Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
grammarly
/
pseudonymization-seq2seq
like
5
Follow
Grammarly
90
Text2Text Generation
grammarly/pseudonymization-data
cnn_dailymail
imdb
English
License:
apache-2.0
Model card
Files
Files and versions
Community
main
pseudonymization-seq2seq
/
flair
3 contributors
History:
2 commits
oleksandryermilov
Upload pytorch_model.bin
fcb756d
over 1 year ago
.DS_Store
Safe
6.15 kB
Initiaal commit
over 1 year ago
config.json
Safe
1.74 kB
Initiaal commit
over 1 year ago
generation_config.json
Safe
297 Bytes
Initiaal commit
over 1 year ago
pytorch_model.bin
Safe
558 MB
LFS
Upload pytorch_model.bin
over 1 year ago
special_tokens_map.json
Safe
280 Bytes
Initiaal commit
over 1 year ago
tokenizer.json
Safe
2.11 MB
Initiaal commit
over 1 year ago
tokenizer_config.json
Safe
344 Bytes
Initiaal commit
over 1 year ago
trainer_state.json
Safe
10.4 kB
Initiaal commit
over 1 year ago
training_args.bin
pickle
Detected Pickle imports (6)
"torch.device"
,
"transformers.training_args_seq2seq.Seq2SeqTrainingArguments"
,
"transformers.trainer_utils.IntervalStrategy"
,
"transformers.trainer_utils.SchedulerType"
,
"transformers.trainer_utils.HubStrategy"
,
"transformers.training_args.OptimizerNames"
How to fix it?
3.71 kB
Initiaal commit
over 1 year ago
vocab.json
Safe
798 kB
Initiaal commit
over 1 year ago