Evaluate the model

Getting all together

In summary, there are several steps for the machine learning project

• load model from the library

• specify device configuration that the model will be using

• initialize the model with all components

  • data suitable for machine learning model

  • dataloader, optimizer, scheduler

  • performance evaluation metrics, logging

• train the model

• evaluate the model performance

model = BertForSequenceClassification.from_pretrained("bert-base-uncased",
                                                      num_labels=len(label_dict),
                                                      output_attentions=False,
                                                      output_hidden_states=False)

model.to(device)
pass

model.load_state_dict(
    torch.load('Models/finetuned_bert_epoch_1_gpu_trained.model',
               map_location = torch.device('cpu')
              )
)

_, predictions, true_vals = evaluate(dataloader_val)

accuracy_per_class(predictions, true_vals)