scenario-KD-SCR-MSV-D2_data-massive01_all_1_1_gamma

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the massive dataset. It achieves the following results on the evaluation set:

• Loss: 4.2569
• Accuracy: 0.6488
• F1: 0.5858

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: 32
• eval_batch_size: 32
• seed: 123621
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
4.1594	2.72	5000	5.2501	0.4736	0.3351
1.8526	5.45	10000	4.7491	0.5816	0.4876
1.2429	8.17	15000	4.8297	0.6011	0.5244
0.9972	10.89	20000	4.7047	0.6111	0.5369
0.8295	13.62	25000	4.6413	0.6268	0.5586
0.7223	16.34	30000	4.5415	0.6323	0.5675
0.6474	19.06	35000	4.3760	0.6384	0.5758
0.5807	21.79	40000	4.4236	0.6403	0.5810
0.5315	24.51	45000	4.2966	0.6446	0.5787
0.4891	27.23	50000	4.2983	0.6463	0.5833
0.4823	29.96	55000	4.2569	0.6488	0.5858

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.13.3