distilbert_lda_v1_book

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

• Loss: 3.1051
• Accuracy: 0.7315

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 OptimizerNames.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
7.6644	0.4215	10000	7.4872	0.1609
4.6569	0.8431	20000	4.2985	0.5609
4.2453	1.2646	30000	3.9352	0.6094
4.0748	1.6861	40000	3.7740	0.6315
3.9464	2.1077	50000	3.6555	0.6453
3.8728	2.5292	60000	3.5809	0.6566
3.814	2.9507	70000	3.5364	0.6635
3.771	3.3723	80000	3.4922	0.6700
3.735	3.7938	90000	3.4582	0.6753
3.7016	4.2153	100000	3.4345	0.6790
3.681	4.6369	110000	3.4123	0.6824
3.6573	5.0584	120000	3.3854	0.6861
3.6373	5.4799	130000	3.3676	0.6889
3.6238	5.9014	140000	3.3501	0.6915
3.6004	6.3230	150000	3.3354	0.6939
3.5931	6.7445	160000	3.3241	0.6959
3.5703	7.1660	170000	3.3077	0.6986
3.5616	7.5876	180000	3.3021	0.6993
3.5502	8.0091	190000	3.2892	0.7014
3.5388	8.4306	200000	3.2785	0.7033
3.5264	8.8522	210000	3.2708	0.7046
3.5212	9.2737	220000	3.2598	0.7061
3.5045	9.6952	230000	3.2526	0.7073
3.4939	10.1168	240000	3.2483	0.7087
3.4934	10.5383	250000	3.2361	0.7101
3.4833	10.9598	260000	3.2301	0.7111
3.4747	11.3814	270000	3.2252	0.7120
3.4753	11.8029	280000	3.2172	0.7129
3.46	12.2244	290000	3.2102	0.7141
3.457	12.6460	300000	3.2041	0.7154
3.4464	13.0675	310000	3.1984	0.7163
3.4446	13.4890	320000	3.1933	0.7171
3.4398	13.9106	330000	3.1897	0.7174
3.436	14.3321	340000	3.1838	0.7185
3.4289	14.7536	350000	3.1784	0.7193
3.4223	15.1751	360000	3.1748	0.7198
3.4187	15.5967	370000	3.1676	0.7208
3.414	16.0182	380000	3.1651	0.7216
3.409	16.4397	390000	3.1609	0.7222
3.4022	16.8613	400000	3.1584	0.7226
3.4019	17.2828	410000	3.1511	0.7238
3.395	17.7043	420000	3.1483	0.7241
3.3878	18.1259	430000	3.1473	0.7248
3.3833	18.5474	440000	3.1439	0.7250
3.3828	18.9689	450000	3.1381	0.7260
3.3795	19.3905	460000	3.1349	0.7265
3.3746	19.8120	470000	3.1318	0.7272
3.3704	20.2335	480000	3.1287	0.7275
3.366	20.6551	490000	3.1248	0.7283
3.3621	21.0766	500000	3.1214	0.7286
3.3582	21.4981	510000	3.1189	0.7291
3.3547	21.9197	520000	3.1174	0.7294
3.3561	22.3412	530000	3.1152	0.7298
3.3516	22.7627	540000	3.1145	0.7300
3.3517	23.1843	550000	3.1110	0.7303
3.349	23.6058	560000	3.1087	0.7309
3.3446	24.0273	570000	3.1080	0.7311
3.342	24.4488	580000	3.1042	0.7317
3.3397	24.8704	590000	3.1048	0.7314

Framework versions

• Transformers 4.46.1
• Pytorch 2.2.0+cu121
• Datasets 3.1.0
• Tokenizers 0.20.1