bert-fine-tune-toxic-comment

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

Loss: 0.5352
Accuracy: 0.935
Recall: 0.4762
Precision: 0.8333
F1: 0.6061

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: 6e-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: 100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	Precision	F1
0.2674	1.0	100	0.3102	0.93	0.4286	0.8182	0.5625
0.1455	2.0	200	0.2689	0.93	0.6190	0.6842	0.65
0.0502	3.0	300	0.3969	0.925	0.6667	0.6364	0.6512
0.011	4.0	400	0.5352	0.935	0.4762	0.8333	0.6061
0.0071	5.0	500	0.5568	0.92	0.6667	0.6087	0.6364
0.0242	6.0	600	0.4414	0.915	0.6190	0.5909	0.6047
0.0003	7.0	700	0.5668	0.93	0.4762	0.7692	0.5882
0.0001	8.0	800	0.6002	0.93	0.4762	0.7692	0.5882
0.0001	9.0	900	0.6258	0.93	0.4762	0.7692	0.5882
0.0	10.0	1000	0.6471	0.93	0.4762	0.7692	0.5882
0.0	11.0	1100	0.6641	0.93	0.4762	0.7692	0.5882
0.0	12.0	1200	0.6767	0.93	0.4762	0.7692	0.5882
0.0	13.0	1300	0.6877	0.93	0.4762	0.7692	0.5882
0.0	14.0	1400	0.6971	0.93	0.4762	0.7692	0.5882
0.0	15.0	1500	0.7063	0.93	0.4762	0.7692	0.5882
0.0	16.0	1600	0.7140	0.93	0.4762	0.7692	0.5882
0.0	17.0	1700	0.7210	0.93	0.4762	0.7692	0.5882
0.0	18.0	1800	0.7285	0.93	0.4762	0.7692	0.5882
0.0	19.0	1900	0.7348	0.93	0.4762	0.7692	0.5882
0.0	20.0	2000	0.7410	0.93	0.4762	0.7692	0.5882
0.0	21.0	2100	0.7467	0.93	0.4762	0.7692	0.5882
0.0	22.0	2200	0.7527	0.93	0.4762	0.7692	0.5882
0.0	23.0	2300	0.7582	0.93	0.4762	0.7692	0.5882
0.0	24.0	2400	0.7636	0.93	0.4762	0.7692	0.5882
0.0	25.0	2500	0.7687	0.93	0.4762	0.7692	0.5882
0.0	26.0	2600	0.7738	0.93	0.4762	0.7692	0.5882
0.0	27.0	2700	0.7784	0.93	0.4762	0.7692	0.5882
0.0	28.0	2800	0.7830	0.93	0.4762	0.7692	0.5882
0.0	29.0	2900	0.7873	0.93	0.4762	0.7692	0.5882
0.0	30.0	3000	0.7912	0.93	0.4762	0.7692	0.5882
0.0	31.0	3100	0.7957	0.93	0.4762	0.7692	0.5882
0.0	32.0	3200	0.8000	0.93	0.4762	0.7692	0.5882
0.0	33.0	3300	0.8041	0.93	0.4762	0.7692	0.5882
0.0	34.0	3400	0.8080	0.93	0.4762	0.7692	0.5882
0.0	35.0	3500	0.8117	0.93	0.4762	0.7692	0.5882
0.0	36.0	3600	0.8156	0.93	0.4762	0.7692	0.5882
0.0	37.0	3700	0.8193	0.93	0.4762	0.7692	0.5882
0.0	38.0	3800	0.8228	0.93	0.4762	0.7692	0.5882
0.0	39.0	3900	0.8263	0.93	0.4762	0.7692	0.5882
0.0	40.0	4000	0.8300	0.93	0.4762	0.7692	0.5882
0.0	41.0	4100	0.8333	0.93	0.4762	0.7692	0.5882
0.0	42.0	4200	0.8365	0.93	0.4762	0.7692	0.5882
0.0	43.0	4300	0.8397	0.93	0.4762	0.7692	0.5882
0.0	44.0	4400	0.8429	0.93	0.4762	0.7692	0.5882
0.0	45.0	4500	0.8461	0.93	0.4762	0.7692	0.5882
0.0	46.0	4600	0.8489	0.93	0.4762	0.7692	0.5882
0.0	47.0	4700	0.8523	0.93	0.4762	0.7692	0.5882
0.0	48.0	4800	0.8554	0.93	0.4762	0.7692	0.5882
0.0	49.0	4900	0.8584	0.93	0.4762	0.7692	0.5882
0.0	50.0	5000	0.8614	0.93	0.4762	0.7692	0.5882
0.0	51.0	5100	0.8641	0.93	0.4762	0.7692	0.5882
0.0	52.0	5200	0.8669	0.93	0.4762	0.7692	0.5882
0.0	53.0	5300	0.8698	0.93	0.4762	0.7692	0.5882
0.0	54.0	5400	0.8729	0.93	0.4762	0.7692	0.5882
0.0	55.0	5500	0.8757	0.93	0.4762	0.7692	0.5882
0.0	56.0	5600	0.8785	0.93	0.4762	0.7692	0.5882
0.0	57.0	5700	0.8812	0.93	0.4762	0.7692	0.5882
0.0	58.0	5800	0.8840	0.93	0.4762	0.7692	0.5882
0.0	59.0	5900	0.8865	0.93	0.4762	0.7692	0.5882
0.0	60.0	6000	0.8893	0.93	0.4762	0.7692	0.5882
0.0	61.0	6100	0.8918	0.93	0.4762	0.7692	0.5882
0.0	62.0	6200	0.8944	0.93	0.4762	0.7692	0.5882
0.0	63.0	6300	0.8968	0.93	0.4762	0.7692	0.5882
0.0	64.0	6400	0.8993	0.93	0.4762	0.7692	0.5882
0.0	65.0	6500	0.9018	0.93	0.4762	0.7692	0.5882
0.0	66.0	6600	0.9044	0.93	0.4762	0.7692	0.5882
0.0	67.0	6700	0.9065	0.93	0.4762	0.7692	0.5882
0.0	68.0	6800	0.9088	0.93	0.4762	0.7692	0.5882
0.0	69.0	6900	0.9113	0.93	0.4762	0.7692	0.5882
0.0	70.0	7000	0.9132	0.93	0.4762	0.7692	0.5882
0.0	71.0	7100	0.9156	0.93	0.4762	0.7692	0.5882
0.0	72.0	7200	0.9181	0.93	0.4762	0.7692	0.5882
0.0	73.0	7300	0.9204	0.93	0.4762	0.7692	0.5882
0.0	74.0	7400	0.9222	0.93	0.4762	0.7692	0.5882
0.0	75.0	7500	0.9244	0.93	0.4762	0.7692	0.5882
0.0	76.0	7600	0.9263	0.93	0.4762	0.7692	0.5882
0.0	77.0	7700	0.9283	0.93	0.4762	0.7692	0.5882
0.0	78.0	7800	0.9304	0.93	0.4762	0.7692	0.5882
0.0	79.0	7900	0.9324	0.93	0.4762	0.7692	0.5882
0.0	80.0	8000	0.9342	0.93	0.4762	0.7692	0.5882
0.0	81.0	8100	0.9359	0.93	0.4762	0.7692	0.5882
0.0	82.0	8200	0.9377	0.93	0.4762	0.7692	0.5882
0.0	83.0	8300	0.9396	0.93	0.4762	0.7692	0.5882
0.0	84.0	8400	0.9414	0.93	0.4762	0.7692	0.5882
0.0	85.0	8500	0.9431	0.93	0.4762	0.7692	0.5882
0.0	86.0	8600	0.9446	0.93	0.4762	0.7692	0.5882
0.0	87.0	8700	0.9461	0.93	0.4762	0.7692	0.5882
0.0	88.0	8800	0.9474	0.93	0.4762	0.7692	0.5882
0.0	89.0	8900	0.9486	0.93	0.4762	0.7692	0.5882
0.0	90.0	9000	0.9497	0.93	0.4762	0.7692	0.5882
0.0	91.0	9100	0.9509	0.93	0.4762	0.7692	0.5882
0.0	92.0	9200	0.9519	0.93	0.4762	0.7692	0.5882
0.0	93.0	9300	0.9528	0.93	0.4762	0.7692	0.5882
0.0	94.0	9400	0.9538	0.93	0.4762	0.7692	0.5882
0.0	95.0	9500	0.9545	0.93	0.4762	0.7692	0.5882
0.0	96.0	9600	0.9551	0.93	0.4762	0.7692	0.5882
0.0	97.0	9700	0.9556	0.93	0.4762	0.7692	0.5882
0.0	98.0	9800	0.9559	0.93	0.4762	0.7692	0.5882
0.0	99.0	9900	0.9562	0.93	0.4762	0.7692	0.5882
0.0	100.0	10000	0.9563	0.93	0.4762	0.7692	0.5882

Framework versions

Transformers 4.41.2
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.19.1

Salmamoori/bert-fine-tune-toxic-comment

bert-fine-tune-toxic-comment

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种