Can test data other than training data or validation data be specified?
You can specify separately the evaluation data for training and the test data to use when you click the Run Evaluation button by adding a new dataset from the DATASET tab, loading test data, and specifying the test dataset added to the dataset under Executor (settings for when the Run Evaluation button is clicked) on the CONFIG tab.
When evaluation is executed, it appears that evaluation calculation is being performed to the end, but nothing appears on the EVALUATION tab.
The I/O pattern may not be supported by the current evaluation function. Right-click the training result list to open a shortcut menu, click Open Result Location, and open the folder containing the training result file. By opening output_result.csv, you can directly view the evaluation results.
In Confusion Matrix, all the data are classified to a certain cell.
It may be that optimization is not being performed properly to begin with, and the obtained results may be the actual values. View the learning curve, and check that the loss and error values are steadily decreasing as epochs are repeated. Consider expanding the number of data samples used in training or changing the network structure.
