Errors in test output
Opened this issue · 2 comments
roopavsm commented
The following are the errors occurring frequently in the output using the image analysis tool:
- Subscripts are not read accurately. Eg: Cs
- The letters in the beginning and ending are missing. Eg: cetyl-C
- Words having hyphen symbols are not read completely(mostly in the end). Eg: hex-3-en, 3-Hydroxy-3-met
- Greek letters are truncated. Eg: -ketoacyl-A
- Numbers are not read completely. Eg: 1.2.1.3 (in the image) -> 1.2. (in the result)
(4,1,3,29) -> (4,1,
petermr commented
Thanks!
~The hyphens are a problem on their own. Suggest you create a small number of images where this is a problem and open a new issue on "subscripts" and point to the examples, as it will need new code.
roopavsm commented
https://user-images.githubusercontent.com/91625756/157455062-e48ee946-16be-47cd-96bd-d81ab05d979a.jpg
https://user-images.githubusercontent.com/91625756/157455156-d2d51d09-2fd5-4ee8-a95e-635f76536d70.png
https://user-images.githubusercontent.com/91625756/157455438-3c608865-79a9-46d0-b34c-cc4ca9f7d55e.jpeg