Happy 8th birthday to the Maveryx Community! Take a walk down memory lane in our birthday blog, and don't miss out on the awesome birthday present that all Maveryx Community members get to take advantage of!
After used the new "Image Recognition Tool" a few days, I think you could improve it :
> by adding the dimensional constraints in front of each of the pre-trained models,
> by adding a true tool to divide the training data correctly (in order to have an equivalent number of images for each of the labels)
> at least, allow the tool to use black & white images (I wanted to test it on the MNIST, but the tool tells me that it necessarily needs RGB images) ?
Question : do you in the future allow the user to choose between CPU or GPU usage ?
In any case, thank you again for this new tool, it is certainly perfectible, but very simple to use, and I sincerely think that it will allow a greater number of people to understand the many use cases made possible thanks to image recognition.
The Data Sources page currently lists all the different data sources that Alteryx supports - however for an administrator it's almost impossible to ensure that their designer users have the drivers for these, or are on the right version.
As an early step - can we add 1 more field to this list which points to the downloader for the driver where applicable
It appears that the Workflow Dependencies window does not report dependencies from all tools. In the example image, you can see that the file input from the Amazon S3 Download tool is not listed. Some tools may have dependencies that do not easily fit the current field structure of the window, but maybe the input/download tools could be listed with an asterisk or partial reference.
Please add support to read and write spatial data from a SpatiaLite database. ESRI and QGIS have supported this format out of the box for quite a while. We have a mixed use environment Alteryx, ESRI, Mapinfo and QGIS and would like a common file based spatial format. You already support SQLite, its container, so expanding to the SpatialLite spec should be a no brainer.
It would be interesting to have the ability to change the orientation of a tool container separate from the orientation of the workflow. For example having a vertical tool container within a horizontal workflow will help keep the workflow short enough to view on one page. Also it would look alot better to collapse and expand vertically in a horizontal workflow.
I love that Alteryx lends itself to good workflow documentation, but I'd really like to be able to add a bit of basic formatting within my comment boxes. I tend to have one large (read: verbose) box at the top/beginning of the workflow describing the purpose of the workflow and quirks of the datasource to watch out for, and it would be easier to read these if I had some simple options like Bold, Italic, Underline, numbered list, bullet list. You know, the sorts of things you can do in basic HTML email? Those. I want them!
But it's still to hard to use. It requires you to have pre-knowledge of a bunch of parameters and different types of knowledge.
Can we improve the interface on this tool so that it can be used by folk who do not have a background in R - for example, take all the different inputs, and make them parameterized on drop-down boxes or input boxes on the tool?
It would be nice if this option would take you to the correct download page relative to the version the user has installed. Currently, this always loads the download page for the current version which is confusing for users of a company who are still required to use an older version.
We have a policy (similar to the best practices guide published recently by the Copenhagen user group - thank you @danielbrun2) that Alteryx canvasses should run without warnings.
However - some of the warnings are tremendously painful to track down. For example - a union that brings together 20 or 30 input streams which gives you a warning saying "not all fields contain field X".
Can we change this to multiple warnings saying "Input X does not contain field X" so that we click straight through to the offending warning?
Additionally - when there's a conversion error - we then have to run the flow again with filters and diagnostics to find the offending row. Can we have a way that alteryx automatically sidelines an example row so that you can solve it in one pass?
If we look at all the warnings with the intention of "how do we make this something that can be solved right now, without having to run complex diagnostics, so that the very next run is 100% clean" that would be a big help.
As Alteryx becomes more focussed on the Enterprise - it is important that we build capabilities that support the needs of large-scale BI.
One of these critical needs is dealing with heterogeneous data from different systems that use different IDs for every critical entity / concept (e.g. client; product)
Here's the example:
- In any large enterprise - there are several thousand different line-of business systems
- Each of these was probably built at a different time, and uses a different key for specific concepts - like Client & Product
- Most large enterprises that I've worked at do not have a pre-built way of transforming these codes so...
- This means that any downstream analytics finds it almost impossible to give single-view-of-customer or single-view-of-product.
Solution option A:
Reengineer all upstream systems. Not feasible
Solution option B:
Expect some reference-data team to fix this by building translations. More feasible but not fast
Remaining Solution Option:
Just as Kimball talked about - the only real way is to define a set of enterprise dimensions, which are the defined master-list of critical concepts that you need to slice-and-dice by (client; product; currency; shipping method; etc) in a way which is source-system agnostic
Then you need a method in the middle to transform incoming data to use these codes. This process is called "Conforming"
What would this look like in Alteryx?
We would use the connect product to define a new dimension - say "Product".
Give this a unique ID which is source-system independant; and then add on the attributes that are important for analytics (product type; category; manufacturer; etc)
Then decide how to handle change (slowly changing dimension or SCD type 0,1,2, etc). Alteryx should take full responsibility for managing this SCD history; as do many of the competitors
We then create a list of possible synonym types (within Connect). For example - a product may have a synonym ID from your supplier; from your ERP system; from your point of sale system. that's 3 different IDs for any product.
We then load up the master data - this is painful but necessary
I read in data into alteryx via any input tool
I bring in a "Conforming" tool off the toolbox (new tool which is needed)
It asks me which column or columns I wish to conform
For each - it asks me which synonym type to use
It then adds a translated column for me to use which ties back to the enterprise dimension - and spits out the errors where the synonym is necessary.
In BI in smaller contexts, or quick rapid-fire BI - you don't have to worry about this. But as soon as you go past a few hundred line-of-business systems and are trying to do enterprise reporting, you really have to take this serious. This is a HUGE part of every BI persons's role in a large enterprise - and it is painful; slow and not very rewarding. If we could create this idea of a simple-to-use and high-velocity conforming process - this would absolutely tear the doors off enterprise BI - and no-one else is doing this yet!
Can we get a more robust read.Alteryx function for mode="data.frame"? If it is reading the stream as a data frame, can we have the option stringsAsFactors = FALSE?
I am getting tripped up a lot because the code will execute in R Studio, but will get mysterious behaviours when it runs within the R Tool. I am manually converting variables to character strings in my R Tool code which I don't have to do in R Studio. However, I'm not a highly detail oriented R developer, so I will miss variable data type conversions and have spent a lot of time going down the wrong path. Also, It makes it difficult to maintain two different scripts for the same routine.
I have started using the glimpse() function in R Tool code, to help catch some data conversions since it writes the output in the message log.
Recently my trial license expired, and after receiving our activation licenses, I needed to upgrade from the 'User' version to the 'Admin' version.
When uninstalling the previous 'User' version, it appeared to complete successfully, and I then started installing the 'Admin' version from the installer.
After installation completed, I noticed that three versions of Alteryx appeared in my Windows 'Programs and Features' in Control Panel - Alteryx (Remove Only), Alteryx (Admin), and Alteryx (User).
When attempting to uninstall the 'User' version, it appeared to complete as expected, but did nothing when examining the application or root directory location, and I could still access and open and use the 'User' version of Alteryx.
To make a long story short, after going into regedit and removing all registry keys named 'Alteryx' including the SRC source key for Alteryx, I was able to manually remove all three of these versions and reinstall the 'Admin' application successfully.
Some kind of an uninstaller to wipe out all Alteryx directories, installations/versions, etc. to perform a clean uninstall/reinstall would be super valuable and probably save some folks a ton of time who are evaluating the product and moving from a Trial(User) version to an activated Admin license.