nllb-200-distilled-600M-finetuned_srimadbhagavatam_sns

This model is a fine-tuned version of facebook/nllb-200-distilled-600M on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9632
• Rouge1: 39.9844
• Rouge2: 15.8187
• Rougel: 24.7601
• Rougelsum: 37.8611

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: 5.6e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
4.2029	1.0	193	3.5530	17.4525	1.8199	14.417	15.7939
3.6789	2.0	386	3.2385	18.4399	2.3063	14.4777	16.8663
3.4121	3.0	579	2.9913	18.6292	2.1671	14.0775	17.4039
3.1958	4.0	772	2.7935	20.9044	3.0869	15.7866	19.4597
3.0238	5.0	965	2.6154	22.9863	3.1733	15.4087	21.6705
2.8546	6.0	1158	2.4343	24.7063	4.0564	16.1424	23.2821
2.7	7.0	1351	2.2810	26.2011	4.6714	16.7887	24.6723
2.5532	8.0	1544	2.1071	30.7319	6.3718	17.4858	28.8254
2.42	9.0	1737	1.9742	28.5217	5.2919	16.9577	26.5686
2.2991	10.0	1930	1.8234	29.8937	6.3088	17.2141	28.0302
2.1851	11.0	2123	1.7177	29.8642	6.9874	18.2935	28.0493
2.0829	12.0	2316	1.5891	30.7551	6.7111	18.1772	28.8555
1.9954	13.0	2509	1.4965	32.6313	8.0662	18.4981	30.8014
1.9055	14.0	2702	1.3996	33.0299	9.6554	19.2763	31.2127
1.8372	15.0	2895	1.3271	35.4767	10.7234	20.2759	33.1856
1.7635	16.0	3088	1.2533	35.5164	11.5198	21.3301	33.2617
1.7052	17.0	3281	1.1865	37.5692	13.6047	22.9496	35.2626
1.6495	18.0	3474	1.1414	37.7493	13.6471	22.6947	35.6368
1.6009	19.0	3667	1.0859	40.251	15.2568	24.4602	37.955
1.5589	20.0	3860	1.0536	37.8875	14.5794	23.4696	35.8989
1.5209	21.0	4053	1.0268	38.4126	14.9535	24.3597	36.435
1.4963	22.0	4246	0.9982	40.9518	16.6418	25.284	38.5787
1.4651	23.0	4439	0.9771	39.4774	16.4189	24.7979	37.3614
1.451	24.0	4632	0.9662	40.4131	16.5895	25.0073	38.3018
1.4351	25.0	4825	0.9632	39.9844	15.8187	24.7601	37.8611

Framework versions

• Transformers 4.28.0
• Pytorch 1.12.1
• Datasets 2.14.4
• Tokenizers 0.13.3