forked from EleutherAI/lm-evaluation-harness
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathrun_loop.sh
More file actions
24 lines (20 loc) · 715 Bytes
/
run_loop.sh
File metadata and controls
24 lines (20 loc) · 715 Bytes
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
#!/bin/bash
#SBATCH --time=72:00:00
#SBATCH --ntasks=1
#SBATCH --nodes=1
#SBATCH --cpus-per-task=16
#SBATCH --gres=gpu:1
#SBATCH --mem=64G
#you can run a job with this command:
#sbatch --partition=xxx --job-name=t5-small --output=./outfile/t5-small.out --error=./outfile/t5-small.err run_loop.sh hf-seq2seq MBZUAI/LaMini-T5-61M eval_res/t5-small-
source ~/.bashrc
conda activate your_evn
for task in race sciq arc_challenge piqa sst mrpc cb rte wsc wsc273 winogrande wic copa webqs mnli mnli_mismatched anli_r1 anli_r2 anli_r3 squad2 record hellaswag qqp triviaqa
do
python main.py \
--model $1 \
--model_args pretrained=$2 \
--tasks $task \
--output_path $3$task.txt \
--device cuda:0
done