pages/6_Logistic_Regression.py

import streamlit as st
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import accuracy_score, confusion_matrix
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense
from tensorflow.keras.utils import plot_model
import io
from PIL import Image

# Load Iris dataset
iris = load_iris()
X = iris.data
y = iris.target

# Only use the first two classes for binary classification
X = X[y != 2]
y = y[y != 2]

# Split the dataset into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Standardize the data
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

# Streamlit interface
st.title('Logistic Regression with Keras on Iris Dataset')
st.write("""
## Introduction
Logistic Regression is a statistical model used for binary classification tasks. 
In this tutorial, we will use the Iris dataset to classify whether a flower is 
**Setosa** or **Versicolor** based on its features.
""")

# Display Iris dataset information
st.write("### Iris Dataset")
st.write("""
The Iris dataset contains 150 samples of iris flowers, each described by four features: 
sepal length, sepal width, petal length, and petal width. There are three classes: Setosa, Versicolor, and Virginica.
For this example, we'll only use the Setosa and Versicolor classes.
""")

# Display flower images
st.write("### Flower Images")
col1, col2 = st.columns(2)
with col1:
    st.image("https://upload.wikimedia.org/wikipedia/commons/thumb/a/a7/Irissetosa1.jpg/1280px-Irissetosa1.jpg", caption="Iris Setosa", use_column_width=True)
with col2:
    st.image("https://upload.wikimedia.org/wikipedia/commons/4/41/Iris_versicolor_3.jpg", caption="Iris Versicolor", use_column_width=True)

# Plotting sample data
st.write("### Sample Data Distribution")
fig, ax = plt.subplots()
for i, color in zip([0, 1], ['blue', 'orange']):
    idx = np.where(y == i)
    ax.scatter(X[idx, 0], X[idx, 1], c=color, label=iris.target_names[i], edgecolor='k')
ax.set_xlabel(iris.feature_names[0])
ax.set_ylabel(iris.feature_names[1])
ax.legend()
st.pyplot(fig)

# User input for number of epochs
epochs = st.slider('Select number of epochs for training:', min_value=10, max_value=200, value=100, step=10)

# Build the logistic regression model using Keras
model = Sequential()
model.add(Dense(1, input_dim=4, activation='sigmoid'))
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

# Display the model architecture
st.write("### Model Architecture")
st.write(model.summary())
try:
    fig, ax = plt.subplots()
    buf = io.BytesIO()
    plot_model(model, to_file=buf, show_shapes=True, show_layer_names=True)
    buf.seek(0)
    st.image(buf, caption='Logistic Regression Model Architecture', use_column_width=True)
except ImportError:
    st.warning('Graphviz is not installed. Model architecture visualization is skipped.')

# Train the model
model.fit(X_train, y_train, epochs=epochs, verbose=0)

# Predict and evaluate the model
y_pred_train = (model.predict(X_train) > 0.5).astype("int32")
y_pred_test = (model.predict(X_test) > 0.5).astype("int32")

train_accuracy = accuracy_score(y_train, y_pred_train)
test_accuracy = accuracy_score(y_test, y_pred_test)

conf_matrix = confusion_matrix(y_test, y_pred_test)

st.write('## Model Performance')
st.write(f'Training Accuracy: {train_accuracy:.2f}')
st.write(f'Testing Accuracy: {test_accuracy:.2f}')

st.write('## Confusion Matrix')
fig, ax = plt.subplots()
ax.matshow(conf_matrix, cmap=plt.cm.Blues, alpha=0.3)
for i in range(conf_matrix.shape[0]):
    for j in range(conf_matrix.shape[1]):
        ax.text(x=j, y=i, s=conf_matrix[i, j], va='center', ha='center')

plt.xlabel('Predicted Label')
plt.ylabel('True Label')
st.pyplot(fig)

st.write('## Make a Prediction')
sepal_length = st.number_input('Sepal Length (cm)', min_value=0.0, max_value=10.0, value=5.0)
sepal_width = st.number_input('Sepal Width (cm)', min_value=0.0, max_value=10.0, value=3.5)
petal_length = st.number_input('Petal Length (cm)', min_value=0.0, max_value=10.0, value=1.4)
petal_width = st.number_input('Petal Width (cm)', min_value=0.0, max_value=10.0, value=0.2)

if st.button('Predict'):
    input_data = np.array([[sepal_length, sepal_width, petal_length, petal_width]])
    input_data_scaled = scaler.transform(input_data)
    prediction = (model.predict(input_data_scaled) > 0.5).astype("int32")
    st.write(f'Prediction: {"Setosa" if prediction[0][0] == 0 else "Versicolor"}')

# Examples of different parameters for each flower type
st.write('## Examples of Parameters')
st.write("""
### Iris Setosa:
- Sepal Length: 5.1 cm
- Sepal Width: 3.5 cm
- Petal Length: 1.4 cm
- Petal Width: 0.2 cm

### Iris Versicolor:
- Sepal Length: 7.0 cm
- Sepal Width: 3.2 cm
- Petal Length: 4.7 cm
- Petal Width: 1.4 cm
""")