phobert_large_fgm

This model is a fine-tuned version of vinai/phobert-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2865
• Micro Precision: 0.9534
• Micro Recall: 0.9578
• Micro F1: 0.9556

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Micro Precision	Micro Recall	Micro F1
No log	1.0	79	0.2884	0.8462	0.8909	0.8680
No log	2.0	158	0.1427	0.9396	0.9504	0.9450
0.6764	3.0	237	0.1380	0.9372	0.9606	0.9488
0.6764	4.0	316	0.1544	0.9502	0.9539	0.9520
0.6764	5.0	395	0.1585	0.9458	0.9540	0.9499
0.0453	6.0	474	0.1704	0.9543	0.9599	0.9571
0.0453	7.0	553	0.1790	0.9489	0.9630	0.9559
0.0238	8.0	632	0.1652	0.9556	0.9592	0.9574
0.0238	9.0	711	0.1925	0.9522	0.9543	0.9532
0.0238	10.0	790	0.1886	0.9474	0.9619	0.9546
0.0140	11.0	869	0.1982	0.9519	0.9604	0.9561
0.0140	12.0	948	0.1997	0.9519	0.9580	0.9549
0.0094	13.0	1027	0.1998	0.9546	0.9614	0.9580
0.0094	14.0	1106	0.2007	0.9523	0.9646	0.9584
0.0094	15.0	1185	0.2277	0.9548	0.9608	0.9578
0.0061	16.0	1264	0.2140	0.9549	0.9622	0.9585
0.0061	17.0	1343	0.2240	0.9549	0.9628	0.9588
0.0031	18.0	1422	0.2366	0.9580	0.9615	0.9598
0.0031	19.0	1501	0.2390	0.9585	0.9584	0.9585
0.0031	20.0	1580	0.2419	0.9561	0.9647	0.9604
0.0030	21.0	1659	0.2392	0.9558	0.9636	0.9597
0.0030	22.0	1738	0.2394	0.9574	0.9620	0.9597
0.0010	23.0	1817	0.2361	0.9569	0.9630	0.9599
0.0010	24.0	1896	0.2412	0.9562	0.9642	0.9602
0.0010	25.0	1975	0.2447	0.9575	0.9635	0.9605
0.0009	26.0	2054	0.2448	0.9577	0.9638	0.9607
0.0009	27.0	2133	0.2445	0.9580	0.9639	0.9609
0.0006	28.0	2212	0.2453	0.9578	0.9642	0.9609
0.0006	29.0	2291	0.2464	0.9576	0.9639	0.9607
0.0006	30.0	2370	0.2463	0.9572	0.9642	0.9607

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.3
• Tokenizers 0.22.2