scenario-TCR-XLMV_data-AmazonScience_massive_all_1_1_gamma2

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

• Loss: 0.7609
• Accuracy: 0.8498
• F1: 0.8128

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: 77
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.6302	0.27	5000	0.7304	0.8189	0.7638
0.4558	0.53	10000	0.6614	0.8412	0.7995
0.3573	0.8	15000	0.6639	0.8461	0.8146
0.2797	1.07	20000	0.7008	0.8485	0.8198
0.2779	1.34	25000	0.7087	0.8484	0.8225
0.2658	1.6	30000	0.7185	0.8509	0.8235
0.2488	1.87	35000	0.7334	0.8486	0.8229
0.2027	2.14	40000	0.8087	0.8458	0.8201
0.2112	2.41	45000	0.7449	0.8501	0.8228
0.2013	2.67	50000	0.7695	0.8502	0.8203
0.2065	2.94	55000	0.7609	0.8498	0.8128

Framework versions

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