forked from jdcla/DNA-transformer
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrun_regulon.sh
41 lines (39 loc) · 801 Bytes
/
run_regulon.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
#!/bin/bash
dataset="data/eco_regulon/eco_regulon_strong.npy"
d_head=32
d_model=32
n_head=6
n_layer=6
dropout=0.10
merge_size=1
tgt_len=512
conv_size=7
output="output/"
echo 'Run training...'
echo $dataset
python train.py --cuda \
--eval-interval 3000 \
--data_path $dataset \
--output_path $output \
--d_embed $d_model \
--n_layer $n_layer \
--d_model $d_model \
--n_head $n_head \
--d_head $d_head \
--dropout $dropout \
--dropatt 0 \
--optim adam \
--lr 0.0001 \
--max_step 85000 \
--warmup_step 3000 \
--tgt_len $tgt_len \
--mem_len 0 \
--varlen \
--eval_tgt_len $tgt_len \
--batch_size 10 \
--shift 20 \
--ext_ds 0 \
--gpu0_bsz 0 \
--conv_size $conv_size \
--not_tied \
--work_dir_path $output