fine_tuned_bert_dreadit

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: 2.6081
• Accuracy: 0.7528

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0037	1.0	178	1.8515	0.7163
0.0017	2.0	356	1.7404	0.7163
0.001	3.0	534	1.2895	0.7921
0.0012	4.0	712	1.3320	0.7669
0.0005	5.0	890	1.3646	0.7949
0.0002	6.0	1068	1.5997	0.7809
0.0001	7.0	1246	1.5772	0.7753
0.0003	8.0	1424	1.7599	0.7556
0.0001	9.0	1602	1.7494	0.7640
0.0001	10.0	1780	1.9942	0.7556
0.0001	11.0	1958	1.9370	0.75
0.0	12.0	2136	1.9671	0.7781
0.0001	13.0	2314	2.1223	0.7640
0.0	14.0	2492	2.1653	0.7472
0.0001	15.0	2670	1.9924	0.75
0.0	16.0	2848	2.1778	0.7528
0.0	17.0	3026	2.3010	0.7612
0.0	18.0	3204	2.2210	0.7669
0.0	19.0	3382	2.3333	0.7556
0.0	20.0	3560	1.8684	0.7697
0.0976	21.0	3738	1.9417	0.7584
0.0	22.0	3916	2.1385	0.7472
0.0	23.0	4094	1.9774	0.7669
0.0	24.0	4272	2.0778	0.75
0.0001	25.0	4450	2.4343	0.7331
0.0	26.0	4628	2.1331	0.7528
0.0	27.0	4806	2.2511	0.7640
0.0	28.0	4984	2.2422	0.7584
0.0	29.0	5162	2.1228	0.7669
0.0006	30.0	5340	2.0973	0.7725
0.0	31.0	5518	1.9392	0.7809
0.0	32.0	5696	2.2996	0.7107
0.4186	33.0	5874	2.2191	0.7584
0.0	34.0	6052	2.2233	0.75
0.0	35.0	6230	2.2263	0.7584
0.0	36.0	6408	2.2205	0.7584
0.0	37.0	6586	2.4488	0.7444
0.0	38.0	6764	2.5616	0.7360
0.0	39.0	6942	2.5941	0.7416
0.0	40.0	7120	2.5129	0.7528
0.0	41.0	7298	2.4978	0.7360
0.0	42.0	7476	2.3089	0.7528
0.0	43.0	7654	2.5056	0.7472
0.0	44.0	7832	2.5786	0.7416
0.0	45.0	8010	2.2956	0.7640
0.0	46.0	8188	2.5265	0.7472
0.0	47.0	8366	2.4396	0.7584
0.0	48.0	8544	2.5547	0.7472
0.0	49.0	8722	2.5556	0.7528
0.0	50.0	8900	2.5732	0.7528
0.0	51.0	9078	2.5062	0.7556
0.0	52.0	9256	2.5504	0.7528
0.0	53.0	9434	2.5602	0.7528
0.0	54.0	9612	2.5627	0.7472
0.0	55.0	9790	2.6575	0.75
0.0	56.0	9968	2.6239	0.7528
0.0	57.0	10146	2.4757	0.7697
0.0	58.0	10324	2.4862	0.7612
0.0	59.0	10502	3.2968	0.6938
0.0	60.0	10680	2.5265	0.7472
0.0	61.0	10858	2.1426	0.7978
0.0	62.0	11036	2.4674	0.7640
0.0	63.0	11214	2.3496	0.7640
0.0	64.0	11392	2.4010	0.7556
0.0	65.0	11570	2.4081	0.7725
0.0	66.0	11748	2.4022	0.7753
0.0	67.0	11926	2.2982	0.7753
0.0	68.0	12104	2.4628	0.7612
0.0	69.0	12282	2.5764	0.7640
0.0	70.0	12460	2.4056	0.7781
0.0	71.0	12638	2.3265	0.7865
0.0	72.0	12816	2.5182	0.7640
0.0	73.0	12994	2.3872	0.7556
0.0	74.0	13172	2.7281	0.7388
0.0	75.0	13350	2.4907	0.7612
0.0	76.0	13528	2.5323	0.7584
0.0	77.0	13706	2.2055	0.7837
0.0	78.0	13884	2.2227	0.7865
0.0	79.0	14062	2.2794	0.7753
0.0	80.0	14240	2.2886	0.7753
0.0	81.0	14418	2.8320	0.7444
0.0	82.0	14596	2.8252	0.7472
0.0	83.0	14774	2.2986	0.7837
0.0	84.0	14952	2.7879	0.7416
0.0	85.0	15130	2.7926	0.7416
0.0	86.0	15308	2.7656	0.7472
0.0	87.0	15486	2.7336	0.7444
0.0	88.0	15664	2.7320	0.7444
0.0	89.0	15842	2.7402	0.7444
0.0	90.0	16020	2.7415	0.7444
0.0	91.0	16198	2.7406	0.7444
0.0	92.0	16376	2.7327	0.7444
0.0	93.0	16554	2.4082	0.7781
0.0	94.0	16732	2.4077	0.7753
0.0	95.0	16910	2.4185	0.7781
0.0	96.0	17088	2.6096	0.7528
0.0	97.0	17266	2.5907	0.7669
0.0	98.0	17444	2.6030	0.7556
0.0	99.0	17622	2.6081	0.7528
0.0	100.0	17800	2.6081	0.7528

Framework versions

• Transformers 4.27.3
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2