finetuned_minilm

This model is a fine-tuned version of nreimers/MiniLM-L6-H384-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6736
• Accuracy: 0.9023

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: 12345
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5371	1.0	619	0.2941	0.8782
0.2763	2.0	1238	0.2590	0.8986
0.1899	3.0	1857	0.3081	0.8959
0.1257	4.0	2476	0.2576	0.9177
0.0929	5.0	3095	0.3949	0.9059
0.0806	6.0	3714	0.3304	0.9173
0.0629	7.0	4333	0.4214	0.9073
0.0474	8.0	4952	0.4625	0.9145
0.0498	9.0	5571	0.4227	0.9236
0.049	10.0	6190	0.5549	0.8945
0.0411	11.0	6809	0.3340	0.9341
0.0272	12.0	7428	0.3317	0.9291
0.0264	13.0	8047	0.4099	0.9305
0.0279	14.0	8666	0.4092	0.9268
0.0242	15.0	9285	0.4418	0.9318
0.0241	16.0	9904	0.4352	0.9273
0.0238	17.0	10523	0.5306	0.9259
0.0216	18.0	11142	0.4267	0.9241
0.0166	19.0	11761	0.5134	0.9255
0.0182	20.0	12380	0.6736	0.9023

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2