pgajo/mdeberta-xlwa-en-it_EW-TT-PE_U0_S1_Tingredient_P0.75_DROP1_mdeberta_E4_DEV98.0

Model description:

Model: pgajo/mdeberta-xlwa-en-it

Dataset: TASTEset

Unshuffled ratio: ['0']

Shuffled ratio: ['1']

Best exact match epoch: 4

Best exact match: 97.8

Best epoch: 4

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_P0.75_DROP1_mdeberta

Results

epoch	train_loss	train_f1	train_exact	dev_loss	dev_f1	dev_exact	test_loss	test_f1	test_exact
1	0.39	90.33	82.92	0.14	97.6	96.43	0	0	0
2	0.09	98.28	96.49	0.15	97.23	96.15	0	0	0
3	0.05	98.86	98.07	0.12	97.96	96.98	0	0	0
4	0.02	99.22	98.9	0.15	98.27	97.8	0	0	0
5	0.04	99.32	98.35	0.15	97.45	96.43	0	0	0
6	0.02	99.64	99.04	0.19	97.74	96.43	0	0	0
7	0.01	99.84	99.59	0.16	97.69	96.7	0	0	0