flat_model

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0506

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	2	5.5895
No log	2.0	4	5.3197
No log	3.0	6	5.0073
No log	4.0	8	4.7018
No log	5.0	10	4.4059
No log	6.0	12	4.1189
No log	7.0	14	3.8437
No log	8.0	16	3.5810
No log	9.0	18	3.3421
No log	10.0	20	3.0996
No log	11.0	22	2.8624
No log	12.0	24	2.6348
No log	13.0	26	2.4103
No log	14.0	28	2.1955
No log	15.0	30	2.0006
No log	16.0	32	1.8247
No log	17.0	34	1.6528
No log	18.0	36	1.4862
No log	19.0	38	1.3387
No log	20.0	40	1.2120
No log	21.0	42	1.1054
No log	22.0	44	1.0292
No log	23.0	46	0.9989
No log	24.0	48	0.9831
No log	25.0	50	0.9650
No log	26.0	52	0.9665
No log	27.0	54	0.9973
No log	28.0	56	1.0356
No log	29.0	58	1.0376
No log	30.0	60	1.0142
No log	31.0	62	0.9909
No log	32.0	64	0.9877
No log	33.0	66	0.9970
No log	34.0	68	1.0117
No log	35.0	70	1.0140
No log	36.0	72	1.0170
No log	37.0	74	1.0477
No log	38.0	76	1.0506
No log	39.0	78	1.0521
No log	40.0	80	1.0616
No log	41.0	82	1.0763
No log	42.0	84	1.0793
No log	43.0	86	1.0772
No log	44.0	88	1.0761
No log	45.0	90	1.0765
No log	46.0	92	1.0721
No log	47.0	94	1.0605
No log	48.0	96	1.0543
No log	49.0	98	1.0511
No log	50.0	100	1.0506

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.1+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0