multilingual_minilm-amazon_massive-intent_eu7

This model is a fine-tuned version of microsoft/Multilingual-MiniLM-L12-H384 on the MASSIVE 1.1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.8238
• Accuracy: 0.8623
• F1: 0.8623

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.3523	1.0	5038	1.3058	0.6937	0.6937
0.7842	2.0	10076	0.8434	0.8059	0.8059
0.5359	3.0	15114	0.7231	0.8302	0.8302
0.4106	4.0	20152	0.7121	0.8443	0.8443
0.3294	5.0	25190	0.7366	0.8497	0.8497
0.2621	6.0	30228	0.7702	0.8528	0.8528
0.2164	7.0	35266	0.7773	0.8577	0.8577
0.1756	8.0	40304	0.8080	0.8569	0.8569
0.1625	9.0	45342	0.8162	0.8624	0.8624
0.1448	10.0	50380	0.8238	0.8623	0.8623

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.1+cu116
• Datasets 2.8.0
• Tokenizers 0.13.2