metadata

license: mit
base_model: cointegrated/rubert-tiny2
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: rubert-tiny2-odonata-extended-ner
    results: []

rubert-tiny2-odonata-extended-ner

This model is a fine-tuned version of cointegrated/rubert-tiny2 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0094
Precision: 0.6364
Recall: 0.7101
F1: 0.6712
Accuracy: 0.9973

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	32	0.1433	0.0	0.0	0.0	0.9961
No log	2.0	64	0.0364	0.0	0.0	0.0	0.9961
No log	3.0	96	0.0310	0.0	0.0	0.0	0.9961
No log	4.0	128	0.0286	0.0	0.0	0.0	0.9961
No log	5.0	160	0.0250	0.0	0.0	0.0	0.9961
No log	6.0	192	0.0183	0.6667	0.0290	0.0556	0.9962
No log	7.0	224	0.0141	0.5581	0.3478	0.4286	0.9965
No log	8.0	256	0.0122	0.6111	0.4783	0.5366	0.9969
No log	9.0	288	0.0111	0.6792	0.5217	0.5902	0.9971
No log	10.0	320	0.0105	0.6154	0.5797	0.5970	0.9970
No log	11.0	352	0.0101	0.5857	0.5942	0.5899	0.9971
No log	12.0	384	0.0097	0.6143	0.6232	0.6187	0.9972
No log	13.0	416	0.0096	0.6203	0.7101	0.6622	0.9972
No log	14.0	448	0.0094	0.6282	0.7101	0.6667	0.9973
No log	15.0	480	0.0094	0.6364	0.7101	0.6712	0.9973

Framework versions

Transformers 4.41.2
Pytorch 2.3.1+cpu
Datasets 2.19.2
Tokenizers 0.19.1