A new 'Early Adopter' (EA) release of the MVTec Deep Learning Tool (DLT) is now available for download.
The Deep Learning Tool v0.4.3 EA allows users to evaluate and compare their trained networks directly in the tool. For this, a new Evaluation Tab has been added. This tab provides information about model precision including a heatmap for the predicted classes of all processed images, as well as an interactive confusion matrix that helps to track down wrong classifications. Users can also calculate the estimated inference time per image and export the evaluation results as a single HTML page.
For more details, please refer to the release notes.
Please also note that HALCON 20.11 includes a new example program that demonstrates how to import a DLT-trained model into HALCON.
How to get it?
To get the new version, please download the updated online installer (MVTec Software Manager) from the MVTec website. Please note that earlier versions of our online installer (mvp.exe) can no longer be used to download new releases and should be replaced with the new MVTec Software Manager. For users who cannot use the online installer, we also provide an offline installer.
what is an early adopter release?
With these intermediate releases we want to encourage interested users to try our latest functionality and to provide valuable feedback to the development team. Early Adopter versions are available in addition to regular releases.