Logeswaransr/T5_MineAI_Prototype

results_T5

This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8755
• Rouge1: 0.2921
• Rouge2: 0.1519
• Rougel: 0.2857
• Rougelsum: 0.2866

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 4
• 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
No log	1.0	109	0.4682	0.2349	0.0878	0.2327	0.2344
No log	2.0	218	0.4153	0.2519	0.0965	0.2481	0.2503
No log	3.0	327	0.4102	0.3011	0.1465	0.2979	0.2990
No log	4.0	436	0.4386	0.2555	0.1138	0.2496	0.2496
0.8199	5.0	545	0.4784	0.2725	0.1188	0.2675	0.2665
0.8199	6.0	654	0.5088	0.2524	0.1066	0.2497	0.2501
0.8199	7.0	763	0.5680	0.2542	0.1093	0.2497	0.2496
0.8199	8.0	872	0.5982	0.2740	0.1375	0.2694	0.2698
0.8199	9.0	981	0.6575	0.2730	0.1368	0.2723	0.2714
0.0653	10.0	1090	0.6753	0.2822	0.1519	0.2798	0.2781
0.0653	11.0	1199	0.6923	0.2795	0.1486	0.2780	0.2774
0.0653	12.0	1308	0.7350	0.2471	0.1209	0.2458	0.2457
0.0653	13.0	1417	0.7698	0.2762	0.1463	0.2720	0.2733
0.0225	14.0	1526	0.7867	0.2771	0.1372	0.2763	0.2755
0.0225	15.0	1635	0.8166	0.3166	0.1689	0.3132	0.3133
0.0225	16.0	1744	0.8085	0.3027	0.1572	0.2998	0.3009
0.0225	17.0	1853	0.8162	0.3090	0.1734	0.3025	0.3038
0.0225	18.0	1962	0.8484	0.2965	0.1627	0.2917	0.2909
0.0105	19.0	2071	0.8610	0.2881	0.1487	0.2813	0.2819
0.0105	20.0	2180	0.8688	0.2811	0.1494	0.2755	0.2770
0.0105	21.0	2289	0.8733	0.2777	0.1453	0.2708	0.2724
0.0105	22.0	2398	0.8776	0.2771	0.1475	0.2709	0.2711
0.0061	23.0	2507	0.8717	0.2829	0.1467	0.2749	0.2749
0.0061	24.0	2616	0.8729	0.2878	0.1467	0.2803	0.2806
0.0061	25.0	2725	0.8755	0.2921	0.1519	0.2857	0.2866

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3