README.md

metadata

language:
  - mn
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: roberta-base-ner-demo
    results: []

roberta-base-ner-demo

This model is a fine-tuned version of bayartsogt/mongolian-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1444
• Precision: 0.9066
• Recall: 0.9148
• F1: 0.9107
• Accuracy: 0.9794

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: 32
• 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	Precision	Recall	F1	Accuracy
0.1657	1.0	477	0.0976	0.8844	0.8947	0.8895	0.9692
0.0631	2.0	954	0.0917	0.8871	0.9084	0.8976	0.9709
0.0387	3.0	1431	0.1079	0.8978	0.9099	0.9038	0.9714
0.0272	4.0	1908	0.1198	0.8993	0.9119	0.9056	0.9716
0.017	5.0	2385	0.1235	0.9038	0.9108	0.9073	0.9783
0.007	6.0	2862	0.1272	0.9085	0.9151	0.9118	0.9795
0.0038	7.0	3339	0.1295	0.9064	0.9172	0.9118	0.9796
0.0029	8.0	3816	0.1368	0.9045	0.9167	0.9106	0.9795
0.0019	9.0	4293	0.1425	0.9076	0.9173	0.9124	0.9796
0.0015	10.0	4770	0.1444	0.9066	0.9148	0.9107	0.9794

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3