Commit History

feat: model config not hardcoded
ad6ad64

boris commited on

Use jax.device_count(), don't assume 8.
a11eff5

Pedro Cuenca commited on

Add missing import for CLIP.
7cc9b38

Pedro Cuenca commited on

add tokenizer save to wandb:
aecf3a7

tmabraham commited on

Merge pull request #30 from pcuenca/clip-score
6567fd7

Tanishq Abraham commited on

Merge pull request #29 from borisdayma/load_checkpoint
862924a
unverified

boris commited on

Score results using pre-trained CLIP.
5a9a1b6

Pedro Cuenca commited on

fix: config used in preprocess
6d252e9

boris commited on

fix: custom reference
e803feb

boris commited on

fix: typo
ca83cca

boris commited on

feat: allow loading a model checkpoint
3d61350

boris commited on

feat: hardcode our full dataset
499ddb2

boris commited on

Merge pull request #25 from borisdayma/fix-config
d811136
unverified

boris commited on

feat: bigger warnings
62dad48

boris commited on

feat: use str mode for json
3e6ab1f

boris commited on

fix: import json
90320ea

boris commited on

fix: output directory must exist
6e89e9e

boris commited on

feat: update scriptst
63249ac

boris commited on

feat: update model config + save optim
a30dbd3

boris commited on

Merge pull request #28 from patil-suraj/tpu-demo
e31a84f
unverified

Ritobrata Ghosh commited on

remove .ipynb
c879290

valhalla commited on

add tpu demo notebook
4b5a542

valhalla commited on

Merge pull request #27 from tmabraham/fix-forced-bos-token-on-demo
8f484d9
unverified

pcuenca commited on

fix forced bos token, also applying BART model to 8 samples now
8d4e13c

tmabraham commited on

Merge pull request #26 from tmabraham/generation-training-demo
fc8c230
unverified

boris commited on

demo for generation, including during training from wandb artifact
c48da33

tmabraham commited on

fix: model config
5aaf9df

boris commited on

Merge pull request #23 from khalidsaifullaah/main
eb591ff
unverified

pcuenca commited on

Merge pull request #24 from borisdayma/feat--log-model-frequently
648e404
unverified

boris commited on

feat: use bart-large-cnn
19d68bb

boris commited on

fix: log metadata
99a1ff5

boris commited on

fix: define function before it is used
d449092

boris commited on

fix: correct arg
283adc6

boris commited on

feat: save model frequently
754f876

boris commited on

feat: split script for small and big runs
5e244d0

boris commited on

feat: update test script
3cccb01

boris commited on

feat: bye bye tensorboard
533b494

boris commited on

feat: use bart large
bb3bfa6

boris commited on

YFCC metadata cleaning and encoding script
2c2f570

khalidsaifullaah commited on

Merge pull request #22 from borisdayma/feat-axis
a1c047b
unverified

boris commited on

fix: use correct key
b20769d

boris commited on

fix: log correct metrics
3fef9c1

boris commited on

feat: hardcode eval_steps
4c5e5a7

boris commited on

fix: eval_steps belongs to training_args
900136f

boris commited on

feat: eval_steps already exists in TrainingArguments
0a0080b

boris commited on

Merge branch 'main'
3ddf1c5

boris commited on

feat: set default x-axis
97a008e

boris commited on

feat: log everything through wandb
19070ab

boris commited on

Merge pull request #21 from borisdayma/feat-no_decay
b29bab7
unverified

boris commited on

feat: eval less often for faster training
f0a53ac

boris commited on