|
|
--- |
|
|
library_name: transformers |
|
|
tags: [toxic-comment] |
|
|
--- |
|
|
|
|
|
# Model Card for Model ID |
|
|
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
|
|
|
This model is fine-tuned on top of best base uncased model for task of performing classification of task as sarcastic or not. Its purpose is to predict whether a given text contains hate speech or not. |
|
|
Class Label are 1 for toxic comment and 0 for not. |
|
|
|
|
|
## Model Details |
|
|
|
|
|
### Model Description |
|
|
|
|
|
<!-- Provide a longer summary of what this model is. --> |
|
|
**Important info** This model works with binary classification and doesn't consider multilabel clssification. It detects it's either a toxic comment or not. |
|
|
- **Developed by:** Ayush Dhoundiyal |
|
|
- **Language(s) (NLP):** English |
|
|
- **Finetuned from model:** Bert Base Uncased |
|
|
|
|
|
### Model Sources [optional] |
|
|
|
|
|
<!-- Provide the basic links for the model. --> |
|
|
- **Paper:** https://github.com/ayushdh96/Natural-Language-Processing/blob/main/Ayush_Dhoundiyal_Project_Report.pdf |
|
|
[More Information Needed] |
|
|
## Training Details |
|
|
### Training Data |
|
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> |
|
|
Pre-processing invloved basic steps like lemmtizing, stemming of words. Removing stop words and lowercasing the text to be classified. It's requested to perform these steps for good results. |
|
|
## Evaluation |
|
|
<!-- This section describes the evaluation protocols and provides the results. --> |
|
|
The model provides the accuracy of 0.95, precision of 0.84. recall of 0.62 and f1 score of 0.71. |
