Embedding Page

The embedding page is a 2D representation of all patches in the system, where patches perceived to be similar by the model are plotted closely together, and updates as the model trains.

Embedding Page UI

---

In the embedding plot, there are (Patch NUmber could be viewed in the information bar)circle dots that represent 256 x 256 image patches each. Circle dots of different colors indicate that these dots belong to different image tiles. When hovering over a dot, a preview of the patch is shown and subsequently clicking on the dot takes the user to the annotation page centered around that patch. This practice helps the user to train a robust classifier in a more efficient approach. See more details about Best Practice to use QA more robustly and efficiently.

Information bar

---

The Information bar contains metadata information for the embedding page.

The user could adjust the numbers of patches (Patch Number) plotted on the embedding table by Embed Slider. The user would like to select all(rightmost) on the slider when there are less than 30 images in the project. More images in the project come with potentially overwhelming numbers of image patches embedded for some users at the project level. In this case, the user would like to use the Embed Slider to adjust the number of patches in the embedding plot. Here is a demo for using Embed Slide.

*Latest ID* is the newest model number trained by the user; *Current ID* is the model number that the current embedding plot employs. The user can change *Selected ID* to *Re_embed* a new embedding plot as the user is interested in looking at the embedding plot of other model numbers. For example, the user usually will *Re_embed* as there is a later classier available. The user is also able to *Select ID* and then *View Embedding*, which is very useful to compare the difference between embedding plots of different models

Note: Re_embed does not automatically update the embedding plot, the user needs to click View Embedding to update the page.