Error in data_layer function "_get_next_minibatch"

[Tushn]

Hello,

I was trying to run this code with this dataset: http://vis-www.cs.umass.edu/lfw/
I needed to fix some problems in code, but has one I could not fix, when I run "train.py" I got this error below:

  File "/home/user/projects/triplet/data_layer.py", line 36, in _get_next_minibatch
    print('lenAttributeError: 'DataLayer' object has no attribute '_data'

(self._data._sample): '+str(len(self._data._sample)))

I can't see where this error occur, because my code execute that function "_shuffle_data" the same object and printed "self._data" without problems, when I was trying to execute "sw.train_model(max_iters)" (in "train.py") this error to occur.

[ly0303521]

Same problem with you , have you solved??

[ly0303521]

@Tushn

[ly0303521]

@Tushn comment test in solver

[Tushn]

@ly0303521 I forked this project few months ago, I could used that, but I have cant remember what I did. After, I was rewrote that layers in my studies with Caffe, unfortunately some codes are not public and I need review how to do.

I recommend that you download it, in: https://github.com/Tushn/triplet

I remember which I used HDF5 compressed file, well I'll try to review my code if you need.

[ly0303521]

comment test in solver.prototxt, I find the answer in the closed issues

[Tushn]

@ly0303521 did you solve your this problem, only commenting test in solver.prototxt?

Actually, that is one problem, I found my private code and could to fix in this net. But, that code is not too beautiful yet, I'm updating in my repository now.

Some problems which I got are:

1. data_layer.py: "self._data", I could solve this calling sample directly in my setup.py

    def setup(self, bottom, top):
        """Setup the DataLayer."""
        print(' --- SETUP --- ')
        self.data = sampledata()
        if cfg.TRIPLET_LOSS:
            self.batch_size = cfg.TRIPLET_BATCH_SIZE
        else:
            self.batch_size = cfg.BATCH_SIZE
        self._name_to_top_map = {
            'data': 0,
            'labels': 1}

        self._index = 0
        self._epoch = 1
        top[0].reshape(self.batch_size, 3, 30, 30)
        top[1].reshape(self.batch_size)

Note: I needed to resize face images

2. I changed "utils/blobs.py" for target_size = 30

3. I changed "sampledata.py" and its paths