results

This model is a fine-tuned version of michellejieli/emotion_text_classifier on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2828
  • F1: 0.7879
  • Roc Auc: nan
  • Hamming: 0.1039

Model description

This model uses a lightweight RoBERTa checkpoint that has been fine-tuned on evaluating emotions to further be trained on recognizing climate disinformation.

Intended uses & limitations

To be used as a submission for the Frugal AI competition

Training and evaluation data

Dataset of text and labels available on Frugal AI competition page.

Training procedure

Used a binarizer to tokenize the text and found a seemingly suitable model checkpoint as a good place to start!

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 4

Training results

Framework versions

  • Transformers 4.47.1
  • Pytorch 2.5.1+cu121
  • Datasets 3.2.0
  • Tokenizers 0.21.0
Downloads last month
1
Safetensors
Model size
82.1M params
Tensor type
F32
ยท
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for nhankins/frugal_ai_submission

Finetuned
(7)
this model

Space using nhankins/frugal_ai_submission 1