Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

3. [SMART SORTING] - Click on the ‘SMART SORTING’ button, and type what’s the prefix for the

Status
colourGreen
titleok
images within the field. There is also an option to use regular expressions instead of prefixes if you check the ‘Regex’ checkbox or you can filter images by their tags.

...

Info

Tip: It is also possible to select multiple images at once using the ‘Shift’ key and classify them in bulk, by selecting the first image, then holding 'Shift' and selecting the last image from the range.

...

After training, the threshold is only calculated based on training annotated images. For better separation of

Status
colourGreen
titleok
and
Status
colourRed
titleng
images, you should click on Recalculate threshold and classify both
Status
colourGreen
titleok
and
Status
colourRed
titleng
(Not Good/defective) images in the dialog window that will be used for the calculation of the new thresholdgo to labeling, annotate more images and click on RECAL MATRIX - you can learn more about the matrix on the Confusion Matrix page.

As a result, you get a graph containing two curves. The green curve shows how many

Status
colourGreen
titleok
images are wrongly classified as
Status
colourRed
titleng
for a given threshold value. The red curve shows how many
Status
colourRed
titleng
images are wrongly classified as
Status
colourGreen
titleok
for a given threshold value. Ideally, you want to choose a threshold at which both curves are minimized. It is possible to choose the threshold by clicking on the graph.

...

Additionally, you get a confusion matrix showing whether the Testing images are evaluated correctly. Again, you can find more on the Confusion Matrix page.

...