forked from mike-b-roberts/morePVs
/
hpc_batch_setup.py
163 lines (142 loc) · 5.8 KB
/
hpc_batch_setup.py
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
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
import os
import en_utilities as um
import pandas as pd
import shutil
import numpy as np
import csv
import sys
import stat
# change
def main(project, study, base_path, maxjobs, node):
# Variables
# ---------
num_threads = 2
# See explanation: `https://stackoverflow.com/questions/51256738/multiple-instances-of-python-running-simultaneously-limited-to-35`
# Establish paths etc
# -------------------
new_project = project+'_hpc'
i_path = os.path.join(base_path,project,'inputs')
i_name ='study_'+study+'.csv'
i_file = os.path.join(i_path,i_name)
np_path =os.path.join(base_path,new_project)
if not os.path.exists (np_path):
os.makedirs(np_path)
new_i_path =os.path.join(np_path,'inputs')
if not os.path.exists (new_i_path):
os.makedirs(new_i_path)
df = pd.read_csv(i_file)
df = df.set_index('scenario')
# Path for bash script files and for script
# ----------------------------------------
bash_path = '/home/z5044992/InputOutput/en/morePVs'
# Split input (s'study_....csv') files
# ------------------------------------
length = len(df)
num_jobs = min(length, maxjobs)
joblength = {}
for job in np.arange (num_jobs):
if job <= round((length/num_jobs - int(length/num_jobs))*num_jobs)-1:
joblength[job] = round(length/num_jobs +0.5)
else:
joblength[job] = int(length/num_jobs)
df1 = df.copy()
csv_list=[]
for job in np.arange(num_jobs):
dfn = pd.DataFrame(df1.iloc[0:joblength[job]],columns=df1.columns)
df1 = df1.iloc[joblength[job]:]
o_name = 'study_'+study+'_hpc'+ str(job).zfill(3) +'.csv'
csv_list += [o_name]
o_file = os.path.join(new_i_path ,o_name)
dfn.to_csv(o_file)
# # Create dict of execution lines:
# # -------------------------------
# execution_line ={}
# for csv_name in csv_list:
# idx = csv_list.index(csv_name)
# study = um.find_between(csv_name,'study_','_')
# execution_line[idx] = \
# 'python /home/z5044992/InputOutput/en/morePVs/morePVs.py -b /home/z5044992/InputOutput/DATA_EN_4 -p '\
# + new_project +' -s ' \
# + um.find_between(csv_name,'study_','.csv')
# num_jobs = len(csv_list)
# if not os.path.exists(bash_path):
# os.makedirs(bash_path)
# for f in os.listdir(bash_path):
# xfile = os.path.join(bash_path, f)
# os.remove(xfile)
# Create single batch bash file
# -----------------------------
bash_content = pd.Series([
'#!/bin/bash',
'#SBATCH --export OPENBLAS_NUM_THREADS='+str(num_threads),
'#SBATCH --mail-user=m.roberts@unsw.edu.au',
'#SBATCH --mail-type=FAIL',
'#SBATCH --job-name='+study,
'#SBATCH --array=0-'+str(num_jobs-1),
'#SBATCH --nodelist=tyrion01b'+node,
'#SBATCH --time=96:00:00',
'#SBATCH --ntasks=1',
'#SBATCH --cpus-per-task=1',
'#SBATCH --mem=8192',
'#SBATCH --output "/home/z5044992/InputOutput/DATA_EN_4/slurm/slurm_%A_%a.out"',
'#SBATCH --error "/home/z5044992/InputOutput/DATA_EN_4/slurm_err/err_%A_%a.err"',
'module load python/3.6',
'source /home/z5044992/python_venv/bin/activate',
'python /home/z5044992/InputOutput/en/morePVs/morePVs.py -b /home/z5044992/InputOutput/DATA_EN_4 -p ' + new_project +' -s ' + study + '_hpc' + '$(printf "%03d" $SLURM_ARRAY_TASK_ID)' + ' -o //share/scratch/z5044992/working/' + new_project,
'deactivate',
'module unload python/3.6',
'cp -pr //share/scratch/z5044992/working/'+new_project+'/'+study +'_hpc'+'$(printf "%03d" $SLURM_ARRAY_TASK_ID) //share/scratch/z5044992/outputs',
'rm -rf //share/scratch/z5044992/working/'+new_project+'/'+study + '_hpc' + '$(printf "%03d" $SLURM_ARRAY_TASK_ID)',
'rm /home/z5044992/InputOutput/DATA_EN_4/studies/' + new_project + '/inputs/study_' + study+'_hpc'+'$(printf "%03d" $SLURM_ARRAY_TASK_ID)'+'.csv'
]).apply(lambda x: x.replace('\r\n', '\n'))
# nb replace unix line ending
bash_name = study+'.bat'
bash_file = os.path.join(bash_path, bash_name)
pd.DataFrame(bash_content).to_csv(bash_file,
index=False,
header=False,
quoting=csv.QUOTE_NONE,
line_terminator='\n')
# Make script file executable:
st = os.stat(bash_file)
os.chmod(bash_file, st.st_mode | stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH)
if __name__ == "__main__":
# Input parameters:
# -----------------
default_project = ''
default_study = ''
default_maxjobs = 60
default_base_path = '/home/z5044992/InputOutput/DATA_EN_4/studies'
# Import arguments - allows multi-processing from command line
# ------------------------------------------------------------
opts = {} # Empty dictionary to store key-value pairs.
while sys.argv: # While there are arguments left to parse...
if sys.argv[0][0] == '-': # Found a "-name value" pair.
opts[sys.argv[0]] = sys.argv[1] # Add key and value to the dictionary.
sys.argv = sys.argv[1:]
# Reduce the argument list by copying it starting from index 1.
if '-p' in opts:
project = opts['-p']
else:
project = default_project
if '-s' in opts:
study = opts['-s']
else:
study = default_study
if '-m' in opts:
maxjobs = int(opts['-m'])
else:
maxjobs = default_maxjobs
if '-b' in opts:
base_path = opts['-b']
else:
base_path = default_base_path
if '-n' in opts:
node = opts['-n']
else:
node = '01'
main(project=project,
study=study,
base_path=base_path,
maxjobs=maxjobs,
node=node)