pgajo/mdeberta-xlwa-en-it_EW-TT-PE_U0_S1_Tingredient_P0.25_DROP1_mdeberta_E5_DEV95.0

Model description:

Model: pgajo/mdeberta-xlwa-en-it

Dataset: TASTEset

Unshuffled ratio: ['0']

Shuffled ratio: ['1']

Best exact match epoch: 5

Best exact match: 94.78

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_P0.25_DROP1_mdeberta

Results

epoch	train_loss	train_f1	train_exact	dev_loss	dev_f1	dev_exact	test_loss	test_f1	test_exact
1	0.55	85.31	75.14	0.25	93.72	90.11	0	0	0
2	0.1	97.69	95.66	0.25	95.5	92.31	0	0	0
3	0.05	98.74	97.73	0.27	95.6	92.86	0	0	0
4	0.04	99.26	98.76	0.3	94.86	93.13	0	0	0
5	0.02	99.8	99.45	0.28	96.53	94.78	0	0	0
6	0.03	99.23	98.21	0.3	94.39	91.21	0	0	0
7	0.06	98.14	96.83	0.3	95.93	93.41	0	0	0
8	0.02	99.6	99.24	0.3	95.51	93.41	0	0	0