issues
search
GateNLP
/
gateplugin-LearningFramework
A plugin for the GATE language technology framework for training and using machine learning models. Currently supports Mallet (MaxEnt, NaiveBayes, CRF and others), LibSVM, Scikit-Learn, Weka, and DNNs through Pytorch and Keras.
https://gatenlp.github.io/gateplugin-LearningFramework/
GNU Lesser General Public License v2.1
26
stars
6
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Exception when training regression problems
#23
johann-petrak
closed
8 years ago
1
Exporting regression problem misses attributes/target
#22
johann-petrak
closed
8 years ago
0
Implement exporting for Mallet
#21
johann-petrak
closed
8 years ago
1
Implement additional feature functions, like wordshape and character n-grams
#20
johann-petrak
opened
8 years ago
4
Consider interfacing with Stanford classification and CRF
#19
johann-petrak
closed
8 years ago
1
Provide a way to access the evaluation result programmatically
#18
ianroberts
closed
8 years ago
1
Support list-valued features as attributes
#17
johann-petrak
closed
8 years ago
3
Allow to limit attribute list to within the same covering annotation
#16
johann-petrak
closed
8 years ago
3
Use a default output annotation set for classification/regression application PRs
#15
johann-petrak
closed
8 years ago
1
LF_ClassificationApplication shows an odd message about training.
#14
johann-petrak
closed
8 years ago
2
LF_ClassificationApplication outputAS parameter does not cause new annotations to be created
#13
johann-petrak
closed
8 years ago
1
Add ordinal regression algorithms and/or problem
#12
johann-petrak
opened
8 years ago
1
Add candidate ranking as a problem
#11
johann-petrak
opened
8 years ago
0
Add a way to lookup the numeric representation of a nominal attribute.
#10
johann-petrak
closed
8 years ago
1
Make it work properly with duplication and parallel processing
#9
johann-petrak
closed
6 years ago
17
Better scaling.
#8
johann-petrak
closed
8 years ago
2
Make it possible to use term frequency counts for ngrams and one-of-k attributes
#7
johann-petrak
closed
8 years ago
1
Move the methods for adding instances from the corpus representation to the engine
#6
johann-petrak
closed
6 years ago
1
Add resource helper like way to call the most important functions on the PRs.
#5
johann-petrak
opened
8 years ago
0
Add USER algorithms.
#4
johann-petrak
closed
6 years ago
2
Make configuring EngineMalletSeq more flexible.
#3
johann-petrak
closed
6 years ago
1
Generalize EngineMalletSeq for all Transducers that take FeatureVectorSequences
#2
johann-petrak
closed
6 years ago
1
Implement all Mallet Classification methods and their parameters.
#1
johann-petrak
closed
6 years ago
1
Previous