metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - go_emotions
metrics:
  - f1
model-index:
  - name: roberta-large-goemotions
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: go_emotions
          type: multilabel_classification
          config: simplified
          split: test
          args: simplified
        metrics:
          - name: F1
            type: f1
            value: 0.5102
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: go_emotions
          type: multilabel_classification
          config: simplified
          split: validation
          args: simplified
        metrics:
          - name: F1
            type: f1
            value: 0.5227

Text Classification GoEmotions

This model is a fine-tuned version of roberta-large on the go_emotions dataset. It achieves the following results on the test set (with a threshold of 0.15):

Accuracy: 0.4175
Precision: 0.4934
Recall: 0.5621
F1: 0.5102

Code

Code for training this model can be found here.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 128
eval_batch_size: 128
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Validation Loss	Accuracy	Precision	Recall	F1
No log	1.0	0.088978	0.404349	0.480763	0.456827	0.444685
0.10620	2.0	0.082806	0.411353	0.460896	0.536386	0.486819
0.10620	3.0	0.081338	0.420199	0.519828	0.561297	0.522716

Framework versions

Transformers 4.20.1
Pytorch 1.12.0
Datasets 2.1.0
Tokenizers 0.12.1