Be sure to review our Idea Submission Guidelines for more information!
Submission GuidelinesHello,
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.
Thank you again
Kévin VANCAPPEL (France ;-))
Thank you again.
Kévin VANCAPPEL
When I add a data connection to my canvas - it's only added to the Data Connections window under certain circumstances (e.g. when I use an alias, or the SQL connection wizard) rather than showing ALL data connections.
Given the importance of data connections for Alteryx flows - it would be better if ALL data connections were grouped together under a Data Connection Manager, which was as visible as the results window not buried deep in the menu system - and you could also then use this spot to change; share; alias etc.
In Microsoft SSIS there's a useful example of how this could be done - where the connections are very visibly a collection of assets that can be seen and updated centrally in one place. So if you have 5 input tools which ALL point to the same database - you only need to update the connection on your designer in one place - irrespective of whether this is a shared connection or not.
One of the biggest areas of time spent is in basic data cleaning for raw data - this can be dramatically simplified by taking a hint from the large ETL / Master data Management vendors and making this core Alteryx.
Server Side
- Allow the users of the server & connect product to define their own Business Types (what Microsoft DQS calls "Domains")
- Example may be a currency code - there are many different synonyms, but in essence you want your data all cleaned back to one master list
- Then allow for different attributes to be added to these business types
- Currency code would have 2 or 3 additional columns: Currency name; Symbol; Country of issue
- Similar to Microsoft DQS - allow users to specify synonyms and cleanup rules. For example - Rupes should be Rupees and should be translated to INR
- You also need cross business type rules - if the country is AUS then $ translates to AUD not to USD.
- These rules are maintained by the Data Steward responsibility for this Business Type.
- This master data needs to be stored and queryable as a slowly changing dimension (preferrably split into a latest & history table with the same ID per entry; and timestamps and user audit details for changes)
Alteryx Designer:
- When you get a raw data set - user can then tag some fields as being one of these business types
- Example: I have a field bal_cur (Balance Currency) - I tag this as Business Type "Currency"
- Then Alteryx automatically checks the data; and applies my cleanup rules which were defined on the server
- For any invalid entries - it marks these as an error in the canvas; and also adds them to a workflow for the data steward for this Business Type on the server - value is set to an "unmapped" value. (ID=-1; all text columns set to "unmapped")
- For any valid entries - it gives you the option to add which normalised (conformed) columns you want - currency code; description; ID; symbol; country of issue
Data Steward Workflow:
- The data steward is notified that there is an invalid value to be checked
- They can either mark this as a valid value (in which case this will be added to the knowledge base for this business type) or a synonym of some other valid value; or an invalid value
Cleanup Audit & Logs:
- In order to drive upstream data cleaning over time - we would need to be able to query and report on data cleanups done by source; by canvas; by user; by business type; and by date - to report back to the source system so that upstream data errors can be fixed at source.
Many thanks
Sean
When using the Cross Tab tool, the Count option under Method for Aggregating Values only appears when the field selected for Values for New Columns is a numeric field. I can add a numeric RecordID before the Cross Tab and then count on that, but why can't the Cross Tab count non-numeric fields? The Summarize tool can.
I suppose I could just bookmark this page, but that wouldn't help others. I frequently forget (I'm getting old) the format strings while creating custom datetime formulas. Is there a quick way to get to these format strings when in the context of creating a datetimeparse/datetimeformat formula?
Cheers,
Mark
Regularly put true in or false in expecting it to work in a formula.
This request is largely based on the implementation found on AzureML; (take their free trial and check out the Deep Convolutional and Pooling NN example from their gallery). This allows you to specify custom convolutional and pooling layers in a deep neural network. This is an extremely powerful machine learning technique that could be tricky to implement, but could perhaps be (for example) a great initial macro wrapped around something in Python, where currently these are more easily implemented than in R.
The COMMENT tool has a number of default settings. It would save me a lot of time if the Designer could remember the last settings used.
This is the panel I am referring to:
Hi there
My idea is to have an option to copy and paste a tools configuration to a different tool of the same type somewhere else on the canvas.
Example:
Say I have four summarise tools dealing with four different data streams, I envision a 'Copy Tool Configuration' option after right clicking on a tool and then a 'Paste Tool Configuration" which can be applied across the multiple instances of the summarise tool by overwriting. This would preserve the tools anchors incoming and outgoing connections.
Benefit:
This would increase the speed of developing workflows. Naturally this would be significantly quicker than copy and pasting tools, and then re-wiring anchors. Additionally, this would potentially reduce human error when iteratively developing workflows.
Regards - Rhys Cooper
Our Alteryx users query a number of different data sources. Some of these include external servers outside our control.
To avoid any issues regarding locking, we use the Read Uncommitted function as part of the Data Input tool as part of our baseline design, probably 95% of the time or more.
It would be very beneficial for our organization if there was a way for us to set this option to be checked by default, so that it was one less thing users needed to remember when configuring a quick data pull.
When moving external data into the database, the underlying SQL looks like:
CREATE GLOBAL TEMPORARY TABLE "AYX16020836880b41e08246b59ee8c"
...
My client would like to add a prefix to the table as:
CREATE GLOBAL TEMPORARY TABLE MMMM999_DM_USER."AYX16020836880b41e08246b59ee8c"
where MMMM999_DM_USER is supplied in the configuration.
A service account automatically sets the current session to something like MMMM999 (alter session set curent schema=MMMM999;)
Hello all,
This is a very interesting feature of the List Box and Drop Down interface tool : the ability to select fields
However such a feature is not available for in-database, highly limiting the use of macros.
Please change.
Best regards,
Simon
I sometimes have to swap (change the order of) two tools in a flow. It is a bothersome task, especially when there are many connections around them. I would like to suggest two new features for such a situation. It would help if either could be realized.
Swap tools
Select two tools, right-click, and select "Swap" option.
Move and connect around
Drag a tool holding down Alt key (or something) to move it from the stream and connect around. After that, we can drag and drop the tool to the right place.
Hi,
I was not aware until today that the REPLACE() function is actually case sensitive.
I opened a thread to try and find an interim workaround for my problem, but Ideally I think that a flag should be added as an optional parameter to this function similar to the Contains() function.
Ever since upgrade to version 10, when you highlight rows in a browse tool, it is almost impossible to see what is highlighted. Previously, I believe it was a shade of green which provided great contrast against the white background of the cells so you could easily see what you were highlighting. With version 10, it was moved to a very very light blue which almost looks white, so it is near impossible to tell which rows you're highlighting. I looked in the settings and do not see any way to change this.
Am myself and my colleagues alone in finding this very straining on the eyes? Would like to either be able to select what color you want in the settings, or move back to something easier to see.
Thanks.
Good morning!
This may be a very simple thing, but would it be possible to add a DateTimeQuarter() function? We have DateTime Second, Minute, Day, Month, and Year, and being able to have an easy formula for the quarter as well would be incredibly convenient.
Thanks,
Kat
User | Likes Count |
---|---|
7 | |
4 | |
4 | |
3 | |
3 |