app.py

import gradio as gr
from transformers import FlaxAutoModelForSeq2SeqLM, AutoTokenizer, AutoModel
import torch
import numpy as np
import random
import json
import os

# Disable Gradio analytics for better performance
os.environ["GRADIO_ANALYTICS_ENABLED"] = "False"

# Load RecipeBERT model (for semantic ingredient combination)
bert_model_name = "alexdseo/RecipeBERT"
bert_tokenizer = AutoTokenizer.from_pretrained(bert_model_name)
bert_model = AutoModel.from_pretrained(bert_model_name)
bert_model.eval()

# Load T5 recipe generation model
MODEL_NAME_OR_PATH = "flax-community/t5-recipe-generation"
t5_tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME_OR_PATH, use_fast=True)
t5_model = FlaxAutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME_OR_PATH)

# Token mapping for T5 model output processing
special_tokens = t5_tokenizer.all_special_tokens
tokens_map = {
    "<sep>": "--",
    "<section>": "\n"
}


def get_embedding(text):
    """Computes embedding for a text with Mean Pooling over all tokens"""
    inputs = bert_tokenizer(text, return_tensors="pt", truncation=True, padding=True)
    with torch.no_grad():
        outputs = bert_model(**inputs)

    # Mean Pooling - take average of all token embeddings
    attention_mask = inputs['attention_mask']
    token_embeddings = outputs.last_hidden_state
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    sum_embeddings = torch.sum(token_embeddings * input_mask_expanded, 1)
    sum_mask = torch.clamp(input_mask_expanded.sum(1), min=1e-9)

    return (sum_embeddings / sum_mask).squeeze(0)


def average_embedding(embedding_list):
    """Computes the average of a list of embeddings"""
    tensors = torch.stack([emb for _, emb in embedding_list])
    return tensors.mean(dim=0)


def get_cosine_similarity(vec1, vec2):
    """Computes the cosine similarity between two vectors"""
    if torch.is_tensor(vec1):
        vec1 = vec1.detach().numpy()
    if torch.is_tensor(vec2):
        vec2 = vec2.detach().numpy()

    # Make sure vectors have the right shape (flatten if necessary)
    vec1 = vec1.flatten()
    vec2 = vec2.flatten()

    dot_product = np.dot(vec1, vec2)
    norm_a = np.linalg.norm(vec1)
    norm_b = np.linalg.norm(vec2)

    # Avoid division by zero
    if norm_a == 0 or norm_b == 0:
        return 0

    return dot_product / (norm_a * norm_b)


def get_combined_scores(query_vector, embedding_list, all_good_embeddings, avg_weight=0.6):
    """Computes combined score considering both similarity to average and individual ingredients"""
    results = []

    for name, emb in embedding_list:
        # Similarity to average vector
        avg_similarity = get_cosine_similarity(query_vector, emb)

        # Average similarity to individual ingredients
        individual_similarities = [get_cosine_similarity(good_emb, emb)
                                   for _, good_emb in all_good_embeddings]
        avg_individual_similarity = sum(individual_similarities) / len(individual_similarities)

        # Combined score (weighted average)
        combined_score = avg_weight * avg_similarity + (1 - avg_weight) * avg_individual_similarity

        results.append((name, emb, combined_score))

    # Sort by combined score (descending)
    results.sort(key=lambda x: x[2], reverse=True)
    return results


def find_best_ingredients(required_ingredients, available_ingredients, max_ingredients=6, avg_weight=0.6):
    """
    Finds the best ingredients based on RecipeBERT embeddings.
    """
    # Clean and prepare ingredient lists
    required_ingredients = [ing.strip() for ing in required_ingredients if ing.strip()]
    available_ingredients = [ing.strip() for ing in available_ingredients if ing.strip()]

    # Remove duplicates
    required_ingredients = list(set(required_ingredients))
    available_ingredients = list(set([i for i in available_ingredients if i not in required_ingredients]))

    # Special case: If no required ingredients, randomly select one from available ingredients
    if not required_ingredients and available_ingredients:
        random_ingredient = random.choice(available_ingredients)
        required_ingredients = [random_ingredient]
        available_ingredients = [i for i in available_ingredients if i != random_ingredient]

    # If still no ingredients or already at max capacity
    if not required_ingredients or len(required_ingredients) >= max_ingredients:
        return required_ingredients[:max_ingredients]

    # If no additional ingredients available
    if not available_ingredients:
        return required_ingredients

    # Calculate embeddings for all ingredients
    embed_required = [(e, get_embedding(e)) for e in required_ingredients]
    embed_available = [(e, get_embedding(e)) for e in available_ingredients]

    # Number of ingredients to add
    num_to_add = min(max_ingredients - len(required_ingredients), len(available_ingredients))

    # Copy required ingredients to final list
    final_ingredients = embed_required.copy()

    # Add best ingredients
    for _ in range(num_to_add):
        # Calculate average vector of current combination
        avg = average_embedding(final_ingredients)

        # Calculate combined scores for all candidates
        candidates = get_combined_scores(avg, embed_available, final_ingredients, avg_weight)

        # If no candidates left, break
        if not candidates:
            break

        # Choose best ingredient
        best_name, best_embedding, _ = candidates[0]

        # Add best ingredient to final list
        final_ingredients.append((best_name, best_embedding))

        # Remove ingredient from available ingredients
        embed_available = [item for item in embed_available if item[0] != best_name]

    # Extract only ingredient names
    return [name for name, _ in final_ingredients]


def skip_special_tokens(text, special_tokens):
    """Removes special tokens from text"""
    for token in special_tokens:
        text = text.replace(token, "")
    return text


def target_postprocessing(texts, special_tokens):
    """Post-processes generated text"""
    if not isinstance(texts, list):
        texts = [texts]

    new_texts = []
    for text in texts:
        text = skip_special_tokens(text, special_tokens)

        for k, v in tokens_map.items():
            text = text.replace(k, v)

        new_texts.append(text)

    return new_texts


def validate_recipe_ingredients(recipe_ingredients, expected_ingredients, tolerance=0):
    """Validates if the recipe contains approximately the expected ingredients."""
    recipe_count = len([ing for ing in recipe_ingredients if ing and ing.strip()])
    expected_count = len(expected_ingredients)
    return abs(recipe_count - expected_count) <= tolerance


def generate_recipe_with_t5(ingredients_list, max_retries=5):
    """Generates a recipe using the T5 recipe generation model with validation."""
    original_ingredients = ingredients_list.copy()

    for attempt in range(max_retries):
        try:
            # For retries after the first attempt, shuffle the ingredients
            if attempt > 0:
                current_ingredients = original_ingredients.copy()
                random.shuffle(current_ingredients)
            else:
                current_ingredients = ingredients_list

            # Format ingredients as a comma-separated string
            ingredients_string = ", ".join(current_ingredients)
            prefix = "items: "

            # Generation settings
            generation_kwargs = {
                "max_length": 512,
                "min_length": 64,
                "do_sample": True,
                "top_k": 60,
                "top_p": 0.95
            }

            # Tokenize input
            inputs = t5_tokenizer(
                prefix + ingredients_string,
                max_length=256,
                padding="max_length",
                truncation=True,
                return_tensors="jax"
            )

            # Generate text
            output_ids = t5_model.generate(
                input_ids=inputs.input_ids,
                attention_mask=inputs.attention_mask,
                **generation_kwargs
            )

            # Decode and post-process
            generated = output_ids.sequences
            generated_text = target_postprocessing(
                t5_tokenizer.batch_decode(generated, skip_special_tokens=False),
                special_tokens
            )[0]

            # Parse sections
            recipe = {}
            sections = generated_text.split("\n")
            for section in sections:
                section = section.strip()
                if section.startswith("title:"):
                    recipe["title"] = section.replace("title:", "").strip().capitalize()
                elif section.startswith("ingredients:"):
                    ingredients_text = section.replace("ingredients:", "").strip()
                    recipe["ingredients"] = [item.strip().capitalize() for item in ingredients_text.split("--") if
                                             item.strip()]
                elif section.startswith("directions:"):
                    directions_text = section.replace("directions:", "").strip()
                    recipe["directions"] = [step.strip().capitalize() for step in directions_text.split("--") if
                                            step.strip()]

            # If title is missing, create one
            if "title" not in recipe:
                recipe["title"] = f"Recipe with {', '.join(current_ingredients[:3])}"

            # Ensure all sections exist
            if "ingredients" not in recipe:
                recipe["ingredients"] = current_ingredients
            if "directions" not in recipe:
                recipe["directions"] = ["No directions generated"]

            # Validate the recipe
            if validate_recipe_ingredients(recipe["ingredients"], original_ingredients, tolerance=1):
                return recipe
            else:
                if attempt == max_retries - 1:
                    return recipe

        except Exception as e:
            if attempt == max_retries - 1:
                return {
                    "title": f"Recipe with {original_ingredients[0] if original_ingredients else 'ingredients'}",
                    "ingredients": original_ingredients,
                    "directions": ["Error generating recipe instructions"]
                }

    # Fallback
    return {
        "title": f"Recipe with {original_ingredients[0] if original_ingredients else 'ingredients'}",
        "ingredients": original_ingredients,
        "directions": ["Error generating recipe instructions"]
    }


def generate_recipe_interface(required_ingredients_text, available_ingredients_text, max_ingredients):
    """Main interface function for Gradio"""
    try:
        # Parse ingredient inputs
        required_ingredients = []
        available_ingredients = []

        if required_ingredients_text:
            required_ingredients = [ing.strip() for ing in required_ingredients_text.split(',') if ing.strip()]

        if available_ingredients_text:
            available_ingredients = [ing.strip() for ing in available_ingredients_text.split(',') if ing.strip()]

        # Validate inputs
        if not required_ingredients and not available_ingredients:
            return "❌ **Error:** Please provide at least some ingredients!", "", "", ""

        # Find best ingredient combination
        optimized_ingredients = find_best_ingredients(
            required_ingredients,
            available_ingredients,
            max_ingredients
        )

        # Generate recipe
        recipe = generate_recipe_with_t5(optimized_ingredients)

        # Format output
        title = f"🍽️ **{recipe['title']}**"

        ingredients_formatted = "## 📋 Ingredients:\n" + "\n".join([f"• {ing}" for ing in recipe['ingredients']])

        directions_formatted = "## 👨‍🍳 Instructions:\n" + "\n".join(
            [f"{i + 1}. {step}" for i, step in enumerate(recipe['directions'])])

        used_ingredients = "## ✅ Used Ingredients:\n" + ", ".join(optimized_ingredients)

        return title, ingredients_formatted, directions_formatted, used_ingredients

    except Exception as e:
        return f"❌ **Error:** {str(e)}", "", "", ""


def generate_recipe_api(required_ingredients_text, available_ingredients_text, max_ingredients):
    """API-compatible function that returns JSON format"""
    try:
        # Parse ingredient inputs
        required_ingredients = []
        available_ingredients = []

        if required_ingredients_text:
            required_ingredients = [ing.strip() for ing in required_ingredients_text.split(',') if ing.strip()]

        if available_ingredients_text:
            available_ingredients = [ing.strip() for ing in available_ingredients_text.split(',') if ing.strip()]

        # Validate inputs
        if not required_ingredients and not available_ingredients:
            return json.dumps({"error": "No ingredients provided"}, indent=2)

        # Find best ingredient combination
        optimized_ingredients = find_best_ingredients(
            required_ingredients,
            available_ingredients,
            max_ingredients
        )

        # Generate recipe
        recipe = generate_recipe_with_t5(optimized_ingredients)

        # Format for API response
        api_response = {
            'title': recipe['title'],
            'ingredients': recipe['ingredients'],
            'directions': recipe['directions'],
            'used_ingredients': optimized_ingredients
        }

        return json.dumps(api_response, indent=2, ensure_ascii=False)

    except Exception as e:
        return json.dumps({"error": f"Error in recipe generation: {str(e)}"}, indent=2)


# Create Gradio interface
with gr.Blocks(title="🍳 AI Recipe Generator", theme=gr.themes.Soft()) as demo:
    gr.Markdown("""
    # 🍳 AI Recipe Generator

    Generate delicious recipes using AI! This tool uses **RecipeBERT** to find the best ingredient combinations and **T5** to generate complete recipes.

    ## How to use:
    1. **Required Ingredients:** Enter ingredients you must use (comma-separated)  
    2. **Available Ingredients:** Enter additional ingredients you have available (comma-separated)  
    3. **Max Ingredients:** Set the maximum number of ingredients for your recipe  
    4. Click **Generate Recipe** to create your personalized recipe!
    """)

    with gr.Tab("🍽️ Recipe Generator"):
        with gr.Row():
            with gr.Column():
                required_ingredients = gr.Textbox(
                    label="🎯 Required Ingredients",
                    placeholder="chicken, rice, onions",
                    info="Ingredients that must be included in the recipe (comma-separated)"
                )
                available_ingredients = gr.Textbox(
                    label="🥕 Available Ingredients",
                    placeholder="garlic, tomatoes, basil, cheese",
                    info="Additional ingredients you have available (comma-separated)"
                )
                max_ingredients = gr.Slider(
                    minimum=3, maximum=12, value=7, step=1,
                    label="📊 Maximum Ingredients",
                    info="Maximum number of ingredients to use in the recipe"
                )
                generate_btn = gr.Button("🚀 Generate Recipe", variant="primary", size="lg")

            with gr.Column():
                recipe_title = gr.Markdown()
                used_ingredients = gr.Markdown()

        with gr.Row():
            with gr.Column():
                recipe_ingredients = gr.Markdown()
            with gr.Column():
                recipe_directions = gr.Markdown()

    with gr.Tab("🔌 API Format"):
        gr.Markdown("""
        ## API Response Format
        This tab shows the response in JSON format, compatible with your Flutter app.
        """)

        with gr.Row():
            with gr.Column():
                api_required = gr.Textbox(
                    label="Required Ingredients",
                    placeholder="chicken, rice, onions"
                )
                api_available = gr.Textbox(
                    label="Available Ingredients",
                    placeholder="garlic, tomatoes, basil"
                )
                api_max = gr.Slider(
                    minimum=3, maximum=12, value=7, step=1,
                    label="Max Ingredients"
                )
                api_generate_btn = gr.Button("Generate JSON", variant="secondary")

            with gr.Column():
                api_output = gr.Code(language="json", label="API Response")

    # Event handlers - WICHTIG: Diese erstellen die API-Endpunkte!
    generate_btn.click(
        fn=generate_recipe_interface,
        inputs=[required_ingredients, available_ingredients, max_ingredients],
        outputs=[recipe_title, recipe_ingredients, recipe_directions, used_ingredients]
    )

    api_generate_btn.click(
        fn=generate_recipe_api,
        inputs=[api_required, api_available, api_max],
        outputs=[api_output]
    )

    # Example inputs
    gr.Examples(
        examples=[
            ["chicken, rice", "onions, garlic, tomatoes, basil", 6],
            ["eggs, flour", "milk, sugar, vanilla, butter", 7],
            ["salmon", "lemon, dill, potatoes, asparagus", 5],
            ["", "beef, potatoes, carrots, onions, garlic", 6]
        ],
        inputs=[required_ingredients, available_ingredients, max_ingredients]
    )

# Launch the application
if __name__ == "__main__":
    demo.launch(
        server_name="0.0.0.0",
        server_port=7860,
        show_error=True
    )