-
Notifications
You must be signed in to change notification settings - Fork 0
/
run_sciama_ADA.py
125 lines (108 loc) · 5.11 KB
/
run_sciama_ADA.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
# -*- coding: utf-8 -*-
"""
Created on Sun Jul 3 12:58:37 2016
@author: moricex
"""
import os
import numpy
import settings
import shutil
from subprocess import Popen, PIPE
import time
#os.chdir(settings.programpath) # Change directory
cwd=os.getcwd()
dirs=os.listdir(cwd)
#MINT = calc_MINT
# Variables to iterate through
n_estimators=numpy.array([8,16,32,64,128,256,512,1024,2048,4096,8192])
#n_estimators=numpy.array([16])
n_train=numpy.array([100,500,1000,2500,5000,10000,25000])
n_learningrate=numpy.array([0.80,0.84,0.88,0.92,0.96,1.00,1.04,1.08,1.12,1.16,1.20])
#n_estimators=numpy.array([8,16])
#n_train=numpy.array([100,500])
#n_depth=numpy.array([3.0,4.0])
#if MINT==1:
## n_proc= numpy.array([4,7,8,9,10,11,12,13,14,15,16])
# n_proc= numpy.array([6,8,8,10,10,12,12])
#else:
n_proc= numpy.array([6,8,8,10,10,12,12])
# CREATE DIRECTORIES AND COPY CODE UP
timestr = time.strftime("%Y%m%d-%H%M%S")
os.chdir(settings.programpath+'runresults')
numpy.savez("stats_%s" %('RUN_'+timestr),n_estimators=n_estimators,n_learningrate=n_learningrate,n_train=n_train)#filters=filters,use_colours=use_colours)
runpaths=[]
runnames=[]
for k in range(0,len(n_estimators)):
for j in range(0,len(n_train)):
for i in range(0,len(n_learningrate)):
run_nameopts='RUN_'+timestr+'_ADA'
run_namewo=run_nameopts+'_n_learningrate_'+str(n_learningrate[i])+'_n_tr_'+str(n_train[j])+'_n_est_'+str(n_estimators[k])#+'max_feat_None'
run_name = '/'+run_namewo+'/'
run_dir=run_nameopts
fullpath=settings.programpath+'runresults'+'/'+run_dir+run_name
dirs_runresults = os.listdir(settings.programpath+'runresults')
if run_dir not in dirs_runresults:
os.mkdir(run_dir)
dirs_runresults = os.listdir(settings.programpath+'runresults/'+run_dir)
if run_namewo not in dirs_runresults:
os.mkdir(fullpath)
runpaths.append(fullpath)
runnames.append(run_namewo)
shutil.copy(settings.programpath+'MLdr12_RF.py',fullpath+'MLdr12_RF.py')
shutil.copy(settings.programpath+'plots.py',fullpath+'plots.py')
shutil.copy(settings.programpath+'settings.py',fullpath+'settings.py')
shutil.copy(settings.programpath+'run_opts.py',fullpath+'run_opts.py')
shutil.copy(settings.programpath+'htmloutput.py',fullpath+'htmloutput.py')
os.chdir(fullpath)
cwd=os.getcwd()
print(cwd)
with open(fullpath+'settings.py',"a") as set_file:
set_file.write("\n# MODIFIED SETTINGS ADDED BY run_sciama.py for run_name:%s"%run_namewo) # Add settings to bottom of settings.py file here
set_file.write("\nprogrampath='"+fullpath+"'")
set_file.write("\n")
#MLAset =
set_file.write("\nMLA = 'sklearn.ensemble.AdaBoostClassifier'")
set_file.write("\nMLAset = {'n_estimators':%s,'learning_rate':%s}"%(n_estimators[k],n_learningrate[i]))
set_file.write("\ntraindatanum=%s"%n_train[j])
set_file.write("\nfeat_outfile = 'ML_RF_feat_%s'" %run_namewo)
set_file.write("\nresult_outfile = 'ML_RF_results_%s'" %run_namewo)
set_file.write("\nprob_outfile = 'ML_RF_probs_%s'" %run_namewo)
set_file.write("\nlog_outfile='ML_RF_log_%s'" %run_namewo)
set_file.write("\nstats_outfile='ML_RF_stats_%s'" %run_namewo)
set_file.write("\nfeatnames_outfile='ML_RF_featnames_%s'" %run_namewo)
set_file.write("\nscores_outfile='ML_RF_scores_%s'" %run_namewo)
set_file.write("\ncalc_MINT = 0")
dirs_run=os.listdir(fullpath)
runfull=fullpath+'MLdr12_RF.py'
# Open a pipe to the qsub command.
p = Popen(["qsub"],stdin=PIPE, stdout=PIPE, close_fds=True,universal_newlines=True)
# output, input = p.stdout, p.stdin
# Customize your options here
job_name = "ML_RF_sciamajob_%s" %run_namewo
walltime = "6:00:00"
processors = "nodes=1:ppn=%s" %n_proc[j]
command = "python MLdr12_RF.py"
# if MINT==1:
## queue = ""
# queue="#PBS -q sciama1.q"
# else:
queue="#PBS -q sciama1.q"
job_string = """#!/bin/bash
#PBS -N %s_numcat*10
#PBS -l walltime=%s
#PBS -l %s
#PBS -j oe
#PBS -V
%s
cd %s
echo "Current working directory is `pwd`"
%s""" % (job_name, walltime, processors,queue,fullpath, command)
# Send job_string to qsub
out,errs=p.communicate(input=job_string)
# input.close()
# Print your job and the system response to the screen as it's submitted
print(job_string)
print(out)
time.sleep(1)
os.chdir(settings.programpath+'runresults')
numpy.save("paths_%s" %run_nameopts,numpy.column_stack((runpaths,runnames)))