distilbert_lda_5_v1_book

This model is a fine-tuned version of on the gokulsrinivasagan/processed_book_corpus-ld-5 dataset. It achieves the following results on the evaluation set:

• Loss: 2.5846
• Accuracy: 0.7334

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
7.1144	0.4215	10000	6.9372	0.1643
4.7967	0.8431	20000	4.3544	0.4792
3.8152	1.2646	30000	3.5063	0.5921
3.5801	1.6861	40000	3.2866	0.6241
3.4416	2.1077	50000	3.1632	0.6409
3.3557	2.5292	60000	3.0762	0.6544
3.2898	2.9507	70000	3.0201	0.6625
3.2439	3.3723	80000	2.9767	0.6697
3.2059	3.7938	90000	2.9382	0.6752
3.1715	4.2153	100000	2.9128	0.6798
3.1509	4.6369	110000	2.8885	0.6834
3.1256	5.0584	120000	2.8609	0.6874
3.1038	5.4799	130000	2.8452	0.6902
3.0895	5.9014	140000	2.8255	0.6932
3.0671	6.3230	150000	2.8121	0.6954
3.0596	6.7445	160000	2.7992	0.6978
3.0371	7.1660	170000	2.7860	0.7002
3.0289	7.5876	180000	2.7773	0.7014
3.0178	8.0091	190000	2.7669	0.7029
3.0064	8.4306	200000	2.7545	0.7053
2.9931	8.8522	210000	2.7466	0.7063
2.9905	9.2737	220000	2.7372	0.7076
2.9751	9.6952	230000	2.7286	0.7091
2.9645	10.1168	240000	2.7234	0.7108
2.9627	10.5383	250000	2.7143	0.7116
2.9517	10.9598	260000	2.7073	0.7128
2.9439	11.3814	270000	2.7033	0.7135
2.944	11.8029	280000	2.6944	0.7151
2.9295	12.2244	290000	2.6887	0.7156
2.9263	12.6460	300000	2.6823	0.7172
2.9172	13.0675	310000	2.6772	0.7180
2.9126	13.4890	320000	2.6722	0.7187
2.9094	13.9106	330000	2.6684	0.7194
2.9054	14.3321	340000	2.6614	0.7204
2.8972	14.7536	350000	2.6573	0.7210
2.8931	15.1751	360000	2.6545	0.7217
2.8894	15.5967	370000	2.6468	0.7227
2.8841	16.0182	380000	2.6425	0.7235
2.8799	16.4397	390000	2.6386	0.7241
2.8742	16.8613	400000	2.6370	0.7245
2.8716	17.2828	410000	2.6301	0.7255
2.8658	17.7043	420000	2.6268	0.7263
2.8605	18.1259	430000	2.6263	0.7266
2.8549	18.5474	440000	2.6233	0.7268
2.8554	18.9689	450000	2.6175	0.7281
2.8499	19.3905	460000	2.6141	0.7286
2.8483	19.8120	470000	2.6111	0.7288
2.8417	20.2335	480000	2.6082	0.7296
2.8365	20.6551	490000	2.6040	0.7302
2.8332	21.0766	500000	2.6011	0.7304
2.8313	21.4981	510000	2.5988	0.7311
2.8267	21.9197	520000	2.5971	0.7314
2.8285	22.3412	530000	2.5954	0.7316
2.8236	22.7627	540000	2.5944	0.7319
2.8246	23.1843	550000	2.5907	0.7323
2.8196	23.6058	560000	2.5883	0.7329
2.8161	24.0273	570000	2.5877	0.7330
2.8153	24.4488	580000	2.5842	0.7335
2.8117	24.8704	590000	2.5848	0.7332

Framework versions

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