Narsil
/
deberta-large-mnli-zero-cls
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update README.md

This commit is contained in:
Nicolas Patry 2021-08-23 13:27:07 +00:00 committed by huggingface-web
parent dfa909d6a9
commit 5a3287ec7c
1 changed files with 2 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
README.md
View File

@ -8,6 +8,7 @@ thumbnail: https://huggingface.co/front/thumbnails/microsoft.png
license: mit license: mit
widget: widget:
- text: "[CLS] I love you. [SEP] I like you. [SEP]" - text: "[CLS] I love you. [SEP] I like you. [SEP]"
pipeline_tag: zero-shot-classification
--- ---
## DeBERTa: Decoding-enhanced BERT with Disentangled Attention ## DeBERTa: Decoding-enhanced BERT with Disentangled Attention
@ -40,9 +41,7 @@ We present the dev results on SQuAD 1.1/2.0 and several GLUE benchmark tasks.
```bash ```bash
cd transformers/examples/text-classification/ cd transformers/examples/text-classification/
export TASK_NAME=mrpc export TASK_NAME=mrpc
python -m torch.distributed.launch --nproc_per_node=8 run_glue.py --model_name_or_path microsoft/deberta-v2-xxlarge \\ python -m torch.distributed.launch --nproc_per_node=8 run_glue.py --model_name_or_path microsoft/deberta-v2-xxlarge \\\n--task_name $TASK_NAME --do_train --do_eval --max_seq_length 128 --per_device_train_batch_size 4 \\\n--learning_rate 3e-6 --num_train_epochs 3 --output_dir /tmp/$TASK_NAME/ --overwrite_output_dir --sharded_ddp --fp16
--task_name $TASK_NAME --do_train --do_eval --max_seq_length 128 --per_device_train_batch_size 4 \\
--learning_rate 3e-6 --num_train_epochs 3 --output_dir /tmp/$TASK_NAME/ --overwrite_output_dir --sharded_ddp --fp16
``` ```
### Citation ### Citation