Jensun Ravichandran
56d554ed83
Remove celluloid.py
2021-06-11 23:07:22 +02:00
Jensun Ravichandran
396d569351
Add utils.py
2021-06-11 23:07:07 +02:00
Jensun Ravichandran
24903b761c
[WIP] Add labels.py
2021-06-11 18:48:43 +02:00
Alexander Engelsberger
e4257ec1f1
Merge branch 'dev' of github.com:si-cim/prototorch into dev
2021-06-11 16:10:04 +02:00
Alexander Engelsberger
aaad2b8626
[BUGFIX] Fix labeled components if initialized
2021-06-11 16:09:51 +02:00
Jensun Ravichandran
c0c0044a42
[REFACTOR] Remove CustomLabelsInitializer
2021-06-11 14:52:09 +02:00
Alexander Engelsberger
47d7f5831f
[Refactor] Add Modules for prior distrbutions
2021-06-08 08:36:55 +02:00
Jensun Ravichandran
4f1c879528
[BUGFIX] Update unit tests
2021-06-04 22:29:30 +02:00
Jensun Ravichandran
2272c55092
[BUGFIX] Fix typo
2021-06-04 22:24:42 +02:00
Jensun Ravichandran
b03c9b1d3c
Add competition and pooling modules
2021-06-04 22:18:46 +02:00
Jensun Ravichandran
0c28eda706
[FEATURE] Remove utility modules and add wrappers instead
2021-06-04 22:16:55 +02:00
Jensun Ravichandran
7bc0bfa3ab
Rename loss functions
2021-06-04 22:15:57 +02:00
Jensun Ravichandran
827958a28a
[FEATURE] Optional transforms in DataAwareInitializers
2021-06-04 22:14:45 +02:00
Jensun Ravichandran
8200e1d3d8
[FEATURE] Allow initialized_components
to be a dataset
2021-06-04 22:13:36 +02:00
Jensun Ravichandran
729b20e9ab
[FEATURE] Add scale to random initializer
2021-06-03 16:35:44 +02:00
Alexander Engelsberger
ca8ac7a43b
[REFACTOR] Probabilistic losses
2021-06-03 14:01:13 +02:00
Alexander Engelsberger
b724a28a6f
[BUGFIX] Stratified functions work on GPU now
2021-06-03 13:19:26 +02:00
Jensun Ravichandran
1e0a8392a2
[QA] Fix for "redefined-builtin" (W0622)
2021-06-02 00:07:44 +02:00
Jensun Ravichandran
2eb7b05653
[FEATURE] Add wrappers for more sklearn datasets
2021-06-01 23:33:51 +02:00
Jensun Ravichandran
d8a0b2dfcc
Minor tweaks
2021-06-01 23:28:01 +02:00
Jensun Ravichandran
2a7394b593
[QA] Remove commented-out torch.jit.script
decorators
2021-06-01 19:46:21 +02:00
Jensun Ravichandran
b1e64c8b8b
[QA] Remove utils.py
2021-06-01 19:41:48 +02:00
Jensun Ravichandran
70cf17607e
[BUGFIX] Fix broken _precheck_initializer
2021-06-01 19:41:21 +02:00
Jensun Ravichandran
b1568a550a
[QA] Fix for "no-self-use" (R0201)
2021-06-01 19:26:05 +02:00
Jensun Ravichandran
e8e803e8ef
[QA] Fix for "dangerous-default-value" (W0102)
2021-06-01 19:24:00 +02:00
Jensun Ravichandran
2c453265fe
[QA] Remove duplicate headings
2021-06-01 19:18:37 +02:00
Jensun Ravichandran
7336d35fee
[QA] Fix "dangerous-default-value" (W0102)
2021-06-01 19:15:06 +02:00
Jensun Ravichandran
bc18952c05
[QA] Fix "dangerous-default-value" (W0102)
2021-06-01 19:10:53 +02:00
Jensun Ravichandran
8e8d0b9c2c
[QA] Fix "list-item-bullet-indent"
2021-06-01 19:08:37 +02:00
Jensun Ravichandran
5a7da2b40b
[QA] Fix for "no-value-for-parameter" (E1120)
2021-06-01 19:03:57 +02:00
Jensun Ravichandran
b6d38f442b
[QA] Remove trailing whitespace
2021-06-01 19:01:20 +02:00
Jensun Ravichandran
8e8851d962
Dynamically remove components
2021-06-01 18:45:47 +02:00
Jensun Ravichandran
27b43b06a7
Rename functions/transform.py -> functions/transforms.py
2021-06-01 17:43:23 +02:00
Jensun Ravichandran
ff69eb1256
Tecator.data is a Tensor and Tecator.targets is a LongTensor
2021-06-01 17:28:37 +02:00
Alexander Engelsberger
4ca581909a
[FEATURE] Change NumpyDataset.data to torch.Tensor
2021-06-01 17:17:42 +02:00
Alexander Engelsberger
2722d976f5
[WIP] Add Growing Neural Gas Energy
2021-06-01 17:16:26 +02:00
Jensun Ravichandran
946cda00d2
Add more competition functions
2021-06-01 12:37:21 +02:00
Jensun Ravichandran
8227525c82
Add LambdaLayer
2021-05-31 16:47:20 +02:00
Jensun Ravichandran
e61ae73749
Make components dynamic
2021-05-31 00:31:40 +02:00
Alexander Engelsberger
040d1ee9e8
Add probabilistic losses
...
Based on Soft LVQ paper by Seo and Obermayer
2021-05-28 20:38:50 +02:00
Alexander Engelsberger
7f0da894fa
Add transformation from distances into gaussian distribution
2021-05-28 16:50:04 +02:00
Alexander Engelsberger
62726df278
Add stratified sum as competition
...
For example used in RSLVQ
2021-05-28 16:49:39 +02:00
Alexander Engelsberger
0ba09db6fe
Bump version: 0.4.5 → 0.5.0
2021-05-28 16:17:49 +02:00
Alexander Engelsberger
87334c11e6
Remove Prototypes1D and its tests
2021-05-28 16:17:49 +02:00
Alexander Engelsberger
40ef3aeda2
Remove usage of Prototype1D
...
Update Iris example to new component API
Update Tecator example to new component API
Update LGMLVQ example to new component API
Update GTLVQ to new component API
2021-05-28 16:17:40 +02:00
Christoph
94fe4435a8
Bump version: 0.4.4 → 0.4.5
2021-05-27 09:58:25 +02:00
Alexander Engelsberger
c204bc8e1f
integrate reviews from ChristophRaab:master
2021-05-27 09:43:02 +02:00
Alexander Engelsberger
00615ae837
refactored gtlvq from ChristophRaab:master
2021-05-27 09:40:42 +02:00
Jensun Ravichandran
9f5f0d12dd
[BUGFIX] Parse dictionary distribution appropirately
2021-05-25 20:52:39 +02:00
Jensun Ravichandran
8a291f7bfb
Overload distribution
argument in component initializers
...
The component initializers behave differently based on the type of the
`distribution` argument. If it is a Python
[list](https://docs.python.org/3/tutorial/datastructures.html ), it is assumed
that there are as many entries in this list as there are classes, and the number
at each location of this list describes the number of prototypes to be used for
that particular class. So, `[1, 1, 1]` implies that we have three classes with
one prototype per class. If it is a Python
[tuple](https://docs.python.org/3/tutorial/datastructures.html ), it a shorthand
of `(num_classes, prototypes_per_class)` is assumed. If it is a Python
[dictionary](https://docs.python.org/3/tutorial/datastructures.html ), the
key-value pairs describe the class label and the number of prototypes for that
class respectively. So, `{0: 2, 1: 2, 2: 2}` implies that we have three classes
with labels `{1, 2, 3}`, each equipped with two prototypes.
2021-05-25 20:05:29 +02:00