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-42
    results: []

phobert-human-tl-seg-seed-42

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: 0.4484
• Accuracy: 0.8428
• Precision: 0.6491
• Recall: 0.4691
• F1: 0.4932

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	0.5133	0.8256	0.5419	0.3574	0.3465
0.5919	2.0	692	0.4692	0.8372	0.6180	0.4317	0.4475
0.4602	3.0	1038	0.4660	0.8361	0.6390	0.4132	0.4307
0.4602	4.0	1384	0.4649	0.8394	0.6767	0.4268	0.4487
0.4568	5.0	1730	0.4581	0.8372	0.6211	0.4259	0.4459
0.4498	6.0	2076	0.4586	0.8402	0.6572	0.4411	0.4606
0.4498	7.0	2422	0.4628	0.8372	0.6626	0.4118	0.4318
0.4516	8.0	2768	0.4520	0.8417	0.6340	0.4522	0.4741
0.4473	9.0	3114	0.4540	0.8387	0.6594	0.4217	0.4441
0.4473	10.0	3460	0.4559	0.8447	0.6911	0.4675	0.4855
0.451	11.0	3806	0.4530	0.8379	0.6461	0.4221	0.4468
0.4436	12.0	4152	0.4484	0.8428	0.6491	0.4691	0.4932
0.4436	13.0	4498	0.4499	0.8394	0.6503	0.4367	0.4648
0.4506	14.0	4844	0.4526	0.8391	0.6419	0.4251	0.4486
0.4418	15.0	5190	0.4523	0.8398	0.6584	0.4261	0.4522
0.4497	16.0	5536	0.4667	0.8372	0.6662	0.4110	0.4317
0.4497	17.0	5882	0.4474	0.8417	0.6431	0.4438	0.4686

Framework versions

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