في هذا الكود عند تشغيل على بيئة pycharm يعطي نتائج مختلفة في العمود engtype وأظن أنها ليست صحيحة
وعند تشغيل هذا العمود في بيئة Anaconda يعطي نتائج صحيحة ؟
import pandas as pd import matplotlib.pyplot as plt import seaborn as sns df = pd.read_csv('car_sales_new.csv') print(df.head()) print(df.info()) print(df.isnull().sum()) percent = (df.isnull().sum()/df.isnull().count()).sort_values(ascending=False) print(percent) print(df.describe()) print("How many value are 0: ", df.price[df.price == 0].count()) print("How many value are 0: ", df.mileage[df.mileage == 0].count()) df.columns = map(str.lower, df.columns) print(df.dtypes) print(df.columns) df['drive'] = df['drive'].fillna('Unspecified') print(df.drive.unique()) df['engv'].fillna(df['engv'].mean(), inplace=True) df['price'] = df['price'].replace(0, df['price'].mean()) millage_avg = sum(df['mileage'] / len(df['mileage'])) df['mileage_level'] = [ "high mileage" if i > millage_avg else "low mileage" for i in df['mileage']] print(df.loc[:10]) df['year'].value_counts().head(20).plot.bar() plt.show() df['body'].value_counts().head(20).plot.bar() plt.show() df['car'].value_counts().head(20).plot.bar() plt.show() usa = df.loc[df['car'] == 'Volkswagen'] top_10_model = usa['model'].value_counts()[:10].to_frame('count').reset_index() plt.figure(figsize=(15, 5)) sns.barplot(x=top_10_model['model'], y=top_10_model.index, palette='PuBuGn_d') plt.title("Top 10 Volkswagen model in terms of contribution", fontsize=18, fontweight="bold") plt.xlabel('') plt.show() usa = df.loc[df['car'] == 'Volkswagen'] top_10_model = usa['engtype'].value_counts()[:10].to_frame('count').reset_index() sns.barplot(x = top_10_model['engtype'], y = top_10_model.index, palette='PuBuGn_d') plt.title('Top 4') plt.xlabel('') plt.show() usa = df.loc[df['car'] == 'Mercedes-Benz'] top_10_model = usa['model'].value_counts()[:10].to_frame('count').reset_index() plt.figure(figsize=(15, 5)) sns.barplot(x=top_10_model['model'], y=top_10_model.index, palette='PuBuGn_d') plt.title("Top 10 Mercedes-Benz model in terms of contribution", fontsize=18, fontweight="bold") plt.xlabel('') plt.show() usa = df.loc[df['car'] == 'Mercedes-Benz'] top_10_model = usa['engtype'].value_counts()[:10].to_frame('count').reset_index() sns.barplot(x = top_10_model['engtype'], y = top_10_model.index, palette='PuBuGn_d') plt.title('Top 4') plt.xlabel('') plt.show() ct = pd.value_counts(df['engtype'].values, sort=False) labels = df['engtype'].unique() labels.sort() sizes = ct ct.sort_index(inplace=True) colors = ['red', 'pink', 'green', 'blue'] plt.pie(sizes, labels=labels, colors=colors, autopct='%1.1f%%', shadow=True, startangle=140) plt.show()
هل المشكلة في بيئة العمل أم الكود ؟