Description

The input structure is:

summarize: [text]. hypothesis: [hypothesis] , and the output is 0 (hypothesis is not supported) or 1 (hypothesis is supported).

This tests whether a hypothesis is true given the preceding text. Currently the model is trained on banking chatbot intent data, such as:

summarize: How old do my kids need to be to use your service?. hypothesis: asking about an age limit

Output: 1

Model Trained Using AutoTrain

• Problem type: Summarization
• Model ID: 840226804
• CO2 Emissions (in grams): 0.02920886926438328

Validation Metrics

• Loss: 0.09617297351360321
• Rouge1: 91.2874
• Rouge2: 0.0
• RougeL: 91.2874
• RougeLsum: 91.4174
• Gen Len: 2.4915

Usage

You can use cURL to access this model:

$ curl -X POST -H "Authorization: Bearer YOUR_HUGGINGFACE_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoTrain"}' https://api-inference.huggingface.co/jeremyccollinsmpi/autotrain-inference_probability_2-840226804