Skip to content

Merge florianmai master to add CNN and LSTM models #10

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
lgalke wants to merge 129 commits into
base: master
Choose a base branch
from
Open

Merge florianmai master to add CNN and LSTM models #10

lgalke wants to merge 129 commits into from

Conversation

@lgalke
Copy link
Member

@lgalke lgalke commented Aug 23, 2019

No description provided.

florianmai added 30 commits June 11, 2017 19:10
@florianmai
Read data from csv format, consider extra samples.
eddf2b2
@florianmai
Removed some unnecessary printouts
d3ed0da
@florianmai
Max features and bugfix
04cf27b
@florianmai
Specify batch size and adjust learning rate accordingly
23f7f0f
@florianmai
Move to keras 2
5bd3286
@florianmai
Fixed a bug occurring when reading data.
9e2e197
@florianmai
Move from deprecated sklearn.cross_validation to sklearn.model_select…
66ae849 
…ion package
@florianmai
Enabled early stopping on validation set.
6c7e22f
@florianmai
Fixed bug in number of steps that occured when moving from keras1.2 t…
57234ca 
…o keras2
@florianmai
Fixed a bug at prediction step where outdated keras functions are used
a7a8022
@florianmai
Learning rate and number of epochs for MLP as command line parameter
cd1a005
@florianmai
Use model checkpoint of best model after early stopping
17dec4b
@florianmai
First version of tensorflow reimplementation of MLP
749e7c0
@florianmai
Wrapper class for tensorflow models in general with mini batching
768b704
@florianmai
MLP_Soph now replicates MLP_Base in tensorflow instead of keras
f4e9c67
@florianmai
Pass sparse tensors directly to tf-models to circumvent tf.constant s…
14ca61a 
…ize limitations
@florianmai
Fixed prediction step for sophmlp
52f4779
@florianmai
Fixed dropout parameter
0d3035d
@florianmai
Switched from tf.contrib.learn to plain tensorflow due to issues with…
007ec5f 
… outdated documentation and deprecated functionality
@florianmai
Adopt keras' categorical xentropy computation
7f8ef02
@florianmai
Fixed a bug in tensorflow prediction step
9d100cb
@florianmai
Compute validation loss in tensorflow-models
a0136a9
@florianmai
Save and use best model in terms of validation loss for prediction
402ce24
@florianmai
Keep TF-reimplementation of keras mlp as 'mlpbase'
e475117
@florianmai
MLP-Soph now has lookup table similar to fastText
c06434e
@florianmai
Skip embedding layer by specifying embedding_size=0
5bc5baa
@florianmai
Self normalizing neural networks for MLP-Soph
3e7f0d9
@florianmai
Bugfix: unable to use 'combined' dataset format with -xX
691485b
@florianmai
Use --fixed_folds and --folds=1 to do only a one fold evaluation with…
d5a90fc 
… fixed folds
@florianmai
Quick-fix for extra-samples not being sampled from the 11th fold
289bf0d
florianmai and others added 24 commits October 27, 2017 15:06
@florianmai
Option for iterating LSTM until max-length
aa2088a
@florianmai
Merge branch 'master' of git.kd.informatik.uni-kiel.de:kd-group/quadf…
1c81d57 
…lor-github
@florianmai
First iteration of Meta-Labeler
4a91c40
@florianmai
Print progress when using meta-labeler
05bb363
@florianmai
Fix in extracting the correct number of labels to predict.
fc0fa6f
@florianmai
Compute score-based Meta-Labeler based on label predictions.
37d8f03
@florianmai
Grid search file now can have string values
592dc09
@florianmai
Option for using batch normalization in MLP.
c0d63c0
@florianmai
Options for choosing number of filters and window sizes for CNN.
3785c6b
@florianmai
Allow lists for grid search
15cad87
@florianmai
Print label statistics when verbose.
1065e98
@florianmai
Merge branch 'master' of git.kd.informatik.uni-kiel.de:kd-group/quadf…
caa0d91 
…lor-github
@florianmai
Documentation for two classes
c220267
@florianmai
Implement OE-LSTMs but not iterate until maxlength but for a set numb…
d0c92c1 
…er of tokens
@florianmai
Remove unneccessary printout
56332ee
@florianmai
OverEager attention: attention only over overeager outputs.
c4915c6
@florianmai
Updated readme and added configuration files for reproducability.
1c7f1e9
@florianmai
Updated requirements
9c1c67e
Update README.md
2cc0984
@florianmai
Merge branch 'master' of github.com:florianmai/Quadflor
f98e49a
@florianmai
Update README.md
d748a5c
@florianmai
Added long version of the paper.
d15414a
@florianmai
Merge branch 'master' of github.com:florianmai/Quadflor
1010936
@lgalke
Resolved conflicts
f2ebfc4
@lgalke lgalke requested a review from florianmai August 23, 2019 23:31
@lgalke
Copy link
Member Author

lgalke commented Aug 24, 2019

We should drop the GloVe model before merging into master

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@florianmai florianmai Awaiting requested review from florianmai

Assignees

No one assigned

Labels

enhancement

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@lgalke @florianmai