metadata

base_model: microsoft/mdeberta-v3-base
library_name: transformers
license: mit
metrics:
  - precision
  - recall
  - f1
  - accuracy
tags:
  - generated_from_trainer
model-index:
  - name: scenario-non-kd-scr-ner-full-mdeberta_data-univner_en55
    results: []

scenario-non-kd-scr-ner-full-mdeberta_data-univner_en55

This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.3347
Precision: 0.5325
Recall: 0.4752
F1: 0.5022
Accuracy: 0.9624

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: 3e-05
train_batch_size: 32
eval_batch_size: 32
seed: 55
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.1623	1.2755	500	0.1875	0.4252	0.3002	0.3519	0.9534
0.0693	2.5510	1000	0.1568	0.4472	0.4689	0.4578	0.9583
0.0358	3.8265	1500	0.1803	0.5255	0.4472	0.4832	0.9611
0.0189	5.1020	2000	0.2008	0.5077	0.4793	0.4931	0.9608
0.0098	6.3776	2500	0.2203	0.4759	0.5207	0.4973	0.9599
0.0072	7.6531	3000	0.2339	0.5105	0.4783	0.4939	0.9613
0.0046	8.9286	3500	0.2526	0.5327	0.4638	0.4958	0.9612
0.003	10.2041	4000	0.2658	0.5150	0.4793	0.4965	0.9604
0.0021	11.4796	4500	0.2805	0.5541	0.4669	0.5067	0.9616
0.0015	12.7551	5000	0.2782	0.5156	0.4959	0.5055	0.9621
0.0018	14.0306	5500	0.3001	0.5448	0.4720	0.5058	0.9614
0.0014	15.3061	6000	0.2938	0.5354	0.5083	0.5215	0.9629
0.0012	16.5816	6500	0.3020	0.5554	0.5135	0.5336	0.9633
0.0008	17.8571	7000	0.3080	0.5345	0.4886	0.5105	0.9627
0.0007	19.1327	7500	0.3184	0.5346	0.4638	0.4967	0.9626
0.0004	20.4082	8000	0.3198	0.5233	0.4762	0.4986	0.9612
0.0005	21.6837	8500	0.3303	0.5583	0.4658	0.5079	0.9617
0.0004	22.9592	9000	0.3305	0.5575	0.4969	0.5255	0.9629
0.0003	24.2347	9500	0.3357	0.5440	0.4669	0.5025	0.9617
0.0002	25.5102	10000	0.3258	0.5404	0.5124	0.5260	0.9624
0.0002	26.7857	10500	0.3360	0.5587	0.4928	0.5237	0.9628
0.0001	28.0612	11000	0.3318	0.5208	0.4803	0.4997	0.9618
0.0001	29.3367	11500	0.3347	0.5325	0.4752	0.5022	0.9624

Framework versions

Transformers 4.44.2
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.19.1