my_awesome_billsum_model_58

This model is a fine-tuned version of google-t5/t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2895
• Rouge1: 0.9839
• Rouge2: 0.9097
• Rougel: 0.944
• Rougelsum: 0.9405
• Gen Len: 4.9167

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	12	0.6061	0.9622	0.8634	0.9034	0.9001	4.9375
No log	2.0	24	0.5474	0.9683	0.8667	0.9081	0.9044	4.8125
No log	3.0	36	0.5017	0.9683	0.8667	0.9081	0.9044	4.8125
No log	4.0	48	0.4739	0.9712	0.8792	0.9167	0.914	4.8333
No log	5.0	60	0.4346	0.9663	0.8708	0.9124	0.9094	4.875
No log	6.0	72	0.3980	0.9663	0.8708	0.9124	0.9094	4.875
No log	7.0	84	0.3772	0.9663	0.8708	0.9124	0.9094	4.875
No log	8.0	96	0.3630	0.9663	0.8708	0.9124	0.9094	4.875
No log	9.0	108	0.3453	0.9651	0.8583	0.9106	0.9064	4.9167
No log	10.0	120	0.3297	0.9651	0.8583	0.9106	0.9064	4.9167
No log	11.0	132	0.3209	0.9651	0.8583	0.9106	0.9064	4.9167
No log	12.0	144	0.3122	0.9651	0.8583	0.9106	0.9064	4.9167
No log	13.0	156	0.3025	0.9738	0.875	0.9232	0.9196	4.9375
No log	14.0	168	0.2975	0.9768	0.8896	0.9339	0.9298	4.9167
No log	15.0	180	0.2979	0.9768	0.8896	0.9339	0.9298	4.9167
No log	16.0	192	0.2983	0.9768	0.8896	0.9339	0.9298	4.9167
No log	17.0	204	0.2967	0.9768	0.8896	0.9339	0.9298	4.9167
No log	18.0	216	0.2930	0.9768	0.8896	0.9339	0.9298	4.9167
No log	19.0	228	0.2877	0.9768	0.8896	0.9339	0.9298	4.9167
No log	20.0	240	0.2861	0.9768	0.8896	0.9339	0.9298	4.9167
No log	21.0	252	0.2896	0.9768	0.8896	0.9339	0.9298	4.9167
No log	22.0	264	0.2940	0.9768	0.8896	0.9339	0.9298	4.9167
No log	23.0	276	0.2912	0.9768	0.8896	0.9339	0.9298	4.9167
No log	24.0	288	0.2849	0.9768	0.8896	0.9339	0.9298	4.9167
No log	25.0	300	0.2879	0.9768	0.8896	0.9339	0.9298	4.9167
No log	26.0	312	0.2953	0.981	0.9125	0.9446	0.9417	4.8958
No log	27.0	324	0.2998	0.981	0.9125	0.9446	0.9417	4.8958
No log	28.0	336	0.2933	0.9839	0.9181	0.9537	0.9512	4.9167
No log	29.0	348	0.2890	0.9798	0.8958	0.9419	0.94	4.9375
No log	30.0	360	0.2895	0.9798	0.8958	0.9419	0.94	4.9375
No log	31.0	372	0.2926	0.9839	0.9181	0.9537	0.9512	4.9167
No log	32.0	384	0.2927	0.9839	0.9181	0.9537	0.9512	4.9167
No log	33.0	396	0.2911	0.9839	0.9181	0.9537	0.9512	4.9167
No log	34.0	408	0.2871	0.976	0.8875	0.9331	0.9296	4.9167
No log	35.0	420	0.2885	0.9827	0.8951	0.9406	0.9384	4.9583
No log	36.0	432	0.2925	0.9869	0.9167	0.9522	0.95	4.9375
No log	37.0	444	0.2902	0.9869	0.9167	0.9522	0.95	4.9375
No log	38.0	456	0.2888	0.9827	0.8951	0.9406	0.9384	4.9583
No log	39.0	468	0.2875	0.9869	0.9167	0.9522	0.95	4.9375
No log	40.0	480	0.2909	0.9869	0.9167	0.9522	0.95	4.9375
No log	41.0	492	0.2920	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	42.0	504	0.2881	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	43.0	516	0.2827	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	44.0	528	0.2777	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	45.0	540	0.2756	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	46.0	552	0.2764	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	47.0	564	0.2799	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	48.0	576	0.2800	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	49.0	588	0.2851	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	50.0	600	0.2896	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	51.0	612	0.2904	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	52.0	624	0.2842	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	53.0	636	0.2826	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	54.0	648	0.2856	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	55.0	660	0.2826	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	56.0	672	0.2881	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	57.0	684	0.2932	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	58.0	696	0.2914	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	59.0	708	0.2936	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	60.0	720	0.2966	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	61.0	732	0.2964	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	62.0	744	0.2948	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	63.0	756	0.2930	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	64.0	768	0.2873	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	65.0	780	0.2879	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	66.0	792	0.2880	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	67.0	804	0.2892	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	68.0	816	0.2894	0.9839	0.9097	0.944	0.9405	4.9167
0.3305	69.0	828	0.2891	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	70.0	840	0.2876	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	71.0	852	0.2877	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	72.0	864	0.2842	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	73.0	876	0.2865	0.9869	0.9167	0.9522	0.95	4.9375
0.3305	74.0	888	0.2840	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	75.0	900	0.2815	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	76.0	912	0.2798	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	77.0	924	0.2813	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	78.0	936	0.2842	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	79.0	948	0.2856	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	80.0	960	0.2863	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	81.0	972	0.2863	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	82.0	984	0.2872	0.98	0.9097	0.9446	0.9413	4.8958
0.3305	83.0	996	0.2879	0.98	0.9097	0.9446	0.9413	4.8958
0.1008	84.0	1008	0.2870	0.98	0.9097	0.9446	0.9413	4.8958
0.1008	85.0	1020	0.2871	0.9869	0.9167	0.9522	0.95	4.9375
0.1008	86.0	1032	0.2868	0.9869	0.9167	0.9522	0.95	4.9375
0.1008	87.0	1044	0.2873	0.9869	0.9167	0.9522	0.95	4.9375
0.1008	88.0	1056	0.2878	0.9869	0.9167	0.9522	0.95	4.9375
0.1008	89.0	1068	0.2887	0.9869	0.9167	0.9522	0.95	4.9375
0.1008	90.0	1080	0.2895	0.9869	0.9167	0.9522	0.95	4.9375
0.1008	91.0	1092	0.2900	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	92.0	1104	0.2908	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	93.0	1116	0.2908	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	94.0	1128	0.2904	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	95.0	1140	0.2901	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	96.0	1152	0.2899	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	97.0	1164	0.2896	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	98.0	1176	0.2895	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	99.0	1188	0.2895	0.9839	0.9097	0.944	0.9405	4.9167
0.1008	100.0	1200	0.2895	0.9839	0.9097	0.944	0.9405	4.9167

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1