diff --git a/README.md b/README.md index 0fac768b028766a06a3d21be1dfeadaa463b3005..77746a7ef828bcfd432b659212a8ec59f5331311 100644 --- a/README.md +++ b/README.md @@ -27,7 +27,7 @@ The MNIST database of handwritten digits, available from this [link](http://yann ## Open Binary MNIST notebook: -[](https://colab.research.google.com/drive/1zW0K7yNF0JIMRs7DUIGazZqNU5IkSppw) +[](https://colab.research.google.com/drive/1RVtgS6NOH5D5ssa19qDt35DYnjPxJCOX) ## Results on MNIST: @@ -67,7 +67,7 @@ The Omniglot data set is designed for developing more human-like learning algori ## Open Binary Omniglot notebook: -[](https://colab.research.google.com/drive/1CF5S1hTIczb4k2hzOyZr5x6ItUeTYrj3) +[](https://colab.research.google.com/drive/1z7ebfh8VWKfgkecpRzY7iiM6NT2ElWBL) ## Results on Omniglot classification with data train (80% train, 10% validation and 10% test): ### Loss/ACC: with 10 epochs.