Is there a guide on how to do document classification with custom models? I'm using Form Recognizer Studio and I can only train one model from all the uploaded images. I want to train 8 different classes/models and then identify uploaded document type.
I don't want to extract any fields, I just want to do document classification. Do I still need to label the documents? I saw there is possibility to train without labels but I can't see the option in Studio.
Related
I need to design a venue/arena layout(for Example MSG Arena) given an SVG or JSON representation of the layout. Much like ticketing systems do but I need further features. I want to change the layout(edit a few sections, change the capacity of a certain section, add a few more seats, etc).
Is there any tool or package that can allow an interactive venue layout(from image or JSON or any other layout representation)? The tool or package needs to integrate with React typescript.
How does the Google Analytics connector add the hierarchical options tree to select which dataset to use? Is it a custom control?
We have the same need right now, where we want the user to choose multiple items, with database->Node->Item tree of over 20000 items. Ideally we would also like to search for elements.
In tableau, you implement a html page with js backing, so you can put any controls you want in this section, but the Data Studio seems to have just it's own subset.
Any hints would be greatly appreciated, as this is a blocker in using the Studio.
For Data Studio Community Connectors, dynamic or hierarchical config is not supported yet. For now, you can either provide a flat list or a type in option.
If you require searchable options and hierarchical configs, please put in feature requests / vote for them in the issue tracker.
Edit: Updated link.
I realized a GUI. My web page consists of 8 graphics. There is a barchart for dates, a map for the country, a datatable for companies, etc.
Now I wish to establish the interactions between graphics. For example, if the user clicks the date 2010, so you have the map and the table displays the results of the 2010 data.
(I think I have to make a $ scope in a controller that will contain all the data, but after how to recover the data for each chart, and then to say "if click on 2010 when ..."?)
Can you please tell me how to do, what steps?
Thank you !
Is it possible to upload multiple files (images in particular) using CakePHP Uploader and HTML5 multiple attribute? And of course save those images with hasMany relation, eg Project hasMany Image.
EDIT:
I am using CakePHP 2.x, this is the Uploader. I have tried adding multiple attribute to input, but seems that is not working, only saves 1 image. Miles is suggesting multiple inputs, I wanted to use only one input field for this matter. So the question above stands.
The simple answer is: Not out of the box. Multiupload widgets are actually kind of tricky (as you need to change the behavior of the upload panel. My weapon of choice for this is: http://blueimp.github.io/jQuery-File-Upload/
A very simple javascript example is provided here: http://kajda.com/scripts/multiupload/js/multiupload.js
Alternatively, you could have a field that takes multiple drag and drops (using HTML5) instead of the upload panel. Check this one out:
http://www.9lessons.info/2012/09/multiple-file-drag-and-drop-upload.html
I have a full-text catalog containing PowerPoint presentations. Some of those presentations contain charts, and in those cases, the chart is usually the main thing on the slide. When doing a search, I therefore want to be able to search for the text within the chart (e.g. the labels on a pie chart, the categories on a bar chart, etc.).
I'm disappointed (and surprised) to find that this does not seem to work. I can search for text within the presentation itself, but not for any of the text on the chart.
Please tell me there's just a setting somewhere that I need to change!
[I've tried this for charts created within PowerPoint, and charts created in Excel and then pasted into PowerPoint.]
Gary, the text extraction from Office documents is handled by the Office Filter Pack and as such, there is nothing Full-text can do for embedded data which the filters do not parse.
I successfully repro-ed the problem in my SQL Server 2012 install with the latest Office Filters. A quick work-around would be to add the labels/categories to the Notes section of each slide which gets parsed and filtered and returned to the full-text indexer.