pgajo/mbert-xlwa-en-it_EW-TT-PE_U1_S0_DROP1_mbert_E8_DEV98.0

Model description:

Model: pgajo/mbert-xlwa-en-it

Dataset: TASTEset

Unshuffled ratio: ['1']

Shuffled ratio: ['0']

Best exact match epoch: 8

Best exact match: 98.07

Best epoch: 8

Drop duplicates: ['1']

Max epochs = 10

Optimizer lr = 3e-05

Optimizer eps = 1e-08

Batch size = 32

Dataset path = pgajo/EW-TT-PE_U1_S0_DROP1_mbert

Results

epoch	train_loss	train_f1	train_exact	dev_loss	dev_f1	dev_exact	test_loss	test_f1	test_exact
1	0.42	88.03	77.33	0.08	97.54	95.58	0	0	0
2	0.05	99.22	97.72	0.05	98.33	97.24	0	0	0
3	0.02	99.66	99.1	0.07	98.37	96.69	0	0	0
4	0.02	99.61	99.1	0.06	98.43	96.96	0	0	0
5	0.01	99.69	99.31	0.05	98.72	97.51	0	0	0
6	0.01	99.75	99.38	0.03	98.62	97.24	0	0	0
7	0.01	99.97	99.86	0.04	98.83	97.79	0	0	0
8	0	99.91	99.86	0.04	98.98	98.07	0	0	0
9	0	99.88	99.79	0.03	99.22	98.07	0	0	0
10	0	99.88	99.72	0.05	98.84	97.51	0	0	0