amazon_helpfulness_classification_on_TAPT_pretrained_full_train

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

• Loss: 0.3275
• Accuracy: 0.8787
• F1 Macro: 0.6852

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.06
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro
0.3175	1.0	7204	0.3037	0.8762	0.6515
0.2996	2.0	14408	0.3404	0.8744	0.6448
0.2624	3.0	21612	0.3263	0.8792	0.6963
0.2322	4.0	28816	0.4314	0.877	0.6752
0.1906	5.0	36020	0.5521	0.8708	0.6836
0.1487	6.0	43224	0.7017	0.8704	0.6941

Framework versions

• Transformers 4.36.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.15.2