S02-PC

This model is a fine-tuned version of Anwaarma/Merged-Server-praj on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5986
• Accuracy: 0.78
• F1: 0.8764

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: 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.0	50	0.5686	0.6	0.5981
No log	0.01	100	0.5715	0.62	0.6205
No log	0.01	150	0.5591	0.64	0.6400
No log	0.01	200	0.5670	0.63	0.6288
No log	0.02	250	0.5568	0.61	0.6106
No log	0.02	300	0.5761	0.64	0.6383
No log	0.02	350	0.5515	0.61	0.6106
No log	0.03	400	0.5567	0.61	0.6087
No log	0.03	450	0.5590	0.62	0.6192
0.606	0.03	500	0.5454	0.64	0.6404
0.606	0.04	550	0.5509	0.63	0.6303
0.606	0.04	600	0.5451	0.64	0.6393
0.606	0.04	650	0.5461	0.65	0.6488
0.606	0.05	700	0.5443	0.62	0.6192
0.606	0.05	750	0.5461	0.66	0.6593
0.606	0.05	800	0.5420	0.66	0.6604
0.606	0.06	850	0.5414	0.65	0.6502
0.606	0.06	900	0.5411	0.65	0.6505
0.606	0.06	950	0.5413	0.69	0.6834
0.584	0.07	1000	0.5432	0.64	0.6353
0.584	0.07	1050	0.5335	0.64	0.6383
0.584	0.07	1100	0.5483	0.67	0.6702
0.584	0.08	1150	0.5548	0.66	0.6605
0.584	0.08	1200	0.5590	0.63	0.6306
0.584	0.09	1250	0.5580	0.67	0.6697
0.584	0.09	1300	0.5616	0.65	0.6502
0.584	0.09	1350	0.5620	0.62	0.6131
0.584	0.1	1400	0.5509	0.61	0.6059
0.584	0.1	1450	0.5473	0.66	0.6605
0.573	0.1	1500	0.5497	0.66	0.6593
0.573	0.11	1550	0.5450	0.65	0.6502
0.573	0.11	1600	0.5484	0.67	0.6689
0.573	0.11	1650	0.5398	0.66	0.6584
0.573	0.12	1700	0.5350	0.65	0.6477
0.573	0.12	1750	0.5333	0.64	0.6370
0.573	0.12	1800	0.5635	0.64	0.6400
0.573	0.13	1850	0.5742	0.63	0.6297

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0