bert_base_lda_100_v1_book

This model is a fine-tuned version of on the gokulsrinivasagan/processed_book_corpus-ld-100 dataset. It achieves the following results on the evaluation set:

• Loss: 4.4757
• Accuracy: 0.7551

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: 0.0001
• train_batch_size: 96
• eval_batch_size: 96
• seed: 10
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10000
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
9.5861	0.4215	10000	9.3593	0.1636
9.2311	0.8431	20000	9.0701	0.1641
9.1362	1.2646	30000	8.9889	0.1662
6.984	1.6861	40000	6.4486	0.4683
5.8878	2.1077	50000	5.5436	0.5941
5.6092	2.5292	60000	5.2862	0.6336
5.4567	2.9507	70000	5.1556	0.6531
5.3612	3.3723	80000	5.0665	0.6665
5.2881	3.7938	90000	4.9997	0.6763
5.2269	4.2153	100000	4.9488	0.6834
5.1854	4.6369	110000	4.9087	0.6889
5.1474	5.0584	120000	4.8727	0.6944
5.1144	5.4799	130000	4.8404	0.6990
5.0914	5.9014	140000	4.8134	0.7026
5.0579	6.3230	150000	4.7935	0.7058
5.044	6.7445	160000	4.7736	0.7089
5.0124	7.1660	170000	4.7533	0.7124
4.9997	7.5876	180000	4.7431	0.7140
4.9809	8.0091	190000	4.7224	0.7163
4.9674	8.4306	200000	4.7087	0.7189
4.9526	8.8522	210000	4.6976	0.7206
4.9422	9.2737	220000	4.6848	0.7224
4.9232	9.6952	230000	4.6721	0.7242
4.9073	10.1168	240000	4.6646	0.7261
4.9072	10.5383	250000	4.6502	0.7276
4.8916	10.9598	260000	4.6415	0.7290
4.8805	11.3814	270000	4.6341	0.7301
4.8775	11.8029	280000	4.6231	0.7318
4.86	12.2244	290000	4.6168	0.7330
4.8539	12.6460	300000	4.6066	0.7346
4.8416	13.0675	310000	4.5990	0.7359
4.8371	13.4890	320000	4.5932	0.7366
4.8312	13.9106	330000	4.5877	0.7373
4.8264	14.3321	340000	4.5786	0.7392
4.8157	14.7536	350000	4.5722	0.7398
4.8097	15.1751	360000	4.5686	0.7404
4.8045	15.5967	370000	4.5596	0.7416
4.7951	16.0182	380000	4.5546	0.7427
4.7906	16.4397	390000	4.5487	0.7434
4.7858	16.8613	400000	4.5452	0.7441
4.7798	17.2828	410000	4.5381	0.7452
4.7725	17.7043	420000	4.5330	0.7461
4.7628	18.1259	430000	4.5300	0.7466
4.7584	18.5474	440000	4.5251	0.7472
4.7573	18.9689	450000	4.5193	0.7483
4.7502	19.3905	460000	4.5154	0.7488
4.7482	19.8120	470000	4.5109	0.7496
4.738	20.2335	480000	4.5065	0.7502
4.7328	20.6551	490000	4.5021	0.7509
4.7289	21.0766	500000	4.4975	0.7515
4.7235	21.4981	510000	4.4950	0.7521
4.7191	21.9197	520000	4.4923	0.7525
4.7191	22.3412	530000	4.4891	0.7530
4.714	22.7627	540000	4.4877	0.7534
4.714	23.1843	550000	4.4832	0.7539
4.7086	23.6058	560000	4.4800	0.7544
4.7055	24.0273	570000	4.4785	0.7547
4.7024	24.4488	580000	4.4747	0.7552
4.6993	24.8704	590000	4.4753	0.7550

Framework versions

• Transformers 4.46.3
• Pytorch 2.2.1+cu118
• Datasets 2.17.0
• Tokenizers 0.20.3