17 lines
490 B
Python
17 lines
490 B
Python
import pandas as pd
|
|
|
|
df = pd.read_csv('new_final_dataset.csv')
|
|
|
|
df_pos = df[df['Sentiment'] == 'positif']
|
|
df_neg = df[df['Sentiment'] == 'negatif']
|
|
df_net = df[df['Sentiment'] == 'netral']
|
|
|
|
target_count = len(df_neg) + len(df_net)
|
|
|
|
df_pos_trimmed = df_pos.sample(n=target_count, random_state=42)
|
|
|
|
df_final = pd.concat([df_pos_trimmed, df_neg, df_net])
|
|
|
|
df_final = df_final.sample(frac=1, random_state=42).reset_index(drop=True)
|
|
|
|
df_final.to_csv('trimmed_sentiment_dataset.csv', index=False) |