我正在使用python脚本进行数据分析，每组分析大约需要2-3个小时。是否可以编写一个可以一个接一个地运行另一个文件的脚本？ 例如我在主文件中指定了要运行的分析文件，而我只运行主文件，它将自动运行分析文件？
或者，有没有任何编辑器可以做到这一点？
来自python和程序设计的初学者。
下面是我要多次运行的脚本（更改了infile和outfile位置之后）
import numpy as np
import pylab
import csv
from scipy.optimize import curve_fit
from scipy.stats import t
import matplotlib.pyplot as plt
#basic information --set up variable
#method infodisgunsh
denatureMethod="CD" # choose 'CD' for chemical denature, 'HD' for heat denature
CD_xaxis_FinalGdmClConcentration_value=[0,0.43,0.87,1.3,1.74,2.17,2.61,3.04,3.48]
CD_conc_upbound=3.48
CD_conc_lowbound=0
CD_conc_range=CD_conc_upbound-CD_conc_lowbound
HD_xasis_normalizedTemp_value=[0,0.129277567,0.224334601,0.338403042,0.429657795,0.539923954,0.646387833,0.775665399,0.882129278,1]
HD_Temp_upbound=63
HD_Temp_lowbound=36.7
HD_temp_range=HD_Temp_upbound-HD_Temp_lowbound
CI_cutoff=0.3
#run info
descriptionOfTheRun="F_diet"# what ever desciption can distinguish
numOfReplicate="1" # number of replicate used to generate 1 fit
plotColor="2"#1=blue,2=organe,3=purple,4=brown
#file selection
infile_locationAndName="G:\BYU OneDrive_G\OneDrive - BYU Office 365\BYU\===Price lab===\Coding\\1023-CD+HD +multi rep\\1023-input_python_test.csv"
outfile_locationAndName="G:\BYU OneDrive_G\OneDrive - BYU Office 365\BYU\===Price lab===\Coding\\1023-CD+HD +multi rep\\1026-CDtest_output.csv"
outImage_location="G:\BYU OneDrive_G\OneDrive - BYU Office 365\BYU\===Price lab===\Coding\\1023-CD+HD +multi rep\\"
#info for each peptide from infile
rowLoc=0
IDLoc=1
NameLoc=2
PeptideLoc=3
ReporterLoc=4
#rep1 info
rep1_notebookCode="LL1091" #the notebook code of the run
rep1_numPtsLocation=5# type in the number get from excel column to python raw translator
rep1_Statrt=6 # type in the number get from excel column to python raw translator
rep1_End=14# type in the number get from excel column to python raw translator
"""#rep2 info
rep2_notebookCode="LL1191" #the notebook code of the run
rep2_numPtsLocation=16# type in the number get from excel column to python raw translator
rep2_Statrt=17 # type in the number get from excel column to python raw translator
rep2_End=26# type in the number get from excel column to python raw translator
#rep3_info
rep3_notebookCode="LL1101" #the notebook code of the run
rep3_numPtsLocation=1# type in the number get from excel column to python raw translator
rep3_Statrt=5 # type in the number get from excel column to python raw translator
rep3_End=6# type in the number get from excel column to python raw translator"""
"""#rep4_info
rep4_notebookCode="LL1101" #the notebook code of the run
rep4_numPtsLocation=1# type in the number get from excel column to python raw translator
rep4_Statrt=5 # type in the number get from excel column to python raw translator
rep4_End=6# type in the number get from excel column to python raw translator"""
#plot color Code, usually don't need to change
color1_dataAndFit="#1f77b4"#blue
color1_Chalf="#aec7e8"
color1_CI="#c6dbef"
color2_dataAndFit="#ff7f0e"#orgnae
color2_Chalf="#ffbb78"
color2_CI="#fdd0a2"
color3_dataAndFit="#5254a3"#purple
color3_Chalf="#9e9ac8"
color3_CI="#dadaeb"
color4_dataAndFit="#8c6d31"#brown
color4_Chalf="#e7ba52"
color4_CI="#e7cb94"
def sigmoid(x,B,A,Chalf, b):#the fitting equation
y =B+((A-B)/(1+np.exp((-1/b)*(Chalf-x))))
return y
infile=open(infile_locationAndName,"r")#"r"=reading mode
reader=csv.reader(infile)#create reader object
header=next(reader)
#base on replication numbers
if numOfReplicate=="1":
xdata1=[]
finalData=[]
for i in range(rep1_Statrt,rep1_End+1):#5 is the data start, 15 is the end of data+1
flow=float(header[i])#making the input value to number
xdata1.append(flow)
for row in reader:
# input y value
ydata1=[]
for i in range(rep1_Statrt,rep1_End+1):# it means row[5] to row[15]
flow=float(row[i])#making the input value to number
ydata1.append(flow)
#baisc information input
title=row[rowLoc]
ID=row[IDLoc]
proteinName=row[NameLoc]
Peptide=row[PeptideLoc]
numOfReporter=row[ReporterLoc]
numOf_rep1_pts=row[rep1_numPtsLocation]
#baisc information ouput
tempRow=[title]#[0]
tempRow.append(ID)#[1]
tempRow.append(proteinName)#[2]
tempRow.append(Peptide)#[3]
tempRow.append(numOfReporter)#[4]
tempRow.append(numOf_rep1_pts)#[5]
try:#if statment, when it is not error
#normalized the data
normalized_ydata1=[]
normalized_xdata1=[]
#normalizing rep1 dataset
if denatureMethod=="CD":#first point is 0, last point is 1
for element in ydata1: # read from the begining, find first data !=0, make it A
if element !=0:
A=element# set 0
break
for element in reversed(ydata1): #read from the end, find first data !=0, make it B
if element !=0:
B=element#set 1
#print ("A",A)
break
for i in range(len(ydata1)):
if ydata1[i] !=0:
element=(ydata1[i]-A)/(B-A) #3 for each data !=0, normalized_ydata=(data-A)/(B-A)
normalized_ydata1.append(element)
normalized_xdata1.append(xdata1[i]) #4 for each normalized_ydata, find the xdata, make it normalized_xdata
else:
normalized_ydata1.append("error")
normalized_xdata1.append(xdata1[i])
if denatureMethod=="HD":#first point is 1, last point is 0
for element in ydata1: #1 read from the begining, find first data !=0, make it B
if element !=0:
B=element# set 1
break
for element in reversed(ydata1): #2 read from the end, find first data !=0, make it A
if element !=0:
A=element#set 0
break
for i in range(len(ydata1)):
if ydata1[i] !=0:
element=(ydata1[i]-A)/(B-A) #3 for each data !=0, normalized_ydata=(data-A)/(B-A)
normalized_ydata1.append(element)
normalized_xdata1.append(xdata1[i]) #4 for each normalized_ydata, find the xdata, make it normalized_xdata
else:
normalized_ydata1.append("error")
normalized_xdata1.append(xdata1[i])
#Set data set for fit and plot
xplot1=[]
yplot1=[]
for i in range (len(normalized_ydata1)):
if normalized_ydata1[i]!="error":
yplot1.append(normalized_ydata1[i])
xplot1.append(normalized_xdata1[i])
#fit Rep1 to s curve
AL_fit_result_output=[]
popt1, pcov1 = curve_fit(sigmoid, xplot1, yplot1)
AL_fit_result_output.extend(popt1)#extend merge two list to one big list
#popt has all the parameter (variable) in regression equation (x,B, A, Chalf, b)
#row [-, 0, 1, 2, 3] in AL_fit_result_output
if denatureMethod=="CD":
x1 = np.linspace(0, 4, 50)
if denatureMethod=="HD":
x1 = np.linspace(0, 1.2, 50)
y1= sigmoid(x1, *popt1)# *popt split the two variable
AL_Chalf=AL_fit_result_output[2]
AL_b=AL_fit_result_output[3]
#calculate AL confidence interval
numOfPoint_AL_repA=float(numOf_rep1_pts)
sumPoints_AL=numOfPoint_AL_repA
tempRow.append(sumPoints_AL)#[6]
AL_Standard_error=np.sqrt(np.diag(pcov1))
AL_fit_result_output.extend(AL_Standard_error)#[4,5,6,7]=[Berror, Aerror, Chalferror, berror]
AL_ConfidenceInterval_chalf=t.ppf(.975, (sumPoints_AL-1))* AL_fit_result_output[6]/np.sqrt(sumPoints_AL)
AL_fit_result_output.append(AL_ConfidenceInterval_chalf)#[8]
if denatureMethod=="CD":
CI_ratioTOrange=AL_ConfidenceInterval_chalf/CD_conc_range
AL_fit_result_output.append(CI_ratioTOrange)#[9]
if denatureMethod=="HD":
CI_ratioTOrange=AL_ConfidenceInterval_chalf/HD_temp_range
AL_fit_result_output.append(CI_ratioTOrange)#[9]
AL_CI_Chalf_lowbound=AL_Chalf-AL_fit_result_output[8]
AL_CI_Chalf_upbound=AL_Chalf+AL_fit_result_output[8]
AL_fit_result_output.append(AL_CI_Chalf_lowbound)#[10]
AL_fit_result_output.append(AL_CI_Chalf_upbound)#[11]
AL_ConfidenceInterval_b=t.ppf(.975, (sumPoints_AL-1))* AL_fit_result_output[7]/np.sqrt(sumPoints_AL)
AL_fit_result_output.append(AL_ConfidenceInterval_b)#[12]
AL_CI_b_lowbound=AL_b-AL_fit_result_output[12]
AL_CI_b_upbound=AL_b+AL_fit_result_output[12]
AL_fit_result_output.append(AL_CI_b_lowbound)#[13]
AL_fit_result_output.append(AL_CI_b_upbound)#[14]
#compute AL r squared
residuals=(yplot1)-sigmoid((xplot1),*popt1)
ss_res=np.sum(residuals**2)
ss_tot=np.sum(((yplot1)-np.mean(yplot1))**2)
r_squared1=1-(ss_res/ss_tot)
AL_fit_result_output.append(r_squared1)#[15]; append add something to the list
if denatureMethod=="HD":
Chalf1_temp=(AL_Chalf*HD_temp_range)+HD_Temp_lowbound
AL_fit_result_output.append(Chalf1_temp)#[16]
if denatureMethod=="CD":
Chalf1_normalized=AL_Chalf/CD_conc_range
AL_fit_result_output.append(Chalf1_normalized)#[16]
tempRow.extend(AL_fit_result_output)
tempRow.extend(normalized_ydata1)
#generate the figure
if CI_ratioTOrange<CI_cutoff:
graphName=descriptionOfTheRun+"_"+"row#"+title+" (has CI)"+"\n"+ID +"||"+ proteinName+"\n"+Peptide
else:
graphName=descriptionOfTheRun+"_"+"row#"+title+"\n"+ID +"||"+ proteinName+"\n"+Peptide
#plot color
if plotColor=="1":#choose from blue, orange, purple, brown
plt.plot(xplot1, yplot1,color=color1_dataAndFit,ls=':',marker='o',label='rep1_'+rep1_notebookCode)
#plt.plot(xplot2, yplot2,color=color1_dataAndFit,ls=':',marker='^',label='AL_1057')
pylab.plot(x1,y1,color1_dataAndFit, label='fit')
pylab.axvline(AL_Chalf,color=color1_Chalf,ls='-',label='C1/2')#chalf
if CI_ratioTOrange<CI_cutoff:
plt.axvspan((AL_Chalf-AL_fit_result_output[8]), (AL_Chalf+AL_fit_result_output[8]),color=color1_CI)#CI
if plotColor=="2":#choose from blue, orange, purple, brown
plt.plot(xplot1, yplot1,color=color2_dataAndFit,ls=':',marker='o',label='rep1_'+rep1_notebookCode)
#plt.plot(xplot2, yplot2,color=color2_dataAndFit,ls=':',marker='^',label='AL_1057')
pylab.plot(x1,y1,color2_dataAndFit, label='fit')
pylab.axvline(AL_Chalf,color=color2_Chalf,ls='-',label='C1/2')#chalf
if CI_ratioTOrange<CI_cutoff:
plt.axvspan((AL_Chalf-AL_fit_result_output[8]), (AL_Chalf+AL_fit_result_output[8]),color=color2_CI)#CI
if plotColor=="3":#choose from blue, orange, purple, brown
plt.plot(xplot1, yplot1,color=color3_dataAndFit,ls=':',marker='o',label='rep1_'+rep1_notebookCode)
#plt.plot(xplot2, yplot2,color=color3_dataAndFit,ls=':',marker='^',label='AL_1057')
pylab.plot(x1,y1,color3_dataAndFit, label='fit')
pylab.axvline(AL_Chalf,color=color3_Chalf,ls='-',label='C1/2')#chalf
if CI_ratioTOrange<CI_cutoff:
plt.axvspan((AL_Chalf-AL_fit_result_output[8]), (AL_Chalf+AL_fit_result_output[8]),color=color3_CI)#CI
if plotColor=="4":#choose from blue, orange, purple, brown
plt.plot(xplot1, yplot1,color=color4_dataAndFit,ls=':',marker='o',label='rep1_'+rep1_notebookCode)
#plt.plot(xplot2, yplot2,color=color4_dataAndFit,ls=':',marker='^',label='AL_1057')
pylab.plot(x1,y1,color4_dataAndFit, label='fit')
pylab.axvline(AL_Chalf,color=color4_Chalf,ls='-',label='C1/2')#chalf
if CI_ratioTOrange<CI_cutoff:
plt.axvspan((AL_Chalf-AL_fit_result_output[8]), (AL_Chalf+AL_fit_result_output[8]),color=color4_CI)#CI
#plot label
if denatureMethod=="CD":
pylab.xlabel('denaturant concentration')
pylab.ylabel('% labeled')
if denatureMethod=="HD":
pylab.xlabel('normalized temperature')
pylab.ylabel('% soluble')
#pylab.ylim(0.3, 2.2)
pylab.legend(loc='best')
pylab.title(graphName,fontsize=10)
pylab.savefig(outImage_location+title+"_"+proteinName+".jpg")
pylab.clf()#clearfig, start a new piece , make sure do it before or after a set of instruction
except :#define the error
print (title)
finalData.append(tempRow)
#output the data file
outfile=open(outfile_locationAndName,"w",newline="")
writer=csv.writer(outfile)#create write object
newheader=["row","protein ID","ProteinNAme","Peptide","numOfReporter","numOf_rep1_pts","sumPoints"]
newheader_AL_fitOutPut_CD=["B","A","Chalf","b","B_err","A_err","Chalf_err","b_err","CI_Chalf","CIratioTOrange","CI_Chalf_low","CI_Chalf_up","CI_b","CI_b_low","CI_b_up","r_square","Chalf_normalized"]
newheader_AL_fitOutPut_HD=["B","A","Chalf","b","B_err","A_err","Chalf_err","b_err","CI_Chalf","CIratioTOrange","CI_Chalf_low","CI_Chalf_up","CI_b","CI_b_low","CI_b_up","r_square","Chalf_temp"]
if denatureMethod=="CD":
newheader.extend(newheader_AL_fitOutPut_CD)
normalized_header1=CD_xaxis_FinalGdmClConcentration_value
newheader.extend(normalized_header1)
if denatureMethod=="HD":
newheader.extend(newheader_AL_fitOutPut_HD)
normalized_header1=HD_xasis_normalizedTemp_value
newheader.extend(normalized_header1)
writer.writerow(newheader)
for addvariable in finalData:
writer.writerow(addvariable)
outfile.close()
infile.close()
答案 0 :(得分：2)
要执行此操作，可以使用下面的“驱动程序”脚本，但是要使用它，还需要修改数据分析脚本，以便它接受输入和输出文件名作为命令行参数。
还要注意如何在文件位置和名称中添加r前缀。这是因为Windows将反斜杠\字符用作路径组件分隔符，但是在Python字符串文字中使用时，它具有特殊的含义。前缀可防止这种解释，并将其视为普通字符-在这种情况下，这是需要的。
或者，您也可以将它们全部更改为正斜杠/字符，这些字符不需要添加r前缀，因为它们在Python脚本中也可以用作路径分隔符。
无论如何，这是驱动程序脚本（如果它支持Python脚本的执行，则可以从编辑器中执行）：
import subprocess
import sys
analysis_script = 'data_analysis.py'
file_list = [
[r"G:\BYU OneDrive_G\OneDrive - BYU Office 365\BYU\===Price lab===\Coding\\1023-CD+HD +multi rep\\1023-input_python_test.csv",
r"G:\BYU OneDrive_G\OneDrive - BYU Office 365\BYU\===Price lab===\Coding\\1023-CD+HD +multi rep\\1026-CDtest_output.csv"
],
[r"G:\BYU OneDrive_G\OneDrive - BYU Office 365\BYU\===Price lab===\Coding\\1023-CD+HD +multi rep\\1036-input_python_test.csv",
r"G:\BYU OneDrive_G\OneDrive - BYU Office 365\BYU\===Price lab===\Coding\\1023-CD+HD +multi rep\\1042-CDtest_output.csv"
],
# etc ...
]
for infile, outfile in file_list:
# Note: It's important to put double quotes around everything in case they
# have embedded space characters in them.
command = '"%s" "%s" "%s" "%s"' % (sys.executable, # command
analysis_script, # argv[0]
infile, # argv[1]
outfile) # argv[2]
subprocess.Popen(command) # Run script and pass it the files.
print('Done')
要使其正常工作，您还需要在数据分析脚本的开头附近添加以下内容，并删除当前使用的硬编码输入和输出文件名。
下面是一个需要在开头添加的示例：
import sys
try:
infile_locationAndName = sys.argv[1]
outfile_locationAndName = sys.argv[2]
except IndexError:
print('ERROR: wrong number of command-line arguments passed')
print(' Usage: "{}" <input file name> <output file name>'.format(
os.path.basename(__file__)))
raise RuntimeError('Missing two required command-line arguments')
# Rest of the (modified) data analysis script ...
答案 1 :(得分：1)
如果您使用的是Linux或Mac，则可以使用bash脚本。例如在.sh文件中：
python script1.py
python script2.py
...
然后在bash yourfile.sh终端上运行
答案 2 :(得分：0)
import os
os.system(python3 <file> <arg1> <arg2> <arg3>)
第一个脚本要等到最后一个脚本执行完毕后才能关闭，但这应该不是问题，因为它是单线程的。