import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import LabelEncoder
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix
import pickle
import statsmodels.api as sm
import matplotlib.pyplot as plt
import seaborn as sns
import os

curdir = os.getcwd()
path_delim = '/'
data = pd.read_excel(curdir + path_delim + 'data/tests_24_10_2023_20_53_cleaned.xlsx', sheet_name="data")
data = data.dropna()
print(data)

data.plot()

label_encoder = LabelEncoder()
categorical_cols = ['calculatedRatio', 'led1Buffer', 'led1Sample', 'led2Buffer', 'led2Sample']
# for col in categorical_cols:
#     data[col] = label_encoder.fit_transform(data[col])

X = data[['calculatedRatio', 'led1Buffer', 'led1Sample', 'led2Buffer', 'led2Sample']]
y = data['classificationResult']

print(X)
corr = X.corr()
print(corr)
sm.graphics.plot_corr(corr, xnames=list(corr.columns))
plt.show()

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Change the classifier to KNeighborsClassifier
model = KNeighborsClassifier(n_neighbors=5)  # You can adjust the number of neighbors as needed
model.fit(X_train, y_train)

y_pred = model.predict(X_test)

accuracy = accuracy_score(y_test, y_pred)
classification_report_result = classification_report(y_test, y_pred)

# Calculate the confusion matrix
confusion = confusion_matrix(y_test, y_pred)

# Plot the confusion matrix using Seaborn
plt.figure(figsize=(8, 6))
sns.heatmap(confusion, annot=True, fmt='d', cmap='Blues', linewidths=0.5)
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.title('Confusion Matrix')
# plt.show()

print(f"Accuracy: {accuracy}")
print("Classification Report:")
print(classification_report_result)

# Accuracy: 0.7502027575020276
# Classification Report:
#                                                                           precision    recall  f1-score   support

#                Inconclusive. Repeat with test with lower volume of blood       0.50      0.17      0.25         6
# Inconclusive. Very low Absorbance - Repeat test with Higher Blood Volume       0.50      0.48      0.49        21
#                                         Negative Borderline. Repeat Test       0.51      0.51      0.51       167
#                                                                   Normal       0.85      0.92      0.88       749
#                          Positive for Sickle Cell. HPLC for Confirmation       0.21      0.17      0.19        41
#                                                      Sickle Cell Disease       0.67      0.32      0.43        38
#                                                        Sickle Cell Trait       0.66      0.55      0.60       211

#                                                                 accuracy                           0.75      1233
#                                                                macro avg       0.56      0.45      0.48      1233
#                                                             weighted avg       0.74      0.75      0.74      1233

# Save the KNeighborsClassifier model
with open('kneighbors_classifier_model.pkl', 'wb') as model_file:
    pickle.dump(model, model_file)

with open('kneighbors_classifier_model.pkl', 'rb') as model_file:
    loaded_model = pickle.load(model_file)

# Define new_data as needed for prediction
# new_data = pd.DataFrame({'calculatedRatio': [0.17500836], 'led1Buffer': [24843.33], 'led1Sample': [19678], 'led2Buffer': [26715.33], 'led2Sample': [13842.67]})
# new_data = pd.DataFrame({'calculatedRatio': [0.231057205], 'led1Buffer': [23776.33], 'led1Sample': [16286], 'led2Buffer': [26401.67], 'led2Sample': [6952.67]})
new_data = pd.DataFrame({'calculatedRatio': [0.175881142], 'led1Buffer': [24256], 'led1Sample': [16303], 'led2Buffer': [27016.33], 'led2Sample': [4612.67]})
# new_data = new_data.apply(lambda col: label_encoder.transform(col))
print(new_data)
predicted_result = loaded_model.predict(new_data)
print(predicted_result)