FAQ

Frequently asked questoins and answers about Neural Network Console are available here.
Check them first before contacting us.

Setup

  • Q.Not all the files can be unzipped corredctly.
  • Q.“Access denied” is displayed, and the program will not start.
  • Q.“The ordinal 4540 could not be located in the dynamic link library LIBEAY32.dll” is displayed when the program is started.
  • Q.“DLL load failed: The specified module could not be found” is displayed, and training cannot be executed.
  • Q.“CUDA driver version is insufficient for CUDA runtime version” is displayed, and training cannot be executed.
  • Q.An error occurs when training is executed with a sample project.
  • Q.I cannot perform training using a GPU.
  • Q.Training is slow. Is there a way to know whether the GPU is working properly?
  • Q.The GPU environment cannot be set up or training using a GPU does not work on a remote desktop environment.

> Answers

Datasets

  • Q.Can image data with 16-bit brightness resolution be used?
  • Q.Can images or data with different sizes be used?

> Answers

During Training Execution

  • Q.“Creating cache data…” takes an extremely long amount of time.
  • Q.Loss and error on the validation data do not decrease monotonically.
  • Q.Training does not progress properly such as NaN being displayed for the evaluation values.
  • Q.“MemoryError,” or “Out of memory” appears in the log output.
  • Q.After loading trained parameters and editing the network, an error occurs when training is started.
  • Q.An error appears in the log output, but I do not understand the meaning.

> Answers

During Evaluation Execution

  • Q.Can test data other than training data or validation data be specified?
  • Q.When evaluation is executed, it appears that evaluation calculation is being performed to the end, but nothing appears on the EVALUATION tab.
  • Q.In Confusion Matrix, all the data are classified to a certain cell.

> Answers

Other FAQs

  • Q.The display driver stops responding, and the GUI display becomes crooked.
  • Q.Loading of trained parameters cannot be performed correctly.

> Answers