|
1 | 1 | # vector-embedding-api |
2 | | -Flask API for generating text embeddings using OpenAI or sentence_transformers |
| 2 | +Flask server that allows users to generate text embeddings using OpenAI's "text-embedding-ada-002" model or the Sentence Transformers library. Text embeddings are dense vector representations of textual data that capture semantic information, making them useful for various natural language processing tasks. |
| 3 | + |
| 4 | +## Features 🎯 |
| 5 | +* POST endpoint access to text embedding models |
| 6 | + * sentence_transformers |
| 7 | + * OpenAI text-embedding-ada-002 |
| 8 | +* Easy setup with configuration file |
| 9 | +* Simple integration with other applications |
| 10 | + |
| 11 | +### Installation 💻 |
| 12 | +To run this Flask server locally, follow the steps below: |
| 13 | + |
| 14 | +**Clone the repository:** 📦 |
| 15 | +```bash |
| 16 | +git clone https://github.com/your-username/text-embeddings-server.git |
| 17 | +cd text-embeddings-server |
| 18 | +``` |
| 19 | + |
| 20 | +**Set up a virtual environment (optional but recommended):** 🐍 |
| 21 | +```bash |
| 22 | +python3 -m venv venv |
| 23 | +source venv/bin/activate |
| 24 | +``` |
| 25 | + |
| 26 | +**Install the required dependencies:** 🛠️ |
| 27 | +```bash |
| 28 | +pip install -r requirements.txt |
| 29 | +``` |
| 30 | + |
| 31 | + |
| 32 | +### Usage |
| 33 | +Before running the server, make sure you have obtained an API key from OpenAI to use their "text-embedding-ada-002" model. Additionally, you need to specify the Sentence Transformers model you want to use in the server.conf file. |
| 34 | + |
| 35 | +**Modify the server.conf configuration file:** ⚙️ |
| 36 | +```ini |
| 37 | +[main] |
| 38 | +openai_api_key = YOUR_OPENAI_API_KEY |
| 39 | +sent_transformers_model = YOUR_SENTENCE_TRANSFORMERS_MODEL_NAME |
| 40 | +``` |
| 41 | + |
| 42 | +**Start the Flask server:** 🚀 |
| 43 | +``` |
| 44 | +python server.py |
| 45 | +``` |
| 46 | + |
| 47 | +The server should now be running on http://127.0.0.1:5000/. |
| 48 | + |
| 49 | + |
| 50 | +### API Endpoints 🌐 |
| 51 | +The server provides the following endpoint: |
| 52 | + |
| 53 | +#### POST /submit |
| 54 | +Submit text to be converted to embeddings. |
| 55 | +The default is to use the sentence transformers model. |
| 56 | +Setting the `ada` field to True will use the OpenAI model |
| 57 | + |
| 58 | + |
| 59 | +**Example Request:** |
| 60 | +```bash |
| 61 | +curl -X POST -H "Content-Type: application/json" -d '{"text": "This is an example text.", "ada": true}' http://127.0.0.1:5000/submit |
| 62 | +``` |
| 63 | + |
| 64 | +**Example Response:** |
| 65 | + |
| 66 | +```json |
| 67 | +{ |
| 68 | + "embedding": [0.123, 0.456, ..., 0.789], |
| 69 | + "status": "success" |
| 70 | +} |
| 71 | +``` |
| 72 | + |
| 73 | +### Error Handling ⚠️ |
| 74 | +The server responds with appropriate error messages if there are any issues with the request or generating embeddings. |
0 commit comments