ium_151636/better_preprocess.py

24 lines
741 B
Python
Raw Normal View History

2023-05-14 19:44:22 +02:00
import pandas as pd
import numpy as np
df=pd.read_csv("25k_IMDb_movie_Dataset.csv")
df=df.drop(columns=['Run Time'])
df['Rating']=df['Rating'].str.replace("no-rating","0")
df['Rating']=pd.to_numeric(df['Rating'])
df['User Rating']=df['User Rating'].str.replace("K","000")
df['User Rating']=df['User Rating'].str.replace("M","000000")
df['User Rating']=pd.to_numeric(df['User Rating'])
df=df.drop("path",axis=1)
df['year']=df['year'].str.replace("-","")
mumbojumbo=["(",")","I","V"," ","X"]
for a in mumbojumbo:
df['year']=df['year'].str.replace(a,"")
df['year']=np.where(df['year']=="","1900",df['year'])
df['year']=df['year'].fillna("1900")
df['year']=df['year']+"-01-01"
df['year']=pd.to_datetime(df['year'])
df.to_csv("data.csv")