metadata

license: mit
base_model: microsoft/deberta-v3-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: aptner_deberta
    results: []

aptner_deberta

This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2929
Precision: 0.5550
Recall: 0.5835
F1: 0.5689
Accuracy: 0.9205

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: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.6136	0.59	500	0.3298	0.5007	0.5287	0.5143	0.9172
0.308	1.19	1000	0.2929	0.5550	0.5835	0.5689	0.9205
0.2428	1.78	1500	0.3124	0.5330	0.6192	0.5729	0.9177
0.2088	2.37	2000	0.3204	0.5356	0.6440	0.5848	0.9147
0.1783	2.97	2500	0.3319	0.5432	0.6760	0.6024	0.9149
0.1434	3.56	3000	0.3371	0.5640	0.6494	0.6037	0.9203
0.1352	4.15	3500	0.3827	0.5425	0.6249	0.5808	0.9135
0.1135	4.74	4000	0.3862	0.5360	0.6760	0.5979	0.9136
0.0987	5.34	4500	0.3978	0.5439	0.6497	0.5921	0.9141
0.0942	5.93	5000	0.3738	0.5791	0.6425	0.6091	0.9225
0.0746	6.52	5500	0.4269	0.5490	0.6479	0.5943	0.9161
0.0727	7.12	6000	0.4236	0.5579	0.6437	0.5977	0.9171
0.0661	7.71	6500	0.4239	0.5650	0.6479	0.6036	0.9200
0.0578	8.3	7000	0.4485	0.5579	0.6332	0.5932	0.9175
0.0505	8.9	7500	0.4553	0.5546	0.6353	0.5922	0.9163
0.0513	9.49	8000	0.4629	0.5587	0.6425	0.5977	0.9171

Framework versions

Transformers 4.36.0.dev0
Pytorch 2.1.0+cu118
Datasets 2.14.6
Tokenizers 0.14.1