pgajo/mdeberta-xlwa-en-it_EW-TT-PE_U0_S1_Tingredient_P1_DROP1_mdeberta_E5_DEV96.0

Model description:

Model: pgajo/mdeberta-xlwa-en-it

Dataset: TASTEset

Unshuffled ratio: ['0']

Shuffled ratio: ['1']

Best exact match epoch: 5

Best exact match: 96.15

Best epoch: 5

Drop duplicates: ['1']

Max epochs = 10

Optimizer lr = 3e-05

Optimizer eps = 1e-08

Batch size = 8

Dataset path = pgajo/EW-TT-PE_U0_S1_Tingredient_DROP1_mdeberta

Results

epoch	train_loss	train_f1	train_exact	dev_loss	dev_f1	dev_exact	test_loss	test_f1	test_exact
1	0.54	84.47	74.31	0.17	96.61	92.86	0	0	0
2	0.13	96.81	94.35	0.16	96.54	94.23	0	0	0
3	0.07	98.08	97.11	0.12	96.72	95.05	0	0	0
4	0.04	98.55	97.59	0.18	96.82	94.23	0	0	0
5	0.04	99	98.28	0.12	97.47	96.15	0	0	0
6	0.05	98.57	97.59	0.14	97.09	95.33	0	0	0
7	0.03	99.3	98.83	0.12	96.78	95.6	0	0	0
8	0.02	99.44	99.04	0.2	96.61	95.05	0	0	0