Real-World Applications

# Medical image classification with TensorFlow
import tensorflow as tf
from tensorflow.keras.applications import ResNet50
from tensorflow.keras.layers import Dense, GlobalAveragePooling2D

# Load pre-trained ResNet50 model
base_model = ResNet50(
    weights='imagenet',
    include_top=False,
    input_shape=(224, 224, 3)
)

# Add custom classification layers
model = tf.keras.Sequential([
    base_model,
    GlobalAveragePooling2D(),
    Dense(128, activation='relu'),
    tf.keras.layers.Dropout(0.5),
    Dense(2, activation='softmax')  # healthy vs disease
])

# Compile the model
model.compile(
    optimizer='adam',
    loss='categorical_crossentropy',
    metrics=['accuracy']
)

# Basic chatbot using Hugging Face Transformers
from transformers import pipeline, Conversation

# Initialize conversational pipeline
chatbot = pipeline("conversational")

# Create a conversation
conversation = Conversation("Hello! How can I help you today?")

# Function to chat with the bot
def chat_with_bot(user_input):
    conversation.add_user_input(user_input)
    response = chatbot(conversation)
    return response.generated_responses[-1]

# Example usage
response = chat_with_bot("What is machine learning?")
print(response)

# Simple collaborative filtering recommendation system
import numpy as np
from sklearn.metrics.pairwise import cosine_similarity

class CollaborativeFilter:
    def __init__(self, user_item_matrix):
        self.matrix = user_item_matrix
        self.user_similarity = cosine_similarity(user_item_matrix)
    
    def recommend(self, user_id, n_recommendations=5):
        # Find similar users
        similar_users = self.user_similarity[user_id].argsort()[-6:-1]
        
        # Get recommendations based on similar users
        recommendations = []
        for similar_user in similar_users:
            user_items = self.matrix[similar_user]
            # Add items not yet rated by target user
            for item_idx, rating in enumerate(user_items):
                if rating > 0 and self.matrix[user_id][item_idx] == 0:
                    recommendations.append((item_idx, rating))
        
        # Sort by rating and return top N
        recommendations.sort(key=lambda x: x[1], reverse=True)
        return recommendations[:n_recommendations]