braintree/array_final_attack_models.sbatch at master · dapello/braintree · GitHub

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
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
#!/bin/bash
#SBATCH -c 20                # Number of cores
#SBATCH -t 5000              # Runtime in D-HH:MM, minimum of 10 minutes
#SBATCH --mem=40g            # Memory pool for all cores '(see also --mem-per-cpu)
#SBATCH --gres=gpu:QUADRORTX6000:1
#SBATCH --array=0-10
#SBATCH -o slurm_logs/score_models_%A_%a.out  # File to which STDOUT will be written, %j inserts jobid
#SBATCH -e slurm_logs/score_models_%A_%a.err  # File to which STDERR will be written, %j inserts jobid

### for multiple GPUS #SBATCH --array=0-2

export HOME='/om2/user/dapello/'
module load openmind/anaconda/3-4.0.0
source activate braintree-0.2

cd /om2/user/dapello/Code/proj_braintree/braintree-0.2/braintree

nvidia-smi
#sp='new_final_models'
sp='new_AT2_final_models'
ds='COCO'
#ds='HVM_var6'

#norm='inf'
#
#epss=(
#   0.00003063725490196
#   0.00006127450980392
#   0.00012254901960784
#   0.00024509803921568
#   0.00049019607843137
#   0.00098039215686274
#   0.00196078431372549
#   0.00392156862745098
#   0.00784313725490196
#   0.01568627450980392
#)

norm=2

epss=(
    0.00390625
    0.0078125
    0.015625
    0.03125
    0.0625
    0.125
    0.25
    0.5
    1
    2
)

# HVM labels
for eps in ${epss[${SLURM_ARRAY_TASK_ID}]}
do
    python evaluate_model.py --seed 6 --dataset=${ds} --logdir=$sp --norm=$norm --eps=$eps
done