Mercurial > ift6266
annotate baseline/conv_mlp/convolutional_mlp.conf @ 185:b9ea8e2d071a
Enlevé ce qui concernait la réutilisation de résultats de préentraînement (trop compliqué pour peu de bénéfice: c'est le finetuning qui est vraiment long
author | fsavard |
---|---|
date | Fri, 26 Feb 2010 17:45:52 -0500 |
parents | d37c944133c3 |
children |
rev | line source |
---|---|
146
33038ab4e799
Reseau a convolution
Jeremy Eustache <jeremy.eustache@voila.fr>
parents:
diff
changeset
|
1 learning_rate=0.01 |
33038ab4e799
Reseau a convolution
Jeremy Eustache <jeremy.eustache@voila.fr>
parents:
diff
changeset
|
2 n_iter=1 |
33038ab4e799
Reseau a convolution
Jeremy Eustache <jeremy.eustache@voila.fr>
parents:
diff
changeset
|
3 batch_size=20 |
33038ab4e799
Reseau a convolution
Jeremy Eustache <jeremy.eustache@voila.fr>
parents:
diff
changeset
|
4 n_kern0=20 |
33038ab4e799
Reseau a convolution
Jeremy Eustache <jeremy.eustache@voila.fr>
parents:
diff
changeset
|
5 n_kern1=50 |
33038ab4e799
Reseau a convolution
Jeremy Eustache <jeremy.eustache@voila.fr>
parents:
diff
changeset
|
6 filter_shape=5 |
33038ab4e799
Reseau a convolution
Jeremy Eustache <jeremy.eustache@voila.fr>
parents:
diff
changeset
|
7 n_layer=3 |