Last active
August 22, 2019 13:31
-
-
Save fulibacsi/158d4d41a1fb77183d6133a169398caa to your computer and use it in GitHub Desktop.
Generate random pandas DataFrame
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import datetime | |
import numpy as np | |
import pandas as pd | |
from pandas.util.testing import makeIntIndex, rands_array | |
class RandomDataFrame: | |
def __init__(self, size, nintcols=1, nfloatcols=0, | |
nstringcols=0, ndatecols=0): | |
self.size = size | |
self.nintcols = nintcols | |
self.maxintvalue = 5000 | |
self.nfloatcols = nfloatcols | |
self.nstringcols = nstringcols | |
self.nchars = 10 | |
self.ndatecols = ndatecols | |
def generate(self): | |
index = makeIntIndex(self.size) | |
istr = 'int_{}' | |
int_cols = {istr.format(i): self.int_col(index, istr.format(i)) | |
for i in range(self.nintcols)} | |
fstr = 'float_{}' | |
float_cols = {fstr.format(i): self.float_col(index, fstr.format(i)) | |
for i in range(self.nfloatcols)} | |
sstr = 'str_{}' | |
string_cols = {sstr.format(i): self.str_col(index, sstr.format(i)) | |
for i in range(self.nstringcols)} | |
dstr = 'date_{}' | |
date_cols = {dstr.format(i): self.date_col(index, dstr.format(i)) | |
for i in range(self.ndatecols)} | |
columns = {} | |
for cols in [int_cols, float_cols, string_cols, date_cols]: | |
columns.update(cols) | |
return pd.DataFrame(columns, index=index) | |
def int_col(self, index, name=None): | |
return pd.Series(np.random.randint(self.maxintvalue, size=self.size), | |
index=index, name=name) | |
def float_col(self, index, name=None): | |
return pd.Series(np.random.rand(self.size), index=index, name=name) | |
def str_col(self, index, name=None): | |
return pd.Series(rands_array(self.nchars, self.size), | |
index=index, name=name) | |
def date_col(self, index, name=None): | |
year = np.random.randint(1980, 2020, self.size) | |
month = np.random.randint(1, 12, self.size) | |
day = np.random.randint(1, 28, self.size) | |
dt = pd.to_datetime({'year': year, 'month': month, 'day': day}) | |
return pd.Series(dt, index=index, name=name) | |
def random_df(size, nintcols=1, nfloatcols=0, nstringcols=0, ndatecols=0): | |
rdf = RandomDataFrame(size, nintcols, nfloatcols, nstringcols, ndatecols) | |
return rdf.generate() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment