File size: 1,383 Bytes
c49b21b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 |
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# Load normalized stock data
csv_path = 'data\\merged\\norm\\stocks_features_improved_normalized.csv'
df = pd.read_csv(csv_path)
# 1. Show basic info and head
print('Data shape:', df.shape)
print(df.head())
# 2. Feature distribution histograms
features = [
'price_momentum', 'volume_price_ratio', 'daily_range', 'avg_sentiment', 'technical_strength'
]
existing_features = [f for f in features if f in df.columns]
if existing_features:
df[existing_features].hist(bins=30, figsize=(12, 8))
plt.suptitle('Feature Distributions')
plt.tight_layout()
plt.show()
else:
print('No engineered features found for distribution plots.')
# 3. Correlation heatmap
if len(existing_features) > 1:
plt.figure(figsize=(8, 6))
sns.heatmap(df[existing_features].corr(), annot=True, cmap='coolwarm')
plt.title('Feature Correlation Heatmap')
plt.show()
# 4. Outlier boxplots for engineered features
for feat in existing_features:
plt.figure(figsize=(6, 2))
sns.boxplot(x=df[feat])
plt.title(f'Boxplot: {feat}')
plt.show()
# 5. Pairplot (if you have a target column, e.g., "target")
# Uncomment and adjust if you have a target/label
# sns.pairplot(df, vars=existing_features, hue='target')
# plt.show()
print('Visualization complete. You can add more plots as needed!')
|