Detecting Toxic Text

How To Use

Under 'Demos'

Select a demo to view an example of a certain type of toxicity in action

Under 'Input Text'

Type the text you want to analyze for toxicity

Under 'Select Model'

My Models: 

Toxicity - 1 Epoch : First version of my finetuned model

Toxicity - 8 Epochs : First version of my model with 8 epochs of training

Woxicity - Weighted : Final version of my model that implemented weights in order to correctly classify underrepresented categories

Base Model: DistilBERT Base Uncased (SST-2) - Analyzes text for either positive or negative content

Output

Hit the submit button to view the output

My Model:

Tweet (portion): The text that was inputted

Toxicity Class:  toxic / severe_toxic / obscene / threat / insult / identity_hate

Probability: Probability of prediction

Base Model: 

Tweet (portion): The text that was inputted

Result: POSITIVE / NEGATIVE

Probability: Probability of prediction

If Model Does Not Load

Visit huggingface site to use model / reload for site