Copia de ConfBert_tokenizerFast.txt

optim = AdamW(model.parameters(), lr=5e-5) #tasa de aprendizaje

# Se inicializa el cargador de datos para los datos de entrenamiento
train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)

for epoch in range(9):

Epoch 0: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=2.06]
Epoch 1: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=2.64]
Epoch 2: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=2.48]
Epoch 3: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=0.638]
Epoch 4: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=0.184]
Epoch 5: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=1.78]
Epoch 6: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=0.288]
Epoch 7: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=0.45]
Epoch 8: 100%|██████████| 94/94 [01:18<00:00,  1.19it/s, loss=0.0356]

Precisión del modelo ajustado: 0.786479250334672