Exercise:Vectorization
From Ufldl
| Line 7: | Line 7: | ||
=== MNIST === | === MNIST === | ||
| - | sparsityParam = 0.1 | + | Use the following parameters for the MNIST dataset: |
| - | lambda = 3e-3 | + | |
| - | beta = 3 | + | sparsityParam = 0.1 |
| - | normalizeData = linear | + | lambda = 3e-3 |
| + | beta = 3 | ||
| + | normalizeData = linear | ||
The autoencoder should learn pen strokes as features. These features should start to become obvious after 400 iterations of minFunc, which takes around 20 - 25 minutes on the Corn cluster. Visualised, the features look as follows: | The autoencoder should learn pen strokes as features. These features should start to become obvious after 400 iterations of minFunc, which takes around 20 - 25 minutes on the Corn cluster. Visualised, the features look as follows: | ||
| Line 19: | Line 21: | ||
<table> | <table> | ||
| - | <tr><td>[[File:MNIST-false- | + | <tr><td>[[File:MNIST-false-bad-1.png|240px]]</td></tr> |
</table> | </table> | ||
