Change dataset path

Author: rdspring1

README.md

@@ -8,6 +8,6 @@ Run data_utils.py to download and pre-process Omniglot dataset
 2. [Learning to Remember Rare Events](https://arxiv.org/abs/1703.03129)
 
 # Other References
-1. [One-shot Learning with Memory-Augmented Neural Networks](https://arxiv.org/pdf/1605.06065.pdf)
-2. [Matching Networks for One Shot Learning](https://arxiv.org/pdf/1606.04080.pdf)
+1. [One-shot Learning with Memory-Augmented Neural Networks](https://arxiv.org/abs/1605.06065.pdf)
+2. [Matching Networks for One Shot Learning](https://arxiv.org/abs/1606.04080.pdf)
   * [Andrej Karpathy's Notes - Matching Networks](https://github.com/karpathy/paper-notes/blob/master/matching_networks.md)

cnn.py

@@ -50,7 +50,7 @@ def forward(self, x, predict):
 episode_length = 30
 episode_width = 5
 validation_frequency = 20
-DATA_FILE_FORMAT = os.path.join(os.getcwd(), 'datasets/omniglot', '%s_omni.pkl')
+DATA_FILE_FORMAT = os.path.join(os.getcwd(), '%s_omni.pkl')
 
 train_filepath = DATA_FILE_FORMAT % 'train'
 trainset = omniglot.OmniglotDataset(train_filepath)