README.md

metadata

library_name: transformers
license: mit
base_model: vinai/phobert-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: phobert-human-tl-seg-seed-6969
    results: []

phobert-human-tl-seg-seed-6969

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

• Loss: 0.4483
• Accuracy: 0.8432
• Precision: 0.6507
• Recall: 0.4682
• F1: 0.4928

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	1.0	346	0.5129	0.8245	0.5325	0.3537	0.3399
0.5757	2.0	692	0.4692	0.8383	0.6238	0.4332	0.4500
0.4598	3.0	1038	0.4671	0.8357	0.6235	0.4116	0.4280
0.4598	4.0	1384	0.4641	0.8394	0.6756	0.4279	0.4498
0.4567	5.0	1730	0.4582	0.8379	0.6325	0.4272	0.4478
0.4499	6.0	2076	0.4588	0.8406	0.6718	0.4424	0.4617
0.4499	7.0	2422	0.4633	0.8372	0.6655	0.4118	0.4320
0.4517	8.0	2768	0.4522	0.8417	0.6340	0.4522	0.4741
0.4477	9.0	3114	0.4539	0.8402	0.6644	0.4267	0.4503
0.4477	10.0	3460	0.4560	0.8439	0.6645	0.4647	0.4813
0.4508	11.0	3806	0.4534	0.8387	0.6499	0.4235	0.4488
0.444	12.0	4152	0.4483	0.8432	0.6507	0.4682	0.4928
0.444	13.0	4498	0.4501	0.8391	0.6491	0.4355	0.4633
0.4506	14.0	4844	0.4532	0.8394	0.6442	0.4253	0.4490
0.4422	15.0	5190	0.4528	0.8394	0.6497	0.4245	0.4494
0.4498	16.0	5536	0.4663	0.8376	0.6743	0.4123	0.4332
0.4498	17.0	5882	0.4476	0.8417	0.6345	0.4435	0.4672

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0