LearningToCompareTF

implementation of the learning to compare article using tensorflow

To start the omniglot experiment run

omniglot_test.py

omniglot dataset will be downloaded in the omniglot directory while results will be stored in pickle files in the results/omniglot/ directory.

check the bottom of the file to see and modify the arguments that you can pass

Results

Learning curves for a model similar to NaiveRN for the omniglot dataset augmented as in the paper:

Name		Name	Last commit message	Last commit date
Latest commit History 27 Commits
plots		plots
README.md		README.md
datasets.py		datasets.py
meta_batch_iterator.py		meta_batch_iterator.py
models.py		models.py
omniglot_test.py		omniglot_test.py
plot_results.py		plot_results.py
utils.py		utils.py