exist-2023-task2

This model is a fine-tuned version of microsoft/Multilingual-MiniLM-L12-H384 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4756
• F1: 0.7027

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	97	1.0175	0.4991
No log	2.0	194	0.8374	0.5695
No log	3.0	291	0.7967	0.5876
No log	4.0	388	0.7797	0.5982
No log	5.0	485	0.7161	0.6424
0.8645	6.0	582	0.6662	0.6302
0.8645	7.0	679	0.6580	0.6385
0.8645	8.0	776	0.6465	0.6491
0.8645	9.0	873	0.8620	0.5650
0.8645	10.0	970	0.5704	0.6852
0.6764	11.0	1067	0.5434	0.6806
0.6764	12.0	1164	0.7109	0.6192
0.6764	13.0	1261	0.5411	0.6708
0.6764	14.0	1358	0.5557	0.6675
0.6764	15.0	1455	0.5483	0.6701
0.56	16.0	1552	0.5155	0.6817
0.56	17.0	1649	0.5375	0.6750
0.56	18.0	1746	0.4858	0.6984
0.56	19.0	1843	0.4571	0.7091
0.56	20.0	1940	0.4756	0.7027

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3