Spaces:
Sleeping
Sleeping
Upload 3 files
Browse files- README.md +27 -6
- app1.py +41 -0
- requirements.txt +5 -0
README.md
CHANGED
@@ -1,13 +1,34 @@
|
|
1 |
---
|
2 |
-
title:
|
3 |
-
emoji:
|
4 |
-
colorFrom:
|
5 |
colorTo: indigo
|
6 |
sdk: gradio
|
7 |
-
sdk_version:
|
8 |
app_file: app.py
|
9 |
pinned: false
|
10 |
-
license: mit
|
11 |
---
|
12 |
|
13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
title: Movie Recommender
|
3 |
+
emoji: 馃幀
|
4 |
+
colorFrom: blue
|
5 |
colorTo: indigo
|
6 |
sdk: gradio
|
7 |
+
sdk_version: "4.16.0"
|
8 |
app_file: app.py
|
9 |
pinned: false
|
|
|
10 |
---
|
11 |
|
12 |
+
# Movie Recommender
|
13 |
+
|
14 |
+
A simple web app to recommend movies based on your text description, using embeddings and cosine similarity.
|
15 |
+
|
16 |
+
## How it works
|
17 |
+
|
18 |
+
- The app uses [SentenceTransformers](https://www.sbert.net/) to encode your text and the movie descriptions into embeddings.
|
19 |
+
- Then it calculates the cosine similarity between your input and all movies in the dataset.
|
20 |
+
- It shows the top 5 most similar movies.
|
21 |
+
|
22 |
+
## How to run
|
23 |
+
|
24 |
+
1. Install dependencies:
|
25 |
+
```bash
|
26 |
+
pip install -r requirements.txt
|
27 |
+
```
|
28 |
+
|
29 |
+
2. Run the app:
|
30 |
+
```bash
|
31 |
+
python app.py
|
32 |
+
```
|
33 |
+
|
34 |
+
Or just run it here on Spaces!
|
app1.py
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from sklearn.metrics.pairwise import cosine_similarity
|
2 |
+
import numpy as np
|
3 |
+
import gradio as gr
|
4 |
+
import pandas as pd
|
5 |
+
from sentence_transformers import SentenceTransformer
|
6 |
+
|
7 |
+
# 讟注谉 讗转 讛诪讜讚诇
|
8 |
+
model = SentenceTransformer("all-MiniLM-L6-v2")
|
9 |
+
|
10 |
+
# 讟注谉 讗转 讛讚讗讟讛住讟 诪讛诇讬谞拽
|
11 |
+
url = "https://huggingface.co/datasets/Pablinho/movies-dataset/resolve/main/9000plus.csv"
|
12 |
+
print("Loading dataset...")
|
13 |
+
dataset = pd.read_csv(url)
|
14 |
+
|
15 |
+
# 讜讚讗 砖讛注诪讜讚讜转 拽讬讬诪讜转
|
16 |
+
assert "Title" in dataset.columns
|
17 |
+
assert "Overview" in dataset.columns
|
18 |
+
|
19 |
+
print("Encoding movie descriptions...")
|
20 |
+
MAX_MOVIES = 1000
|
21 |
+
dataset = dataset.head(MAX_MOVIES)
|
22 |
+
dataset["embeddings"] = dataset["Overview"].apply(lambda x: model.encode(x).tolist())
|
23 |
+
print("Done encoding!")
|
24 |
+
|
25 |
+
def recommend_similar_movies(input_text, top_n=5):
|
26 |
+
input_embedding = model.encode([input_text])
|
27 |
+
similarities = cosine_similarity(input_embedding, np.vstack(dataset['embeddings'].to_numpy()))[0]
|
28 |
+
top_indices = similarities.argsort()[::-1][:top_n]
|
29 |
+
results = dataset.iloc[top_indices][['Title', 'Overview']]
|
30 |
+
return "\n\n".join(f"馃幀 **{row['Title']}**\n{row['Overview']}" for _, row in results.iterrows())
|
31 |
+
|
32 |
+
demo = gr.Interface(
|
33 |
+
fn=recommend_similar_movies,
|
34 |
+
inputs=gr.Textbox(lines=2, placeholder="Describe a movie..."),
|
35 |
+
outputs="text",
|
36 |
+
title="Movie Recommender",
|
37 |
+
description="Get movie recommendations based on your description. Powered by sentence-transformers and cosine similarity."
|
38 |
+
)
|
39 |
+
|
40 |
+
if __name__ == "__main__":
|
41 |
+
demo.launch()
|
requirements.txt
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
numpy
|
2 |
+
pandas
|
3 |
+
scikit-learn
|
4 |
+
gradio
|
5 |
+
sentence-transformers
|