Commits · rshakked/safe-talk

feat: add support for evaluating saved model without retraining

1294c96

rshakked commited on May 13

feat: add Gradio progress bar and streaming log updates during training

9364099

rshakked commited on May 13

fix: pass tokenizer explicitly to AbuseDataset and safeguard evaluation step

5c16708

rshakked commited on May 13

feat(app): run training on button click and display logs after completion

2032430

rshakked commited on May 13

fix: reduce batch size and enable gradient checkpointing to prevent GPU OOM crashes

fedc8f2

rshakked commited on May 13

chore(debug): move device check to top of script for clearer startup logs

285a433

rshakked commited on May 12

chore(debug): print tensor devices and CUDA availability for troubleshooting

d4ac0ac

rshakked commited on May 12

fix: remove .to(device) from dataset to avoid pin_memory error in Trainer

92846e1

rshakked commited on May 12

Add version print for transformers to debug dependency issues

fea5367

rshakked commited on May 12

updated requirments.txt

609058f

rshakked commited on May 12

updated requirments.txt

c5027d1

rshakked commited on May 12

Add accelerate for Trainer compatibility

aaf83ad

rshakked commited on May 12

updated train_abuse_model to save the traind model after training process

5524706

rshakked commited on May 12

Update app.py to stream training logs live

d04cf51

rshakked commited on May 12

Add sentencepiece to support DebertaV2Tokenizer

5d99999

rshakked commited on May 12

Fresh start: safe version with token moved to env secret

80b417a

rshakked commited on May 12

Add all dependencies for model training

046bfc4

rshakked commited on May 12

updated app.py to include clear logging

a3bf050

rshakked commited on May 12

updated app.py to show live outputs and updated gitignore to ignore excel and doc files

a9706bf

rshakked commited on May 12

Remove dataset folder from commit and update .gitignore

536c7e2

rshakked commited on May 12

Update training code

7a679b4

rshakked commited on May 12

added app file and enable GPU training

58be06f

rshakked commited on May 11

Force Space rebuild

643d98c

rshakked commited on May 10

initial commit

e1be091

rshakked commited on May 9

Initial commit

17abdd2

rshakked commited on May 9

created app.py

840e37c

rshakked commited on May 11

Force Space rebuild

cf896a1

rshakked commited on May 10

Add model training script

0618de2

rshakked commited on May 9

Initial commit

f60e547

rshakked commited on May 9

initial commit

e398a5f

rshakked commited on May 9

Spaces:

rshakked
/

safe-talk

Running

Commit History

feat: add support for evaluating saved model without retraining

1294c96

feat: add Gradio progress bar and streaming log updates during training

9364099

fix: pass tokenizer explicitly to AbuseDataset and safeguard evaluation step

5c16708

feat(app): run training on button click and display logs after completion

2032430

fix: reduce batch size and enable gradient checkpointing to prevent GPU OOM crashes

fedc8f2

chore(debug): move device check to top of script for clearer startup logs

285a433

chore(debug): print tensor devices and CUDA availability for troubleshooting

d4ac0ac

fix: remove .to(device) from dataset to avoid pin_memory error in Trainer

92846e1

Add version print for transformers to debug dependency issues

fea5367

updated requirments.txt

609058f

updated requirments.txt

c5027d1

Add accelerate for Trainer compatibility

aaf83ad

updated train_abuse_model to save the traind model after training process

5524706

Update app.py to stream training logs live

d04cf51

Add sentencepiece to support DebertaV2Tokenizer

5d99999

Fresh start: safe version with token moved to env secret

80b417a

Add all dependencies for model training

046bfc4

updated app.py to include clear logging

a3bf050

updated app.py to show live outputs and updated gitignore to ignore excel and doc files

a9706bf

Remove dataset folder from commit and update .gitignore

536c7e2

Update training code

7a679b4

added app file and enable GPU training

58be06f

Force Space rebuild

643d98c

initial commit

e1be091

Initial commit

17abdd2

created app.py

840e37c

Force Space rebuild

cf896a1

Add model training script

0618de2

Initial commit

f60e547

initial commit

e398a5f

Commit History

feat: add support for evaluating saved model without retraining 1294c96

feat: add Gradio progress bar and streaming log updates during training 9364099

fix: pass tokenizer explicitly to AbuseDataset and safeguard evaluation step 5c16708

feat(app): run training on button click and display logs after completion 2032430

fix: reduce batch size and enable gradient checkpointing to prevent GPU OOM crashes fedc8f2

chore(debug): move device check to top of script for clearer startup logs 285a433

chore(debug): print tensor devices and CUDA availability for troubleshooting d4ac0ac

fix: remove .to(device) from dataset to avoid pin_memory error in Trainer 92846e1

Add version print for transformers to debug dependency issues fea5367

updated requirments.txt 609058f

updated requirments.txt c5027d1

Add accelerate for Trainer compatibility aaf83ad

updated train_abuse_model to save the traind model after training process 5524706

Update app.py to stream training logs live d04cf51

Add sentencepiece to support DebertaV2Tokenizer 5d99999

Fresh start: safe version with token moved to env secret 80b417a

Add all dependencies for model training 046bfc4

updated app.py to include clear logging a3bf050

updated app.py to show live outputs and updated gitignore to ignore excel and doc files a9706bf

Remove dataset folder from commit and update .gitignore 536c7e2

Update training code 7a679b4

added app file and enable GPU training 58be06f

Force Space rebuild 643d98c

initial commit e1be091

Initial commit 17abdd2

created app.py 840e37c

Force Space rebuild cf896a1

Add model training script 0618de2

Initial commit f60e547

initial commit e398a5f

feat: add support for evaluating saved model without retraining

1294c96

feat: add Gradio progress bar and streaming log updates during training

9364099

fix: pass tokenizer explicitly to AbuseDataset and safeguard evaluation step

5c16708

feat(app): run training on button click and display logs after completion

2032430

fix: reduce batch size and enable gradient checkpointing to prevent GPU OOM crashes

fedc8f2

chore(debug): move device check to top of script for clearer startup logs

285a433

chore(debug): print tensor devices and CUDA availability for troubleshooting

d4ac0ac

fix: remove .to(device) from dataset to avoid pin_memory error in Trainer

92846e1

Add version print for transformers to debug dependency issues

fea5367

updated requirments.txt

609058f

updated requirments.txt

c5027d1

Add accelerate for Trainer compatibility

aaf83ad

updated train_abuse_model to save the traind model after training process

5524706

Update app.py to stream training logs live

d04cf51

Add sentencepiece to support DebertaV2Tokenizer

5d99999

Fresh start: safe version with token moved to env secret

80b417a

Add all dependencies for model training

046bfc4

updated app.py to include clear logging

a3bf050

updated app.py to show live outputs and updated gitignore to ignore excel and doc files

a9706bf

Remove dataset folder from commit and update .gitignore

536c7e2

Update training code

7a679b4

added app file and enable GPU training

58be06f

Force Space rebuild

643d98c

initial commit

e1be091

Initial commit

17abdd2

created app.py

840e37c

Force Space rebuild

cf896a1

Add model training script

0618de2

Initial commit

f60e547

initial commit

e398a5f