ.
Thanks ahead!

Hi, thank you for sharing your code.

In the conv_block() function in layers.py, tf.contrib.layers.batch_norm() is used without the is_training flag modification.

Since the flag controls to update the running means, the normalization layers collect query data information during the meta test stage.

Could you explain about the batch normalization layer in your code? Thank you.

I am doing training based on your code and match the results mentioned in your paper. However, when I tried to train on mimgnet_5shot.sh file it gives an ambiguous error without any reason. I checked all the parameters and path, everything is fine.

Command
./runfiles/metadrop_mimg_5shot.sh

Traceback (most recent call last):
  File "main.py", line 227, in <module>
    meta_test()
  File "main.py", line 167, in meta_test
    saver.restore(sess, os.path.join(args.savedir, 'model'))
  File "/home/khawar/anaconda3/envs/py35/lib/python3.5/site-packages/tensorflow/python/training/saver.py", line 1538, in restore
    + compat.as_text(save_path))
ValueError: The passed save_path is not a valid checkpoint: ./results/metadrop/mimgnet_5shot/model