pgajo/mbert_EW-TT-PE_U0_S1_Tingredient_P0.25_DROP1_mbert_E10_DEV80.0

Model description:

Model: bert-base-multilingual-cased

Dataset: TASTEset

Unshuffled ratio: ['0']

Shuffled ratio: ['1']

Best exact match epoch: 10

Best exact match: 80.22

Best epoch: 10

Drop duplicates: ['1']

Max epochs = 10

Optimizer lr = 3e-05

Optimizer eps = 1e-08

Batch size = 32

Dataset path = pgajo/EW-TT-PE_U0_S1_Tingredient_P0.25_DROP1_mbert

Results

epoch	train_loss	train_f1	train_exact	dev_loss	dev_f1	dev_exact	test_loss	test_f1	test_exact
1	3.18	11.43	2.48	2.18	29.53	20.6	0	0	0
2	1.31	55.71	40.91	1.12	72.8	58.24	0	0	0
3	0.57	81.62	72.45	1.05	79.04	71.7	0	0	0
4	0.29	90.38	84.37	1.01	81.04	74.73	0	0	0
5	0.19	93.63	88.98	0.92	79.94	75	0	0	0
6	0.11	96.49	94.21	0.97	81.43	75.82	0	0	0
7	0.08	97.23	95.59	0.94	83.21	78.57	0	0	0
8	0.04	98.64	97.87	1.13	83.51	78.3	0	0	0
9	0.04	98.8	97.87	1.11	83.88	78.57	0	0	0
10	0.05	98.23	97.52	0.89	84.79	80.22	0	0	0