Error when training starts. ZeroDivisionError: integer division or modulo by zero

[mpwsh]

I know you pretty much abandoned this project, but i'm trying to make it work with tensorflow 0.12.1 and i'm getting this when the training "starts" (actually it freezes at 0% and then shows this error)

Model creation...
WARNING:tensorflow:From /media/sata/MusicGenerator/deepmusic/model.py:246 in _build_network.: scalar_summary (from tensorflow.python.ops.logging_ops) is deprecated and will be removed after 2016-11-30.
Instructions for updating:
Please switch to tf.summary.scalar. Note that tf.summary.scalar uses the node name instead of the tag. This means that TensorFlow will automatically de-duplicate summary names based on the scope they are created in. Also, passing a tensor or list of tags to a scalar summary op is no longer supported.
E tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:00:00.0/numa_node
Your kernel may have been built without NUMA support.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:885] Found device 0 with properties: 
name: NVIDIA Tegra X1
major: 5 minor: 3 memoryClockRate (GHz) 0.9984
pciBusID 0000:00:00.0
Total memory: 3.89GiB
Free memory: 2.85GiB
I tensorflow/core/common_runtime/gpu/gpu_device.cc:906] DMA: 0 
I tensorflow/core/common_runtime/gpu/gpu_device.cc:916] 0:   Y 
I tensorflow/core/common_runtime/gpu/gpu_device.cc:975] Creating TensorFlow device (/gpu:0) -> (device: 0, name: NVIDIA Tegra X1, pci bus id: 0000:00:00.0)
E tensorflow/core/common_runtime/gpu/gpu_device.cc:586] Could not identify NUMA node of /job:localhost/replica:0/task:0/gpu:0, defaulting to 0.  Your kernel may not have been built with NUMA support.
Initialize variables...
WARNING: No previous model found, but some files/folders found at /media/sata/MusicGenerator/save/model. Cleaning...
Removing /media/sata/MusicGenerator/save/model/train/events.out.tfevents.1511396622.tegra-ubuntu
Start training (press Ctrl+C to save and exit)...

------- Epoch 1 (lr=0.0001) -------
Subsampling the songs (train)...
Shuffling the dataset...
Generating batches...
Subsampling the songs (test)...
Shuffling the dataset...
Generating batches...
Training:   0%|                                                                                                | 0/2 [00:00<?, ?it/s]Traceback (most recent call last):
  File "main.py", line 29, in <module>
    composer.main()
  File "/media/sata/MusicGenerator/deepmusic/composer.py", line 197, in main
    self._main_train()
  File "/media/sata/MusicGenerator/deepmusic/composer.py", line 255, in _main_train
    next_batch_test = batches_test[self.glob_step % len(batches_test)]  # Generate test batches in a cycling way (test set smaller than train set)
ZeroDivisionError: integer division or modulo by zero

Any ideas?
Does this relates to the tags warning?

[BLACKMogus]

File "/media/sata/MusicGenerator/deepmusic/composer.py", line 255, in _main_train
You should open this script composer.py and find this line :
next_batch_test = batches_test[self.glob_step % len(batches_test)]
the parameter len(batches_test) here is zero because the batches_test is empty.
the batches_test is got by this code,in the same script:
batches_train, batches_test = self.music_data.get_batches()
Now is the solution:
Open the batchbuilder.py script and find :
def get_list(self, dataset, name):
""" See parent class for more details
Args:
dataset (list[Song]): the training/testing set
name (str): indicate the dataset type
Return:
list[Batch]: the batches to process
"""
focus on the code
for i in range((nb_samples//self.args.batch_size)):
yield extracts[i*self.args.batch_size:(i+1)*self.args.batch_size]
the main reason is that nb_samples is too small to divided by self.args.batch_size（equal 0）
so this code don't work，you should input some long songs
What I input are the Bach Piano Songs ，which I download from Internet