Edit model card

Model description:

Model: pgajo/mbert-xlwa-en-it

Dataset: TASTEset

Unshuffled ratio: ['0']

Shuffled ratio: ['1']

Best exact match epoch: 6

Best exact match: 87.91

Best epoch: 6

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_P0.5_DROP1_mbert

Results

epoch train_loss train_f1 train_exact dev_loss dev_f1 dev_exact test_loss test_f1 test_exact
1 0.98 73.38 55.72 0.51 86.17 77.2 0 0 0
2 0.28 92.14 84.71 0.51 88.01 83.24 0 0 0
3 0.11 96.68 93.46 0.5 89.94 85.71 0 0 0
4 0.06 98.29 96.07 0.48 90.93 86.54 0 0 0
5 0.06 98.54 97.52 0.53 89.68 84.89 0 0 0
6 0.02 99.16 98.62 0.53 90.77 87.91 0 0 0
7 0.03 98.97 98.07 0.52 91.21 87.64 0 0 0
8 0.03 99.05 98.48 0.48 90.62 85.44 0 0 0
9 0.02 99.44 98.9 0.44 91.72 87.09 0 0 0
Downloads last month
2
Safetensors
Model size
177M params
Tensor type
F32