my_awesome_billsum_model_68

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: 0.2186
• Rouge1: 0.9718
• Rouge2: 0.8861
• Rougel: 0.9312
• Rougelsum: 0.9298
• Gen Len: 5.0625

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	2.0043	0.3937	0.2702	0.3788	0.3776	17.75
No log	2.0	24	1.4138	0.4258	0.2978	0.4009	0.3998	16.8333
No log	3.0	36	0.8103	0.5858	0.4637	0.5658	0.5614	12.7083
No log	4.0	48	0.5156	0.9539	0.8354	0.8948	0.8934	4.8542
No log	5.0	60	0.4552	0.9539	0.8354	0.8948	0.8934	4.8542
No log	6.0	72	0.4053	0.965	0.8514	0.9092	0.9055	4.8958
No log	7.0	84	0.3565	0.968	0.8653	0.9144	0.9144	4.9167
No log	8.0	96	0.3263	0.968	0.8653	0.9144	0.9144	4.9167
No log	9.0	108	0.2998	0.968	0.8653	0.9144	0.9144	4.9167
No log	10.0	120	0.2807	0.967	0.8788	0.9273	0.9253	4.8958
No log	11.0	132	0.2694	0.967	0.8788	0.9273	0.9253	4.8958
No log	12.0	144	0.2622	0.967	0.8795	0.9273	0.9253	4.9375
No log	13.0	156	0.2490	0.967	0.8795	0.9273	0.9253	4.9375
No log	14.0	168	0.2427	0.967	0.8795	0.9273	0.9253	4.9375
No log	15.0	180	0.2385	0.967	0.8795	0.9273	0.9253	4.9375
No log	16.0	192	0.2350	0.967	0.8795	0.9273	0.9253	4.9375
No log	17.0	204	0.2284	0.967	0.8795	0.9273	0.9253	4.9375
No log	18.0	216	0.2212	0.967	0.8795	0.9273	0.9253	4.9375
No log	19.0	228	0.2173	0.97	0.892	0.936	0.9343	4.9583
No log	20.0	240	0.2177	0.97	0.892	0.936	0.9343	4.9583
No log	21.0	252	0.2161	0.97	0.892	0.936	0.9343	4.9583
No log	22.0	264	0.2111	0.97	0.892	0.936	0.9343	4.9583
No log	23.0	276	0.2072	0.967	0.8799	0.9273	0.9271	4.9792
No log	24.0	288	0.2066	0.97	0.892	0.936	0.9343	4.9583
No log	25.0	300	0.2068	0.973	0.9146	0.9464	0.9435	4.9792
No log	26.0	312	0.2080	0.97	0.892	0.936	0.9343	4.9583
No log	27.0	324	0.2078	0.97	0.892	0.936	0.9343	4.9583
No log	28.0	336	0.1976	0.973	0.8993	0.9346	0.9328	4.9792
No log	29.0	348	0.1921	0.973	0.8993	0.9346	0.9328	4.9792
No log	30.0	360	0.1896	0.973	0.8993	0.9346	0.9328	4.9792
No log	31.0	372	0.1906	0.9686	0.8792	0.9223	0.9204	5.0
No log	32.0	384	0.1942	0.973	0.8993	0.9346	0.9328	4.9792
No log	33.0	396	0.1976	0.97	0.8868	0.926	0.9253	5.0
No log	34.0	408	0.2006	0.97	0.9021	0.9363	0.9353	5.0
No log	35.0	420	0.1983	0.97	0.9021	0.9363	0.9353	5.0
No log	36.0	432	0.2010	0.967	0.8799	0.9273	0.9271	4.9792
No log	37.0	444	0.2014	0.97	0.9021	0.9363	0.9353	5.0
No log	38.0	456	0.2027	0.97	0.9021	0.9363	0.9353	5.0
No log	39.0	468	0.2059	0.97	0.9021	0.9363	0.9353	5.0
No log	40.0	480	0.2035	0.97	0.9021	0.9363	0.9353	5.0
No log	41.0	492	0.1989	0.97	0.8937	0.9363	0.9353	5.0
0.4765	42.0	504	0.1969	0.973	0.892	0.9346	0.933	5.0208
0.4765	43.0	516	0.1958	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	44.0	528	0.1937	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	45.0	540	0.1922	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	46.0	552	0.1940	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	47.0	564	0.1944	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	48.0	576	0.1943	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	49.0	588	0.1985	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	50.0	600	0.2034	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	51.0	612	0.2071	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	52.0	624	0.2113	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	53.0	636	0.2115	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	54.0	648	0.2104	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	55.0	660	0.2109	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	56.0	672	0.2114	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	57.0	684	0.2127	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	58.0	696	0.2149	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	59.0	708	0.2154	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	60.0	720	0.2187	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	61.0	732	0.2193	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	62.0	744	0.2200	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	63.0	756	0.2203	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	64.0	768	0.2192	0.9718	0.8861	0.9326	0.9296	5.0625
0.4765	65.0	780	0.2185	0.9718	0.8708	0.9204	0.9193	5.0625
0.4765	66.0	792	0.2189	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	67.0	804	0.2186	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	68.0	816	0.2181	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	69.0	828	0.2176	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	70.0	840	0.2193	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	71.0	852	0.2198	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	72.0	864	0.2202	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	73.0	876	0.2193	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	74.0	888	0.2191	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	75.0	900	0.2208	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	76.0	912	0.2206	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	77.0	924	0.2193	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	78.0	936	0.2183	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	79.0	948	0.2185	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	80.0	960	0.2176	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	81.0	972	0.2175	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	82.0	984	0.2181	0.9718	0.8861	0.9312	0.9298	5.0625
0.4765	83.0	996	0.2184	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	84.0	1008	0.2172	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	85.0	1020	0.2177	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	86.0	1032	0.2175	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	87.0	1044	0.2180	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	88.0	1056	0.2180	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	89.0	1068	0.2181	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	90.0	1080	0.2179	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	91.0	1092	0.2178	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	92.0	1104	0.2179	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	93.0	1116	0.2175	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	94.0	1128	0.2179	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	95.0	1140	0.2181	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	96.0	1152	0.2182	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	97.0	1164	0.2184	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	98.0	1176	0.2186	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	99.0	1188	0.2186	0.9718	0.8861	0.9312	0.9298	5.0625
0.1106	100.0	1200	0.2186	0.9718	0.8861	0.9312	0.9298	5.0625

Framework versions

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