vuiseng9
/

bert-base-uncased-mnli

Text Classification PyTorch Transformers bert

Model card Files Files and versions Community

bert-base-uncased-mnli / README.md

Chua, Vui Seng

Add readme 8e0524e 9 months ago

raw history blame contribute delete

Safe

1.01 kB

	This model is developed with transformers v4.10.3.

	# Train
	```bash
	#!/usr/bin/env bash

	export CUDA_VISIBLE_DEVICES=0

	OUTDIR=bert-based-uncased-mnli
	WORKDIR=transformers/examples/pytorch/text-classification
	cd $WORKDIR

	nohup python run_glue.py \
	--model_name_or_path bert-base-uncased \
	--task_name mnli \
	--do_eval \
	--do_train \
	--per_device_train_batch_size 16 \
	--per_device_eval_batch_size 16 \
	--max_seq_length 128 \
	--num_train_epochs 3 \
	--overwrite_output_dir \
	--output_dir $OUTDIR 2>&1 \| tee $OUTDIR/run.log &
	```

	# Eval
	```bash
	export CUDA_VISIBLE_DEVICES=0

	OUTDIR=eval-bert-based-uncased-mnli
	WORKDIR=transformers/examples/pytorch/text-classification
	cd $WORKDIR

	nohup python run_glue.py \
	--model_name_or_path vuiseng9/bert-base-uncased-mnli \
	--task_name mnli \
	--do_eval \
	--per_device_eval_batch_size 16 \
	--max_seq_length 128 \
	--overwrite_output_dir \
	--output_dir $OUTDIR 2>&1 \| tee $OUTDIR/run.log &
	```