Skip to content
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

Merge master into new_MT_branch. #977

Merged
merged 30 commits into from
Dec 18, 2019
Merged

Merge master into new_MT_branch. #977

merged 30 commits into from
Dec 18, 2019

Conversation

Kelina
Copy link
Contributor

@Kelina Kelina commented Dec 18, 2019

No description provided.

HaokunLiu and others added 30 commits September 19, 2019 11:45
@HaokunLiu @sleepinyourhat
Readme update for bert npi paper (#915)
10fb192 
* Update README.md

* minor fix

* Typo fix

* typo fix
@HaokunLiu @sleepinyourhat
Fixing index problem & minor pytorch_transformers_interface cleanup (#…
c36b74e 
…916)

* update boundry func with offsets

* update tasks that use indexes

* remove outdated temporary fix
@sleepinyourhat
Prepare for 1.2.1 release.
706b652
@zphang @sleepinyourhat
QA-SRL (#716)
b19ca78 
* Initial QASRL

* Updated pred writing for QASRL

* Add validation shuffle to QASRL

* Remove tqdm, modify class check in preds

* qasrl rebase cleanup

* Update QA-SRL to new repo changes

* Removing src

* QASRL Cleanup

* updating to new model format

* csv to tsv

* QASRL update
Implementing Data Parallel (#873)
2553c2d 
* implemented data parallel

* black style

* Resolve last of merge marks

* deleting irrelevant logs

* adding new way to get attribute

* updating to master

* torch.Tensor -> torch.tensor for n_exs

* black style

* black style

* Merge master

* adapting other tasks to multiple GPU"

* adding helper function for model attributes

* adding get_model_attribute to main.py

* deleting unecessary n_inbput for span_module

* black style

* revert comment change

* fixing batch size keys

* opt_params -> optimizer_params

* Remove extraneous cahnges

* changed n_exs to one-liner

* adapting args.cuda to multi-GPU setting

* adding use_cuda variable

* Fixing parsing for case of args.cuda=subset

* fixing tests

* fixing nits, cleaning up parse_cuda function

* additional nit

* deleted extra space

* Revert nit

* refactoring into get_batch_size

* removing use_cuda

* adding options.py

* removing use_cuda in tests, deleting extra changes

* change cuda default

* change parse_cuda_list_args import

* jiant.options -> jiant.utils.options

* change demo.conf cuda setting

* fix bug -> make parse_cuda return int if only one gpu

* fix bug

* fixed tests

* revert test_retokenize change

* cleaning up code

* adding addiitonal jiant.options

* Separating cuda_device = int case with multiple cuda_device case

* deleting remains of uses_cuda

* remove time logging

* remove use_cuda from evaluate

* val_interval -> validation_interval

* adding cuda comment to tutorial

* fixed typo
@HaokunLiu
replace correct_sent_indexing with non inplace version (#921)
7508bea 
* replace correct_sent_indexing with non inplace version

* Update modules.py

* Update modules.py
@zphang
Abductive NLI (aNLI) (#922)
9d4baf3 
* anli

* anli fix

* Adding aNLI link, additional test/dev warning
SocialIQA (#924)
254dc37 
* black style

* adding SocialQA

* black style

* black style

* fixed socialQA task

* black style

* Update citation

* Nit

* senteval

* socialIQA naming

* reverse unnecessary add
Fixing bug with restoring checkpoint with two gpus + cleaning CUDA pa…
f0ef3f7 
…rsing related code (#928)

* black style

* remove

* cleaning up code around cuda-parsing

* adding defaulting to -1 if there is no cuda devices detected

* fixing nits, throw error instead of log warning for cuda not found
@zphang
Updating CoLA inference script (#931)
8f46d4f
Adding Senteval Tasks (#926)
303a733 
* black style

* adding initial senteval, senteval preprocessing script

* black

* adding senteval to registry

* fixing bigram-shift

* adding label_namespace arg, fixing the ksenteval tasks

* revert extra changes

* black style

* change name -> senteval-probing

* fixing senteval-probing tasks

* renamed senteval -> sentevalprobing

* delete extra imports

* black style

* renaming files and cleaning up preprocessing code

* nit

* black

* deleting pdb

* Senteval -> SE shorthand

* fixing code style
Speed up retokenization (#935)
787e78b 
* black style

* pre-loading tokenizer before retokenization function
@phu-pmh
Scitail (#943)
2ed6802 
* scitail

* Scitail

* Scitail

* update Scitail, removed config

* update Scitail, removed config
@pitrack
Add corrected data stastistics (#941)
fb1eec1 
Thanks to #936, we've discovered errors in our data statistics reporting in the edge probing paper. This table contains the corrected values. As there is more space here, the full (unrounded) values are reported instead. This was generated by a script that read the stats.tsv file and the diff vs. the paper should match my comment on the issue yesterday.
@HaokunLiu
CommonsenseQA+hellaswag (#942)
1d40f23 
* add commonsenseqa task

* add hellaswag task

* dabug

* from #928

* add special tokens to CommensenseQA input

* format

* revert irrelevant change

* Typo fix

* delete

* rename stuff

* Update qa.py

* black
@HaokunLiu
fix name (#945)
3b07a5e
@HaokunLiu
CCG update (#948)
347f743 
* generalize ccg to other transformer models

* debug

* I don't know who broke this at what time, but let's just fix it here now
Fixing senteval-probing preprocessing (#951)
41abe5f 
* Copying configs from superglue

* adding senteval probing config commands

* adding meta-script for transfer and probing exps

* Adding meta bash script fixed

* give_permissions script

* small fix transfer_analysis.sh (#946)

model_*.th might indicate several models; fixed to model_*.best.th

* lr_patience fix

* target_task training -> pretrain training

* adding edgeprobing configs and command

* adding edge probing conf

* fix load_target_train bug

* add hyperparameter sweeping

* val_interval change

* adding sweep function

* Task specific val_intervals

* add reload_vocab to hyperparameter sweep

* adding batch_size specification

* fixing senteval-word-content

* fixing senteval preprocess script

* revert extra delete

* remove extra files

* black format

* black formatting trainer.py

* remove load_data()

* removing extra changes
Adding tokenizer alignment function (#953)
98b1dc8 
* Copying configs from superglue

* adding senteval probing config commands

* adding meta-script for transfer and probing exps

* Adding meta bash script fixed

* give_permissions script

* small fix transfer_analysis.sh (#946)

model_*.th might indicate several models; fixed to model_*.best.th

* lr_patience fix

* target_task training -> pretrain training

* adding edgeprobing configs and command

* adding edge probing conf

* fix load_target_train bug

* add hyperparameter sweeping

* val_interval change

* adding sweep function

* Task specific val_intervals

* add reload_vocab to hyperparameter sweep

* adding batch_size specification

* fixing senteval-word-content

* fixing senteval preprocess script

* revert extra delete

* remove extra files

* black format

* black formatting trainer.py

* remove load_data()

* removing extra changes

* adding alignment mapping function

* fix comment nits

* comment nit

* adding example of token_alignment
@HaokunLiu
Function words probing (#949)
d769338 
* add nli prob task template

* Create acceptablity_probing.py

* specify nli probing tasks

* port acceptablity probing tasks

* add directory name

* debug

* debug

* format

* black

* revert unintended change
@phu-pmh
CosmosQA (#952)
8af068d 
* misc run scripts

* cosmosqa

* cosmosqa

* cosmosqa

* cosmosqa run

* cleaned up repo

* cleaned up repo

* reformatted
@zphang
qqp fix (#956)
1ee0d95
@zphang
QAMR + QA-SRL Update (#932)
2a9230b 
* qamr

* tokenization

* temp qamr

* qamr

* QASRL

* Undo slicing

* quick hack to bypass bad qasrl examples

* f1 em fix

* tokenization fixes

* average

* New tokenization aligner

* update example counts

* Cleanup

* Typography
@njjiang @sleepinyourhat
Set _unk_id in Roberta module (#959)
39b234e 
Currently the `_unk_id` for Roberta is not set correctly, which triggers the assertion error on line 118.
Fixing load_target_train_checkpoint with mixing setting (#960)
7dc9965 
* adding loading for mix

* black style
@pyeres @sleepinyourhat
update pytorch and numpy version requirements (#965)
daec5cf
@HaokunLiu
CCG update (#955)
8a059b8 
* generalize ccg to other transformer models

* debug

* I don't know who broke this at what time, but let's just fix it here now

* ccg lazy iterator

* debug

* clean up

* debug

* debug ccg, minor cleanup
@pyeres @sleepinyourhat
add adversarial_nli tasks (#966)
c181273
@sleepinyourhat
Update README.md
42b389f
@sleepinyourhat
Citation fix
18ca100
@pep8speaks
Copy link

Hello @Kelina! Thanks for opening this PR. We checked the lines you've touched for PEP 8 issues, and found:

Line 1:24: W291 trailing whitespace

Line 1004:101: E501 line too long (110 > 100 characters)

Line 177:101: E501 line too long (103 > 100 characters)

Line 135:101: E501 line too long (106 > 100 characters)

Line 118:101: E501 line too long (125 > 100 characters)
Line 129:101: E501 line too long (104 > 100 characters)
Line 132:101: E501 line too long (114 > 100 characters)
Line 134:101: E501 line too long (124 > 100 characters)
Line 135:101: E501 line too long (120 > 100 characters)
Line 174:101: E501 line too long (110 > 100 characters)
Line 176:101: E501 line too long (104 > 100 characters)
Line 179:101: E501 line too long (109 > 100 characters)

Line 795:101: E501 line too long (104 > 100 characters)

Line 1:4: W291 trailing whitespace

Line 102:76: W291 trailing whitespace
Line 103:12: W291 trailing whitespace
Line 104:1: W293 blank line contains whitespace
Line 111:45: W291 trailing whitespace

Line 51:101: E501 line too long (148 > 100 characters)
Line 52:101: E501 line too long (136 > 100 characters)
Line 59:101: E501 line too long (189 > 100 characters)
Line 60:101: E501 line too long (186 > 100 characters)
Line 63:101: E501 line too long (248 > 100 characters)
Line 64:101: E501 line too long (245 > 100 characters)
Line 67:101: E501 line too long (177 > 100 characters)
Line 68:101: E501 line too long (172 > 100 characters)
Line 71:101: E501 line too long (219 > 100 characters)
Line 72:101: E501 line too long (215 > 100 characters)
Line 75:101: E501 line too long (178 > 100 characters)
Line 76:101: E501 line too long (167 > 100 characters)
Line 79:101: E501 line too long (188 > 100 characters)
Line 80:101: E501 line too long (183 > 100 characters)
Line 83:101: E501 line too long (171 > 100 characters)
Line 84:101: E501 line too long (155 > 100 characters)
Line 87:101: E501 line too long (178 > 100 characters)
Line 88:101: E501 line too long (175 > 100 characters)
Line 95:101: E501 line too long (233 > 100 characters)
Line 96:101: E501 line too long (222 > 100 characters)
Line 99:101: E501 line too long (149 > 100 characters)
Line 100:101: E501 line too long (146 > 100 characters)
Line 103:101: E501 line too long (238 > 100 characters)
Line 104:101: E501 line too long (235 > 100 characters)
Line 107:101: E501 line too long (211 > 100 characters)
Line 108:101: E501 line too long (208 > 100 characters)
Line 111:101: E501 line too long (125 > 100 characters)
Line 112:101: E501 line too long (121 > 100 characters)
Line 115:101: E501 line too long (215 > 100 characters)
Line 116:101: E501 line too long (213 > 100 characters)
Line 123:101: E501 line too long (127 > 100 characters)
Line 124:101: E501 line too long (126 > 100 characters)
Line 127:101: E501 line too long (159 > 100 characters)
Line 128:101: E501 line too long (156 > 100 characters)
Line 131:101: E501 line too long (248 > 100 characters)
Line 132:101: E501 line too long (242 > 100 characters)
Line 135:101: E501 line too long (110 > 100 characters)
Line 136:101: E501 line too long (107 > 100 characters)
Line 139:101: E501 line too long (212 > 100 characters)
Line 140:101: E501 line too long (206 > 100 characters)
Line 143:101: E501 line too long (218 > 100 characters)
Line 144:101: E501 line too long (215 > 100 characters)
Line 147:101: E501 line too long (147 > 100 characters)
Line 148:101: E501 line too long (145 > 100 characters)
Line 151:101: E501 line too long (149 > 100 characters)
Line 152:101: E501 line too long (138 > 100 characters)
Line 155:101: E501 line too long (195 > 100 characters)
Line 156:101: E501 line too long (188 > 100 characters)
Line 159:101: E501 line too long (131 > 100 characters)
Line 160:101: E501 line too long (128 > 100 characters)
Line 163:101: E501 line too long (221 > 100 characters)
Line 164:101: E501 line too long (218 > 100 characters)
Line 167:101: E501 line too long (366 > 100 characters)
Line 168:101: E501 line too long (362 > 100 characters)
Line 175:101: E501 line too long (171 > 100 characters)
Line 176:101: E501 line too long (166 > 100 characters)
Line 179:101: E501 line too long (199 > 100 characters)
Line 180:101: E501 line too long (198 > 100 characters)
Line 183:101: E501 line too long (214 > 100 characters)
Line 184:101: E501 line too long (210 > 100 characters)
Line 187:101: E501 line too long (212 > 100 characters)
Line 188:101: E501 line too long (209 > 100 characters)
Line 195:101: E501 line too long (144 > 100 characters)
Line 196:101: E501 line too long (138 > 100 characters)
Line 199:101: E501 line too long (234 > 100 characters)
Line 200:101: E501 line too long (232 > 100 characters)
Line 203:101: E501 line too long (207 > 100 characters)
Line 204:101: E501 line too long (204 > 100 characters)
Line 207:101: E501 line too long (301 > 100 characters)
Line 208:101: E501 line too long (289 > 100 characters)
Line 211:101: E501 line too long (269 > 100 characters)
Line 212:101: E501 line too long (264 > 100 characters)
Line 215:101: E501 line too long (171 > 100 characters)
Line 216:101: E501 line too long (169 > 100 characters)
Line 219:101: E501 line too long (240 > 100 characters)
Line 220:101: E501 line too long (235 > 100 characters)
Line 223:101: E501 line too long (117 > 100 characters)
Line 224:101: E501 line too long (115 > 100 characters)
Line 227:101: E501 line too long (232 > 100 characters)
Line 228:101: E501 line too long (227 > 100 characters)
Line 231:101: E501 line too long (232 > 100 characters)
Line 232:101: E501 line too long (221 > 100 characters)
Line 235:101: E501 line too long (241 > 100 characters)
Line 236:101: E501 line too long (224 > 100 characters)
Line 239:101: E501 line too long (146 > 100 characters)
Line 240:101: E501 line too long (140 > 100 characters)
Line 243:101: E501 line too long (235 > 100 characters)
Line 244:101: E501 line too long (227 > 100 characters)
Line 251:101: E501 line too long (219 > 100 characters)
Line 252:101: E501 line too long (197 > 100 characters)

You can repair most issues by installing black and running: black -l 100 ./*. If you contribute often, have a look at the 'Contributing' section of the README for instructions on doing this automatically.

@Kelina Kelina merged commit f0bc972 into new_MT_branch Dec 18, 2019
@jeswan jeswan mentioned this pull request Sep 17, 2020
Closed
@jeswan jeswan added the jiant-v1-legacy Relevant to versions <= v1.3.2 label Sep 17, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@iftenney iftenney Awaiting requested review from iftenney

@pruksmhc pruksmhc Awaiting requested review from pruksmhc

@sleepinyourhat sleepinyourhat Awaiting requested review from sleepinyourhat

@W4ngatang W4ngatang Awaiting requested review from W4ngatang

Assignees
No one assigned
Labels
jiant-v1-legacy Relevant to versions <= v1.3.2
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
10 participants
@Kelina @pep8speaks @HaokunLiu @jeswan @sleepinyourhat @zphang @phu-pmh @pitrack @njjiang @pyeres