Commit ed6470ed authored by MAIRAL Julien's avatar MAIRAL Julien

update

parent 8c56f424
......@@ -48,8 +48,14 @@ You may download these files from [here](http://pascal.inrialpes.fr/data2/mairal
or [here](http://pascal.inrialpes.fr/data2/mairal/data/cifar10.mat) (for the raw dataset).
See also [here](https://www.cs.toronto.edu/~kriz/cifar.html) for the original dataset.
For the supervised variant, we feature a 14-layer architecture that achieves about 90.6% accuracy. For the unsupervised one, we provide a two-layer architecture. By changing the number of filters, the accuracy ranges from 77.2% accuracy (with 64 and 256 filters on the first and second layer respectively) to 86.2% (with 2048 and 32768 filters).
## Q: Why do the results seem to be slightly better than in the original paper?
There are several reasons for that:
--* This is a reimplementation which features different architectures than the original paper.
--* Whereas the results of the NIPS paper were obtained by optimizing parameters on a validation set, before reporting results on the test set.
[1]: https://hal.inria.fr/hal-01387399/document
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment