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 enriching your date data, week information is pretty critical, I suggest adding %w to the format stings as well..
Several month based expressions can also be made weekly, data time first of week #31, last of week #45 etc....
Also came across a lot of questions at our community on how to turn week based data...
Writing the formula is obvious (Ceil(DateTimeDiff([Today],[Date],'days')/7))
but having a DateTimeDiff([Today],[Date],'weeks') is preferable don't you think?
Cheers
I'd like to see the size of the data processing through each tool, in say GB. This would be helpful in understanding the volume of data being written to connected systems and would also be useful when cleansing data to determine if a meaningful impact was made to the data cleanliness.
When giving a user a dropdown list, it is difficult to give them an option of the data in a specific column. So if I have a "State" column, I would like to give the user a dropdown of all of the states currently represented in the table. So if there was "Nebraska, Nebraska, Iowa, Kansas, South Dakota", the dropdown list would give them an option to select Nebraska, Iowa, Kansas, and South Dakota. This could be useful on both List boxes and Dropdown lists. Currently there is a workaround that lets you do it, but it is not a great solution because if there is a space (Like in South Dakota), it puts an underscore in it so parsing is required.
Please enhance the Join Multiple tool to include an option of:
"left outer join all on input 1".
Such a tool would be massively advantageous when data granularity is defined by one dataset (input 1) and many subsequent tables just need to be joined onto that input.
As-is, I have had to manually perform dozens of join + union sequences which seems both untidy and inefficient.
Such a feature would likely be more useful than existing options of full outer joins / inner joins which currently the tool allows.
I love the new (relatively) ConsumerView Decoder Tool! I used to do it the hard way, and it was fragile.
However, one thing is still missing: the Mosaic fields (MOSAIC HOUSEHOLD and MOSAIC ZIP4) - these are output from the tool as nulls. So, not only do you not get it decoded, you have to join back to the input to get the fields back as they were.
First, at least please pass them through as they were.
But preferably, decode them to the Mosaic Segment/Group names.
I realize (or couldn't find) the source for the Mosaic segment definitions is not currently in a Calgary database, but the tool is in the Calgary group.
The ability to Sort and Filter in the Results window is a huge time saver. Please allow the same functionality when viewing results in a new window.
I was able to add the following lines of R script to get the importance of the variables used in the cluster analysis. This will allow the user to see what variables are important in determining the clusters they have.
The script I added is below. It is pretty basic and could used spruced up by an Alteryx engineer as far as column naming, accounting for contingencies, and making it a reporting function. I think this would be a valuable feature for future versions of this tool.
library(FeatureImpCluster)#load library
FeatureImp_res <- FeatureImpCluster(clus.sol,as.data.table(the.matrix)) #Use FeatureImpCluster to take the cluster model (clus.sol) and data (the.matrix) to get variable imp.
FeatureImp_df <- as.data.frame(FeatureImp_res$featureImp) #turns features from a list to dataframe
FeatureImp_df_rn <- tibble::rownames_to_column(FeatureImp_df, "Variable") #Adds the variable name to the importance scores
write.Alteryx(FeatureImp_df_rn, 3) #outputs dataframe in output #3
The new enhancements to the Input tool, File, Excel file type being able to input a named range is fantastic!
One trick I use often when creating a template Excel file for user input is I give the form sheet a "Code Name" - sadly in Excel this cannot be set when creating the template using code or from Alteryx - it needs to be set manually in the <alt><F11> IDE, select the sheet in browser then set the code name in the properties window. Advantage: the user can change the visible label/name of the sheet to their liking but the code name does not change - therefore, an automated pipeline with a fixed sheet name for import (like a workflow picking up files dropped into a drop folder to ETL new results into a database), could reference the sheet code name and not be prone to workflow failure if the user changes the name of the sheet.
Please consider making the Count Record tool configurable so that users can receive a visual read of results in canvas. This would quickly assist in ensuring a user's ability to verify a workflow is functioning as expected or not. Currently there is no in canvas visual cue tied to the counter. If a user is expecting a certain count result (e.g. zero), the user has to click on the count records function to see if the result meets expectations or not. Users may spend a lot of time checking each stage of a workflow to ensure everything is flowing appropriately. A visual cue of results would reduce that time to check. Outside of runtime errors, there is currently no visual cue to indicate a possible problem or unexpected exception to a filter or other macro.
Two suggestions:
1) Allow users to color code count results in specific count results or ranges. (e.g. if a user is expecting a zero count, allow them to change the color of the counter to red for anything not resulting in zero and green if count is zero). Or allow them to set a color range depending on the count total (e.g. a user may need some visual tolerance indicator with a count under 100 as green, 101-200 as yellow, 201-300 as orange and 300+ as red).
2) Show the actual count in the Count Record macro icon.
Thank you!
Can we have a tool to optimize another tool's configuration based on an output target? For example optimize the fuzzy logic setup to find the optimal tool configuration that yields the best matching score for a given data set.
Would like to be able to connect to the Stibo STEP system/database as a Data Source. Some people have the Stibo server on-premise while others have it hosted in Amazon (AWS).
Not sure what else I could provide at this point for further details.
Hello Alteryx Gurus -
I've got some workflows that run daily, but there are times, depending on the breaks, wherein I don't get any data from one of my data sources. Which is actually fine, nobody did Job X today. But it makes Alteryx puke out and I get an error message emailed to me. Ultimately, I've got to hop into the rather voluminous log entries to determine if this was a data stream not initialized / was empty error, or something else that I actually need to care about.
That being said, in the coding realm, it is relatively simple to look for specific flavors of exceptions and then just eat them without notifying people. So, why not add something to the runtime / events panel for emailing at error time to allow for ignoring data stream not initialized errors? In this way, I could get notified when there is a real error I need to pay attention to, and not get notified when there is no new data, which isn't really that big a deal.
Thank you for attending my TED talk on enhanced error reporting and exception classification capabilities.
Hello
I have searched the community but haven't found any obvious solutions to this.
When using a cross tab I often find that there shouldn't be any aggregated values and if there is it means there is an issue with my data or workflow.
Therefore I think a useful feature would be an option for the cross tab tool to be able to return an error if it trying to aggregate any values.
I have a work around by using a summarize tool to count the non unique records and then a test to see if there are any duplicates but I think this could be a useful addition to the tool.
Thanks
Could you add a presentation mode to Alteryx Designer. When I demonstrate Alteryx Workflow to my colleagues, what Alteryx is doing is great, but it doesn't look great. I'd like to see more clarity and flair when running workflows. In presentation mode, all tools should be dimly grayed out immediately after run, and tools that are 100% complete should be bright and clear.
It would be nice to have the option to open a macro from the configuration window. I often use the find tool to locate macros and sometimes they are in collapsed tool containers or buried under other tools and it makes it difficult to right click on the tool itself. The find tool brings up the configuration window so if there was a button on that window it would make opening the macro more efficient.
The macro input tool asks for a template input that is either a text or file input. A perfect template can be hard to come by and so I wonder if the macro input could just ask how many columns are expected and the type and size that this column should be. I feel this would be more intuitive and easier to debug.
While scrolling through the analytics gallery, if I click on a tool on page 5 to read the details and then hit back key, it takes me back to page 1. It would be nice to just go back to my previous page.
Thanks!
In the next product version, can the parameter options for the topic modelling be changed to allow the output of both word relevance summary and interactive chart? It's a bit strange to run the tool twice to get this output.
Over the period of time, it is quite natural to forget why we have used certain filter in the filter tool or why we have used certain transformation etc....
I am looking for an option for every tool that I use in the canvas... .by right click and choose "maintain legacy" kind of option and it opens up a pop-up or appears from the far right side of the canvas, and date and time is already populated there and provides a place to maintain 'note' of each of the transformation why we have done that and when we have done that. So, it helps us to track what changes have been done to the tool or workflow over period of time.
In our situation, we have been making several changes to the workflows and implementing new business rules as and when it becomes necessary. After a year or so, when we revisit the workflow, forgetting what kind of decisions made us make those changes to the rules or filters or joins etc...
Thank You,
Suresh Saguturu.
I often encounter situations where I would love to be able to use the tool container name as a constant within my workflow. This would be a huge time savor for me as I use containers often to segment my work and those segments often have a meaningful purpose within my workflow. Having the ability to use the name for decision making or as a file prefix/suffix would be a great enhancement
User | Likes Count |
---|---|
7 | |
6 | |
6 | |
5 | |
5 |