r10521708/albert-base-chinese-finetuned-qqp-FHTM-5x

albert-base-chinese-finetuned-qqp

This model is a fine-tuned version of ckiplab/albert-base-chinese on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3385688364505768
• Accuracy: 0.8357142857142857
• F1: 0.8244274809160306

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=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	30	0.654749	0.642857	0.719101
No log	2.0	60	0.614816	0.728571	0.707692
No log	3.0	90	0.443354	0.807143	0.802920
No log	4.0	120	0.338569	0.835714	0.824427
No log	5.0	150	0.339324	0.828571	0.806452

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1
• Datasets 2.9.0
• Tokenizers 0.13.0.dev0