metadata
tags:
- text-regression
- emotion
- sentiment
- emotion intensity
language:
- unk
widget:
- text: I'm scared
datasets:
- SemEval-2018-Task-1-Text-Regression-Task
co2_eq_emissions:
emissions: 0.17201402406362057
pipeline_tag: text-classification
twitter-roberta-base-fear-intensity
This model is a fine-tuned version of cardiffnlp/twitter-roberta-base-2022-154m on the SemEval 2018 - Task 1 Affect in Tweets (subtask: El-reg / text regression).
- Warning: Hosted inference API produces inaccurate values
Model Trained Using AutoTrain
- Problem type: Single Column Regression
- Model ID: 68748137460
- CO2 Emissions (in grams): 0.1720
Validation Metrics
- Loss: 0.011
- MSE: 0.011
- MAE: 0.083
- R2: 0.712
- RMSE: 0.107
- Explained Variance: 0.743
Usage
You can use cURL to access this model:
$ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I'm scared"}' https://api-inference.huggingface.co/models/garrettbaber/twitter-roberta-base-fear-intensity
Or Python API:
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model = AutoModelForSequenceClassification.from_pretrained("garrettbaber/twitter-roberta-base-fear-intensity")
tokenizer = AutoTokenizer.from_pretrained("garrettbaber/twitter-roberta-base-fear-intensity")
inputs = tokenizer("I'm scared", return_tensors="pt")
outputs = model(**inputs)