Edit model card

xml-roberta-HU-Com

This model is a fine-tuned version of daveni/twitter-xlm-roberta-emotion-es on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.3693
  • Accuracy: 0.7911
  • F1: 0.7440
  • Precision: 0.7415
  • Recall: 0.7466

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: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
0.6717 1.0 90 0.5918 0.6852 0.5272 0.6774 0.4315
0.453 2.0 180 0.5358 0.7465 0.6403 0.7570 0.5548
0.2631 3.0 270 0.7088 0.7744 0.7273 0.7152 0.7397
0.1936 4.0 360 0.7078 0.7939 0.7566 0.7278 0.7877
0.1273 5.0 450 1.1057 0.7772 0.7436 0.6988 0.7945
0.066 6.0 540 1.1990 0.7799 0.7168 0.7519 0.6849
0.0286 7.0 630 1.2457 0.7994 0.7584 0.7434 0.7740
0.0261 8.0 720 1.3297 0.7799 0.7106 0.7638 0.6644
0.0097 9.0 810 1.3733 0.7855 0.7354 0.7379 0.7329
0.0071 10.0 900 1.3693 0.7911 0.7440 0.7415 0.7466

Framework versions

  • Transformers 4.43.0.dev0
  • Pytorch 2.0.1+cu117
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
8
Safetensors
Model size
278M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for raulgdp/xml-roberta-HU-Com

Finetuned
(2)
this model