visu.py
No OneTemporary
Actions

Subscribers

None

File Metadata

	import matplotlib.pyplot as plt
	import seaborn as sns
	from sklearn.inspection import permutation_importance


	def plot_feature_importance(model, X_train, y_train, feature_names, output_path):
	result = permutation_importance(model, X_train, y_train, n_repeats=10, random_state=42)
	sorted_idx = result.importances_mean.argsort()

	plt.figure(figsize=(12, 8))
	plt.barh(range(len(sorted_idx)), result.importances_mean[sorted_idx], align='center')
	plt.yticks(range(len(sorted_idx)), [feature_names[i] for i in sorted_idx])
	plt.xlabel("Permutation Importance")
	plt.title("Feature Importance")
	plt.savefig(output_path)
	plt.close()


	def plot_feature_correlation(dataframe, output_path):
	plt.figure(figsize=(12, 10))
	corr_matrix = dataframe.corr()
	sns.heatmap(corr_matrix, annot=True, fmt=".2f", cmap="coolwarm")
	plt.title("Feature Correlation Matrix")
	plt.savefig(output_path)
	plt.close()


	def plot_predicted_counts(predictions, df, output_path):
	df['predictions'] = predictions
	result = df.groupby('antibiotics_quantity')['predictions'].value_counts().unstack().fillna(0)

	# Plotting side-by-side bars
	result.plot(kind='bar', color=['orange', 'blue'], width=0.8)
	plt.xlabel('Antibiotics Quantity')
	plt.ylabel('Count')
	plt.title('Predicted Dead/Alive Bacteria per Antibiotic Level')
	plt.legend(title='Predictions', labels=['Dead', 'Alive'])
	plt.savefig(output_path)
	plt.close()