python How to Read. mtx Files
- 2021-11-01 03:59:28
- OfStack
The mtx file is matrix data stored in a sparse matrix format and can be read by following these steps:
1. Install the scanpy package
pip install scanpy
2. File reading
import scanpy as sc
adata = sc.read(filename)
data = adata.X
After line 1 read, annData is returned, and line 2 obtains matrix data through. X operation
3. Convert to dense matrix
data = data.todense()
The matrix obtained directly is sparse and can be transformed into dense matrix by todense function
Supplement: python reads various files
Json:
use_time=[]
with open(address,'r') as f: #ubuntu
mobile = json.load(f)
calls = mobile["transactions"][0]["calls"]
for call in calls:
use_time.append(str(call['use_time']))
Excel:
rawdata1=open_workbook(address)
rawdata=rawdata1.sheet_by_index(0)
for i in range(1,rawdata.nrows):
if rawdata.cell(i,date_index).value=="": # Skip blank lines
continue
else:
if ctype==3: # If yes 3 , then use datetime Module processing date
date1=rawdata.cell(i,date_index).value
date2 = xldate_as_tuple(date1,0)
date3=datetime(*date2)
if "." in str(rawdata.cell(i,phone_index).value):
phone1=str(rawdata.cell(i,phone_index).value)[:-2]
else:
phone1=str(rawdata.cell(i,phone_index).value)
Write EXCEL:
Excel_file = xlwt.Workbook()
sheet = Excel_file.add_sheet('sheet0')
header=[u' Number ',' Date top1',' Date top2',' Date top3']
# Write header row:
for i in range(len(header)):
sheet.write(0,i,header[i])
# Start writing data by row:
for i in range(len(phonelist)):
sheet.write(i+1,0,phonelist[i])
sheet.write(i+1,1,dic[str(phonelist[i])])
# Save EXCEL :
Excel_file.save("C:/Users/Desktop/100 File output xls/"+str(fileName)+".xls")
CSV:
rawdata=pd.read_csv(address,skip_blank_lines=True) # Parameter is to remove blank lines
if 'start_time' or 'begin_time' in rawdata.columns:
if 'start_time' in rawdata.columns:
start_time=rawdata['start_time']
elif 'begin_time' in rawdata.columns:
start_time=rawdata['begin_time']
txt:
rawdata=open(address,'r')
i=0
a=[] #c Deposit number 1 Column name of row
for line in rawdata:
if i==1: # Default number 2 Row starts to store call data
a=line.split(',') # Comma as separator
for j in range(len(a)): # Finds the column subscript of the specified column name
if (('-' in str(a[j]))or('/' in str(a[j]))): # Determine the number of columns in which the date is located
date_index=j # Column subscript of save date
elif str(a[j]).isdigit() and len(str(a[j]))>5: # The default string consisting of all numbers is the phone number
phone_index=j
else:
pass
break
else:
i+=1
i=0
for line in rawdata:# Begin to dump data:
if len(line)<10: # Skip blank lines
continue
data_line=line.split(',') #txt Default to ',' Separated data
if i==0:
pass # No. 1 1 Behavior column name, skipping
i+=1
else: # From the first 2 Row starts to save data
start_time.append(data_line[date_index])