---
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: bert-hateful-memes-expanded
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# bert-hateful-memes-expanded

This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on texts from the following datasets:
- [Hateful Memes](https://hatefulmemeschallenge.com/), `train`, `dev_seen` and `dev_unseen`
- [HarMeme](https://github.com/di-dimitrov/harmeme), `train`, `val` and `test`
- [MultiOFF](https://github.com/bharathichezhiyan/Multimodal-Meme-Classification-Identifying-Offensive-Content-in-Image-and-Text), `Training`, `Validation` and `Testing`

It achieves the following results on the evaluation set:
- Loss: 3.7600

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0

### Training results



### Framework versions

- Transformers 4.11.0
- Pytorch 1.8.1+cu102
- Datasets 1.8.0
- Tokenizers 0.10.2