Demonstration of Sentiment Analysis Board
In this demonstration, we will showcase one of the boards we’ve been developing, which is designed for Text Classification, specifically Sentiment Analysis.
Overview
The primary goal of this board is to perform sentiment analysis on given text inputs using the Hugging Face inference API.
Inputs and Configuration
- API Key: This key is required to access the inference API.
- Text Input: A piece of text that we want to analyse for sentiment.
- Use Cache: This option speeds up requests by using cached responses if the API has encountered the same input before.
- Wait for Model: This flag indicates whether we should wait for the model to fully load before receiving a response.
Example
Let’s consider the input: “I really like the work that I saw at the conference.”
By performing sentiment analysis on this text, we expect a positive result.
Results
Upon analysing the input, the API returns scores for specific labels. In this case, the “positive” label received a very high confidence level, indicating that the text was perceived as highly positive.
In a subsequent video, we will demonstrate a use case of this board using our Chrome Breadboard extension.