Commit History

Add multilingual to the language tag
f39fef7

lbourdois commited on

check if ds is load
5e0ceba

marinone94 commited on

reset script
5e05341

marinone94 commited on

add save total
4f87524

marinone94 commited on

fix params
7036d6a

marinone94 commited on

train on all data
93b0d35

marinone94 commited on

avoid pushing checkpoints
98dfb11

marinone94 commited on

fix decoder storing
a9f9b4a

marinone94 commited on

and push model, ofc
f80fd50

marinone94 commited on

fix language hard coding
b9d77dc

marinone94 commited on

hard code language
fa4a655

marinone94 commited on

reset to set prefix tokens
bb38ae2

marinone94 commited on

fix tensor dim
0713f7f

marinone94 commited on

fix tensor dim
c4280a8

marinone94 commited on

fix torch dim
072b53c

marinone94 commited on

use decode to inspect ds
aac3fbb

marinone94 commited on

use tokenizer to batch_decode
9e467b3

marinone94 commited on

debug dataset
e023bb4

marinone94 commited on

ifx overwriting var
b9bdaf4

marinone94 commited on

print preds for debugging
0157ee1

marinone94 commited on

fix casting single dataset
27673c0

marinone94 commited on

use take instead of select
5920347

marinone94 commited on

debug decoding
a057c82

marinone94 commited on

Merge branch 'main' of https://huggingface.co/marinone94/whisper-tiny-sv
f850b55

marinone94 commited on

add readme only
4de1fa4

marinone94 commited on

skip config ds
5c64556

marinone94 commited on

add readme only
84181fe

marinone94 commited on

Merge branch 'main' of https://huggingface.co/marinone94/whisper-tiny-sv into main
dda4080

marinone94 commited on

fix redme issues
bd3f125

marinone94 commited on

add readme only
6d6b33c

marinone94 commited on

test model card
124e0f4

marinone94 commited on

fix model cards
e07fd66

marinone94 commited on

login wandb
4e6639e

marinone94 commited on

upgrade torchaudio
a254b8a

marinone94 commited on

run 50 steps
90f3365

marinone94 commited on

use multiple datasets
cbbd9a3

marinone94 commited on

WIP: mix datasets
dbe5e43

marinone94 commited on

fix raw dataset features
b7db389

marinone94 commited on

reduce max steps
b9efe48

marinone94 commited on

allow multiple languages and datasets
6ea4d4a

marinone94 commited on

skip push to hub during training
ee5b1b2

marinone94 commited on

allowing multiple datasets
23bb45c

marinone94 commited on

handle PermissionError
6d4cdd4

marinone94 commited on

get also wandb and email creds from file
d6fb5e8

marinone94 commited on

get token from creds if not set in venv
fd7be5b

marinone94 commited on

force train + eval
2affbb8

marinone94 commited on