Unnecessary Model Download Due to Incorrect File Path

While using the Topaz Video AI software, I encountered a bug that causes the software to attempt to re-download a model file from the network each time it is launched, even though the corresponding model file already exists in the local path. This not only wastes a significant amount of time but also unnecessarily burdens the network bandwidth.

Reproduction Steps:

  1. Launch the Topaz Video AI software.
  2. The software attempts to load the model file alq-v13-fgnet-fp16-576x384-1x-ox.tz.
  3. Due to a minor error in the file path, the software fails to correctly identify the existing local model file, resulting in a re-download from the network.

Expected Behavior: The software should be able to recognize and load the model file that exists in the local path without downloading from the network.

Actual Behavior: The software fails to recognize the local model file because an extra forward slash / incorrectly appears before the filename in the file path.

Log Information: Here is the excerpt of the relevant log that shows the process of the error:

2024-06-13 09:22:57 19556 CRITICAL: Loading default error reading file: C:\ProgramData\Topaz Labs LLC\Topaz Video AI\models\/alq-v13-fgnet-fp16-576x384-1x-ox.tz
:AIERROR: Model Failed

2024-06-13 09:23:05 19556 INFO: RESULT for https://veai-models.topazlabs.com/alq-v13-fgnet-fp16-576x384-1x-ox.tz is 200
2024-06-13 09:23:05 19556 INFO: Compressed Buffer for alq-v13-fgnet-fp16-576x384-1x-ox.tz verified
2024-06-13 09:23:05 19556 INFO: putting fetched file into: C:\ProgramData\Topaz Labs LLC\Topaz Video AI\models\
:AIINFO: Model Downloaded

Additional Information:

  • Software Version: [5.1.2]
  • Operating System: [win10]

Suggestion: Fix the file path recognition issue to ensure that the software can correctly load the existing local model file and avoid unnecessary network downloads.

Does setting the folder in the preferences not clear the problem?

YES, this folder have all the model.

Can you share the logs from the app for review?