flan-t5-rouge-durga-q5-clean-3

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

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0003
train_batch_size: 6
eval_batch_size: 6
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 30

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
2.4075	1.0	25	1.3986	0.2974	0.0955	0.2908	0.2922
1.7776	2.0	50	0.9113	0.3446	0.1314	0.3321	0.3327
1.1018	3.0	75	0.5636	0.3736	0.1990	0.3647	0.3664
0.62	4.0	100	0.3676	0.4405	0.2807	0.4313	0.4312
0.5543	5.0	125	0.2240	0.4433	0.3035	0.4374	0.4391
0.6031	6.0	150	0.1842	0.4202	0.2714	0.4186	0.4199
0.3551	7.0	175	0.1245	0.4640	0.3485	0.4639	0.4645
0.3913	8.0	200	0.1189	0.4526	0.3303	0.4542	0.4543
0.1866	9.0	225	0.0837	0.5110	0.3965	0.5065	0.5078
0.2824	10.0	250	0.0652	0.5409	0.4593	0.5417	0.5430
0.1881	11.0	275	0.0579	0.5475	0.4585	0.5471	0.5489
0.0936	12.0	300	0.0487	0.6001	0.5283	0.6013	0.6025
0.1443	13.0	325	0.0401	0.6235	0.5599	0.6251	0.6254
0.1081	14.0	350	0.0331	0.6252	0.5568	0.6271	0.6271
0.0567	15.0	375	0.0295	0.6447	0.5794	0.6439	0.6449
0.1883	16.0	400	0.0205	0.6969	0.6496	0.6939	0.6963
0.0633	17.0	425	0.0230	0.6875	0.6370	0.6864	0.6870
0.0601	18.0	450	0.0152	0.7016	0.6554	0.7006	0.7014
0.0938	19.0	475	0.0150	0.7094	0.6648	0.7081	0.7095
0.05	20.0	500	0.0104	0.7209	0.6805	0.7204	0.7219
0.0556	21.0	525	0.0108	0.6988	0.6495	0.6961	0.6985
0.0248	22.0	550	0.0075	0.7191	0.6794	0.7168	0.7189
0.0247	23.0	575	0.0059	0.7255	0.6888	0.7238	0.7256
0.0186	24.0	600	0.0064	0.7283	0.6945	0.7271	0.7280
0.0338	25.0	625	0.0051	0.7277	0.6936	0.7261	0.7268
0.0377	26.0	650	0.0044	0.7252	0.6885	0.7237	0.7251
0.022	27.0	675	0.0044	0.7273	0.6936	0.7265	0.7272
0.0104	28.0	700	0.0056	0.7249	0.6886	0.7239	0.7258
0.0117	29.0	725	0.0052	0.7286	0.6950	0.7274	0.7286
0.0228	30.0	750	0.0047	0.7284	0.6952	0.7275	0.7286