app.py

from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
import torch
import pandas as pd
import numpy as np
from sentence_transformers import SentenceTransformer
from torch_geometric.data import Data
from torch_geometric.nn import GATConv
from sklearn.metrics.pairwise import cosine_similarity

# FastAPI App
app = FastAPI()

# Data and Model Initialization
data = torch.load("graph_data.pt", map_location=torch.device("cpu"))

# Corrected state dictionary for GATConv model
original_state_dict = torch.load("graph_model.pth", map_location=torch.device("cpu"))
corrected_state_dict = {}
for key, value in original_state_dict.items():
    if "lin.weight" in key:
        corrected_state_dict[key.replace("lin.weight", "lin_src.weight")] = value
        corrected_state_dict[key.replace("lin.weight", "lin_dst.weight")] = value
    else:
        corrected_state_dict[key] = value

# Define GATConv Model
class ModeratelySimplifiedGATConvModel(torch.nn.Module):
    def __init__(self, in_channels, hidden_channels, out_channels):
        super().__init__()
        self.conv1 = GATConv(in_channels, hidden_channels, heads=2)
        self.dropout1 = torch.nn.Dropout(0.45)
        self.conv2 = GATConv(hidden_channels * 2, out_channels, heads=1)

    def forward(self, x, edge_index, edge_attr=None):
        x = self.conv1(x, edge_index, edge_attr)
        x = torch.relu(x)
        x = self.dropout1(x)
        x = self.conv2(x, edge_index, edge_attr)
        return x

# Initialize GATConv model and BERT-based sentence transformer model
gatconv_model = ModeratelySimplifiedGATConvModel(
    in_channels=data.x.shape[1], hidden_channels=32, out_channels=768
)
gatconv_model.load_state_dict(corrected_state_dict)

model_bert = SentenceTransformer("all-mpnet-base-v2")

# Ensure DataFrame is loaded properly
df = pd.read_feather("EmbeddedCombined.feather")

# Function to get most similar video and recommend top 10 based on GNN embeddings
def get_similar_and_recommend(input_text):
    embeddings_matrix = np.array(df["embeddings"].tolist())
    input_embedding = model_bert.encode([input_text])[0]
    similarities = cosine_similarity([input_embedding], embeddings_matrix)[0]
    most_similar_index = np.argmax(similarities)

    most_similar_video = {
        "title": df["title"].iloc[most_similar_index],
        "description": df["description"].iloc[most_similar_index],
        "similarity_score": similarities[most_similar_index],
    }

    # Function to recommend top 10 videos based on GNN embeddings
    def recommend_next_10_videos(given_video_index, all_video_embeddings):
        dot_products = [
            torch.dot(all_video_embeddings[given_video_index].cpu(), all_video_embeddings[i].cpu())
            for i in range(all_video_embeddings.shape[0])
        ]
        dot_products[given_video_index] = -float("inf")

        top_10_indices = np.argsort(dot_products)[::-1][:10]
        recommendations = [df["title"].iloc[idx] for idx in top_10_indices]
        return recommendations

    top_10_recommendations = recommend_next_10_videos(
        most_similar_index, gatconv_model(data.x, data.edge_index, data.edge_attr).cpu()
    )

    return {
        "most_similar_video_title": most_similar_video["title"],
        "top_10_recommendations": top_10_recommendations,
    }

# Define the endpoint for FastAPI to get video title and recommendations
class UserInput(BaseModel):
    text: str  # The string input from the user

@app.post("/recommendations")
def recommend_videos(user_input: UserInput):
    if not user_input.text:
        raise HTTPException(status_code=400, detail="Input text cannot be empty.")
    
    result = get_similar_and_recommend(user_input.text)
    
    return result