distilbert_lda_50_v1_book

This model is a fine-tuned version of on the gokulsrinivasagan/processed_book_corpus-ld-50 dataset. It achieves the following results on the evaluation set:

• Loss: 4.1487
• Accuracy: 0.7281

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
8.9162	0.4215	10000	8.5853	0.1955
5.7575	0.8431	20000	5.3762	0.5702
5.3383	1.2646	30000	5.0020	0.6136
5.1543	1.6861	40000	4.8293	0.6350
5.0355	2.1077	50000	4.7268	0.6473
4.9536	2.5292	60000	4.6440	0.6583
4.8934	2.9507	70000	4.5936	0.6648
4.8459	3.3723	80000	4.5508	0.6706
4.8083	3.7938	90000	4.5127	0.6758
4.7736	4.2153	100000	4.4858	0.6791
4.7514	4.6369	110000	4.4608	0.6827
4.7242	5.0584	120000	4.4364	0.6861
4.7039	5.4799	130000	4.4180	0.6886
4.6926	5.9014	140000	4.3993	0.6911
4.6677	6.3230	150000	4.3836	0.6933
4.6595	6.7445	160000	4.3707	0.6953
4.6349	7.1660	170000	4.3575	0.6975
4.6275	7.5876	180000	4.3497	0.6983
4.6149	8.0091	190000	4.3355	0.7002
4.6043	8.4306	200000	4.3248	0.7019
4.5925	8.8522	210000	4.3177	0.7030
4.586	9.2737	220000	4.3079	0.7043
4.5691	9.6952	230000	4.2980	0.7058
4.5581	10.1168	240000	4.2940	0.7070
4.5611	10.5383	250000	4.2824	0.7081
4.5492	10.9598	260000	4.2768	0.7090
4.5384	11.3814	270000	4.2734	0.7095
4.5396	11.8029	280000	4.2642	0.7106
4.5237	12.2244	290000	4.2593	0.7117
4.5205	12.6460	300000	4.2518	0.7127
4.5105	13.0675	310000	4.2450	0.7138
4.5085	13.4890	320000	4.2409	0.7144
4.5037	13.9106	330000	4.2371	0.7149
4.5007	14.3321	340000	4.2299	0.7161
4.4931	14.7536	350000	4.2254	0.7165
4.4864	15.1751	360000	4.2212	0.7171
4.4842	15.5967	370000	4.2133	0.7182
4.4762	16.0182	380000	4.2096	0.7190
4.471	16.4397	390000	4.2059	0.7194
4.467	16.8613	400000	4.2037	0.7198
4.4651	17.2828	410000	4.1972	0.7208
4.4579	17.7043	420000	4.1947	0.7213
4.4519	18.1259	430000	4.1922	0.7217
4.4457	18.5474	440000	4.1885	0.7220
4.4465	18.9689	450000	4.1830	0.7230
4.441	19.3905	460000	4.1797	0.7234
4.4385	19.8120	470000	4.1763	0.7240
4.4331	20.2335	480000	4.1728	0.7246
4.4282	20.6551	490000	4.1693	0.7250
4.4242	21.0766	500000	4.1657	0.7254
4.4206	21.4981	510000	4.1630	0.7259
4.4165	21.9197	520000	4.1615	0.7262
4.4196	22.3412	530000	4.1590	0.7266
4.412	22.7627	540000	4.1587	0.7267
4.4122	23.1843	550000	4.1542	0.7271
4.4097	23.6058	560000	4.1519	0.7277
4.4042	24.0273	570000	4.1513	0.7278
4.4037	24.4488	580000	4.1470	0.7284
4.401	24.8704	590000	4.1480	0.7281

Framework versions

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