anton/mlpractical
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
237 Commits 2 Branches 0 Tags 91 MiB
534c4744ca
Commit Graph

237 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Matt Graham
534c4744ca Fixing incorrect use of module relative class reference. 2016-11-02 03:39:13 +00:00
Matt Graham
891b942890 Merge branch 'mlp2016-7/master' of https://github.com/CSTR-Edinburgh/mlpractical into mlp2016-7/master 2016-11-02 02:55:37 +00:00
Matt Graham
a575efb0ee Adding instructions on running notebooks remotely. 2016-11-02 02:55:10 +00:00
Matt Graham
e0114b9aab Adding notebook with convolutional layer tests. 2016-11-02 01:14:27 +00:00
Matt Graham
da9ee58609 Adding reshape layer. 2016-11-02 01:13:22 +00:00
Matt Graham
d5ae8b2caf Merge pull request #43 from jfainberg/patch-1 
Minor typos in momentum docstring
 2016-10-29 21:17:24 +01:00
Joachim
bea3502210 Minor typos in momentum docstring 2016-10-29 15:57:00 +01:00
Matt Graham
a6f775b1ff Adding notebook for lab 6. 2016-10-27 21:43:35 +01:00
Matt Graham
1efe9de847 Adding code for lab 5 notebook answers. 2016-10-27 21:43:08 +01:00
Matt Graham
95d4028bb7 Adding stochastic layer functionality in preparation for dropout implementation. 2016-10-27 21:36:57 +01:00
Matt Graham
7e1250a82b Adding bash script to secure Jupyter notebook server on DICE. 2016-10-27 21:35:55 +01:00
Matt Graham
dfcce94242 Clarifying docstring in scheduler. 2016-10-27 11:24:00 +01:00
Matt Graham
ae999b43e4 Fixing spacing typo. 2016-10-21 22:07:41 +01:00
Matt Graham
9a182d4be1 Correcting regularisation coefficient values to test. 2016-10-21 22:07:25 +01:00
Matt Graham
cf6417f917 Removing old transfer function notebook. 2016-10-21 22:06:37 +01:00
Matt Graham
586c7477eb Removing old regularisation notebook. 2016-10-21 01:25:08 +01:00
Matt Graham
167508cc63 Adding new lab 5 notebook. 2016-10-21 01:24:37 +01:00
Matt Graham
ab060d556c Switching to parameter penalty rather than cost naming. 2016-10-21 01:24:04 +01:00
Matt Graham
a353526790 Adding augmented MNIST data provider. 2016-10-20 13:26:09 +01:00
Matt Graham
a021ce585b Applying softmax stability fix to cost gradient. 2016-10-20 13:23:19 +01:00
Matt Graham
b4b00e0c65 Using more stable softmax implementation by subtracting max. 2016-10-18 13:54:39 +01:00
Matt Graham
870c04f664 Modifying optmiser.train call in notebook to use new interface with run_time. 2016-10-18 13:45:28 +01:00
Matt Graham
0f496306c2 Fixing RadialBasisFunctionLayer __repr__ bug. 2016-10-18 13:45:05 +01:00
Matt Graham
237cff12db Making sure input_dims set converted to list. 2016-10-18 13:44:30 +01:00
Matt Graham
5d4c979ef3 Changing code blocks to use small p Python. 2016-10-17 23:36:21 +01:00
Matt Graham
f13ba91f82 Dealing with case where number of epochs less than stats interval. 2016-10-17 23:32:23 +01:00
Matt Graham
0cec9920f3 Adding coursework files to master branch. 2016-10-14 03:56:06 +01:00
Matt Graham
4863599bed Renaming old notebooks to reflect new numbering. 2016-10-14 03:40:43 +01:00
Matt Graham
0b9898d72c Adding new fourth lab notebook. 2016-10-14 03:39:19 +01:00
Matt Graham
43ad2a00ba Adding RBF layer definition for 4th lab. 2016-10-14 03:36:46 +01:00
Matt Graham
2e5a1a4c4f Renaming MNIST eval set to test set for consistency with slides. 2016-10-14 03:36:16 +01:00
Matt Graham
4c4cf2cf27 Bug fix: incorrect use of zip in MomentumLearningRule. 2016-10-12 23:11:38 +01:00
Matt Graham
a59d37dc69 Fixing softmax index typo. 2016-10-11 15:39:35 +01:00
Matt Graham
cafed216d7 Fixing typo in description of back-propagation. 2016-10-11 13:40:29 +01:00
Matt Graham
a6fcdc2894 Fixing typo in initialiser assert. 2016-10-10 10:09:29 +01:00
Matt Graham
cfad36c91f Removing final backpropagation of gradients to inputs as unnecessary. 2016-10-10 09:27:02 +01:00
Matt Graham
a24ff90f71 Adding Nesterov momentum learning rule. 2016-10-10 09:25:52 +01:00
Matt Graham
259b000cba Adding schedulers to optimiser. 2016-10-10 09:25:33 +01:00
Matt Graham
18f893d864 Adding Glorot and Bengio parameter initialiser. 2016-10-10 09:24:32 +01:00
Matt Graham
302d44371a Adding new schedulers module. 2016-10-10 09:23:53 +01:00
Matt Graham
fa2f6b195c Correcting docstring typo. 2016-10-08 22:19:49 +01:00
Matt Graham
1fd2b84353 Renaming third lab notebook. 2016-10-07 07:01:26 +01:00
Matt Graham
ee0c6ce1d5 Updating third lab notebook on multilayer models. 2016-10-07 07:00:47 +01:00
Matt Graham
b1a031deac Adding solutions for second lab notebook. 2016-10-07 07:00:10 +01:00
Matt Graham
9863c75b05 Adding back propagation block diagram figure and source. 2016-10-07 06:24:18 +01:00
Matt Graham
25263e1391 Including initial pre-training statistics. 2016-10-07 06:23:15 +01:00
Matt Graham
e05717e3d0 Adding SoftmaxLayer definition. 2016-10-07 06:22:12 +01:00
Matt Graham
8d6f37668f Adding ability to adjust batch size after initialisation and resetting to initial state (order). 2016-10-07 06:21:34 +01:00
Matt Graham
f979aae573 Adding lab start up instructions. 2016-10-06 12:30:05 +01:00
Matt Graham
e8333c02f1 Squashed commit of fixes to allow lab 1 notebook to render on Github. 
commit eb66a44a9f899e20457576a3d34ceb33e0b46004
Author: Matt Graham <m.m.graham@ed.ac.uk>
Date:   Mon Oct 3 12:17:55 2016 +0100

    Restoring output without static function definitions.

commit 623748b334b7f0667c0b2d6e2f829a49978e87eb
Author: Matt Graham <m.m.graham@ed.ac.uk>
Date:   Mon Oct 3 12:16:29 2016 +0100

    Removing loaded static function definitions to test if helps Github rendering.

commit 9b4c21b97e7766ba20db00f80bbd31081417c269
Author: Matt Graham <m.m.graham@ed.ac.uk>
Date:   Mon Oct 3 12:14:42 2016 +0100

    Removing outputs from notebook to check if allows Github to render.
 2016-10-06 12:30:05 +01:00