deberta-v3-xsmall-cotat-recommened-hp

This model is a fine-tuned version of microsoft/deberta-v3-xsmall on the DandinPower/review_cleanonlytitleandtext dataset. It achieves the following results on the evaluation set:

• Loss: 0.8783
• Accuracy: 0.6263
• Macro F1: 0.6285

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: 4.5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
1.61	0.4571	100	1.6076	0.22	0.1631
1.5063	0.9143	200	1.2854	0.4094	0.2942
1.2016	1.3714	300	1.0481	0.5529	0.5311
1.0219	1.8286	400	0.9338	0.6093	0.6020
0.9362	2.2857	500	0.8919	0.6261	0.6239
0.9097	2.7429	600	0.8783	0.6263	0.6285

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1