Code for the Memory Module as described in "Learning to Remember Rare Events" by Lukasz Kaiser, Ofir Nachum, Aurko Roy, and Samy Bengio published as a conference paper at ICLR 2017.

Requirements: * TensorFlow (see tensorflow.org for how to install) * Some basic command-line utilities (git, unzip).

Description:

The general memory module is located in memory.py. Some code is provided to see the memory module in action on the standard Omniglot dataset. Download and setup the dataset using data_utils.py and then run the training script train.py (see example commands below).

Note that the structure and parameters of the model are optimized for the data preparation as provided.

Quick Start:

First download and set-up Omniglot data by running

python data_utils.py

Then run the training script:

python train.py --memory_size=8192 
  --batch_size=16 --validation_length=50 
  --episode_width=5 --episode_length=30

The first validation batch may look like this (although it is noisy):

0-shot: 0.040, 1-shot: 0.404, 2-shot: 0.516, 3-shot: 0.604,
  4-shot: 0.656, 5-shot: 0.684

At step 500 you may see something like this:

0-shot: 0.036, 1-shot: 0.836, 2-shot: 0.900, 3-shot: 0.940,
  4-shot: 0.944, 5-shot: 0.916

At step 4000 you may see something like this:

0-shot: 0.044, 1-shot: 0.960, 2-shot: 1.000, 3-shot: 0.988,
  4-shot: 0.972, 5-shot: 0.992

Maintained by Ofir Nachum (ofirnachum) and Lukasz Kaiser (lukaszkaiser).

链接：https://github.com/tensorflow/models/tree/master/research/learning_to_remember_rare_events