my_awesome_billsum_model_78

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: 0.5080
• Rouge1: 0.9792
• Rouge2: 0.8868
• Rougel: 0.9405
• Rougelsum: 0.94
• Gen Len: 4.9792

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	0.4089	0.9821	0.9104	0.9484	0.9484	4.9583
No log	2.0	24	0.4068	0.9821	0.9104	0.9484	0.9484	4.9583
No log	3.0	36	0.4284	0.9792	0.8868	0.9405	0.94	4.9792
No log	4.0	48	0.4548	0.9792	0.8903	0.9395	0.9405	5.0208
No log	5.0	60	0.4590	0.9792	0.8903	0.9395	0.9405	5.0208
No log	6.0	72	0.4543	0.9792	0.8903	0.9395	0.9405	5.0208
No log	7.0	84	0.4863	0.9752	0.8708	0.9311	0.9311	5.0417
No log	8.0	96	0.4935	0.9732	0.8569	0.9221	0.9216	5.0208
No log	9.0	108	0.4931	0.9762	0.8691	0.9311	0.9311	5.0
No log	10.0	120	0.4817	0.9762	0.8691	0.9311	0.9311	5.0
No log	11.0	132	0.4741	0.9762	0.8691	0.9311	0.9311	5.0
No log	12.0	144	0.4732	0.9762	0.8691	0.9311	0.9311	5.0
No log	13.0	156	0.4742	0.9762	0.8691	0.9311	0.9311	5.0
No log	14.0	168	0.4736	0.9792	0.8903	0.9395	0.9405	5.0208
No log	15.0	180	0.4680	0.9792	0.8903	0.9395	0.9405	5.0208
No log	16.0	192	0.4534	0.9821	0.9007	0.9479	0.9494	5.0
No log	17.0	204	0.4412	0.9821	0.9007	0.9479	0.9494	5.0
No log	18.0	216	0.4341	0.9821	0.9007	0.9479	0.9494	5.0
No log	19.0	228	0.4317	0.9821	0.9007	0.9479	0.9494	5.0
No log	20.0	240	0.4315	0.9821	0.9007	0.9479	0.9494	5.0
No log	21.0	252	0.4313	0.9792	0.8903	0.9395	0.9405	5.0208
No log	22.0	264	0.4277	0.9792	0.8868	0.9405	0.94	4.9792
No log	23.0	276	0.4376	0.9792	0.8868	0.9405	0.94	4.9792
No log	24.0	288	0.4432	0.9792	0.8868	0.9405	0.94	4.9792
No log	25.0	300	0.4450	0.9792	0.8868	0.9405	0.94	4.9792
No log	26.0	312	0.4468	0.9792	0.8868	0.9405	0.94	4.9792
No log	27.0	324	0.4415	0.9792	0.8868	0.9405	0.94	4.9792
No log	28.0	336	0.4560	0.9792	0.8868	0.9405	0.94	4.9792
No log	29.0	348	0.4713	0.9792	0.8868	0.9405	0.94	4.9792
No log	30.0	360	0.4732	0.9792	0.8868	0.9405	0.94	4.9792
No log	31.0	372	0.4726	0.9792	0.8868	0.9405	0.94	4.9792
No log	32.0	384	0.4682	0.9792	0.8868	0.9405	0.94	4.9792
No log	33.0	396	0.4647	0.9792	0.8868	0.9405	0.94	4.9792
No log	34.0	408	0.4644	0.9792	0.8868	0.9405	0.94	4.9792
No log	35.0	420	0.4657	0.9821	0.9007	0.9479	0.9494	5.0
No log	36.0	432	0.4643	0.9821	0.9007	0.9479	0.9494	5.0
No log	37.0	444	0.4572	0.9821	0.9007	0.9479	0.9494	5.0
No log	38.0	456	0.4447	0.9821	0.9007	0.9479	0.9494	5.0
No log	39.0	468	0.4437	0.9821	0.9007	0.9479	0.9494	5.0
No log	40.0	480	0.4684	0.9821	0.9007	0.9479	0.9494	5.0
No log	41.0	492	0.4722	0.9821	0.9007	0.9479	0.9494	5.0
0.0088	42.0	504	0.4716	0.9821	0.9007	0.9479	0.9494	5.0
0.0088	43.0	516	0.4803	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	44.0	528	0.4854	0.9762	0.8691	0.9311	0.9311	5.0
0.0088	45.0	540	0.4830	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	46.0	552	0.4819	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	47.0	564	0.4812	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	48.0	576	0.4806	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	49.0	588	0.4762	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	50.0	600	0.4737	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	51.0	612	0.4735	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	52.0	624	0.4738	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	53.0	636	0.4736	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	54.0	648	0.4738	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	55.0	660	0.4776	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	56.0	672	0.4866	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	57.0	684	0.4926	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	58.0	696	0.4938	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	59.0	708	0.4902	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	60.0	720	0.4962	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	61.0	732	0.5033	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	62.0	744	0.5043	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	63.0	756	0.5025	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	64.0	768	0.5176	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	65.0	780	0.5708	0.9762	0.8691	0.9311	0.9311	5.0
0.0088	66.0	792	0.5707	0.9762	0.8691	0.9311	0.9311	5.0
0.0088	67.0	804	0.5278	0.9762	0.8691	0.9311	0.9311	5.0
0.0088	68.0	816	0.5179	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	69.0	828	0.5164	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	70.0	840	0.5504	0.9762	0.8691	0.9311	0.9311	5.0
0.0088	71.0	852	0.5584	0.9762	0.8691	0.9311	0.9311	5.0
0.0088	72.0	864	0.5281	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	73.0	876	0.5198	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	74.0	888	0.5176	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	75.0	900	0.5103	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	76.0	912	0.5068	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	77.0	924	0.5030	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	78.0	936	0.5025	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	79.0	948	0.4968	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	80.0	960	0.5113	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	81.0	972	0.5083	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	82.0	984	0.5031	0.9792	0.8868	0.9405	0.94	4.9792
0.0088	83.0	996	0.5066	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	84.0	1008	0.5177	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	85.0	1020	0.5192	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	86.0	1032	0.5104	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	87.0	1044	0.5085	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	88.0	1056	0.5130	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	89.0	1068	0.5116	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	90.0	1080	0.5081	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	91.0	1092	0.5074	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	92.0	1104	0.5090	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	93.0	1116	0.5097	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	94.0	1128	0.5123	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	95.0	1140	0.5118	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	96.0	1152	0.5089	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	97.0	1164	0.5080	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	98.0	1176	0.5079	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	99.0	1188	0.5076	0.9792	0.8868	0.9405	0.94	4.9792
0.0059	100.0	1200	0.5080	0.9792	0.8868	0.9405	0.94	4.9792

Framework versions

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