fnsa-mlserver

FNSA-MLServer

This README provides instructions on how to deploy and query a sentiment analysis model using SeldonIO MLServer with a pre-trained Hugging Face model (

). The guide includes steps to set up the server using Docker Compose and send inference requests using

curl

.

Prerequisites

• Docker
• Docker Compose

Setup

Run the Server:

docker-compose up

Querying the Model

You can send an inference request to the model using

. Here's an example of how to do it:

curl -X POST http://localhost:8080/v2/models/financial-news-sentiment/infer -H "Content-Type: application/json" -d '{
  "inputs": [
    {
      "name": "input",
      "shape": [1],
      "datatype": "BYTES",
      "data": ["The company\'s stock price surged after the positive earnings report."]
    }
  ]
}'

Expected Response

The server will respond with the sentiment analysis results. The expected response format is:

{
    "model_name": "financial-news-sentiment",
    "id": "1e993fa8-10d8-4ce3-a012-5c79ab8b127b",
    "parameters": {},
    "outputs": [
        {
            "name": "output",
            "shape": [
                1
            ],
            "datatype": "BYTES",
            "data": [
                {
                    "label": "positive",
                    "score": 0.9996684789657593
                }
            ]
        }
    ]
}

This indicates that the input text was classified as "positive" with a high confidence score.