This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
At the moment, we are not able to use input data field names and its values in Output tool, mainly in the Pre-SQL and Post-SQL statement. I see some discussions on this in the community and in many scenarios we require that. It will be great if we have this option.
I was just thinking... they might not need to fully build out a python ide, but could still reach the same objective.
You should be able to keep a python file on its own and call it in r. By doing this, you might be able to have the json/xml handling of python with the visual/stats power of R while it being nicely bundled in your workflow. This uses base functions in r and does a good job turning a pandas dataset to an r dataframe you can move along your workflow.
You could always just use this same idea to write a file somewhere and once it's written, your workflow will continue. If you do, the code is literally 1 line in r... Anyway, let me know your thoughts!
As a method of deploying preprocessing and ML models it would be awesome to be able to convert a workflow to java...
models are needed to be deployed into Complex event processes or decision systems. Even for SAS there is a need to implement the datastep algorithms and procs to run in JVM.
It is possible to convert a workflow into a PMML file and then use JMML package to convert that to Java. Yet the full workflow with all preprocessing alternatives and a series of ML methods may not be captured fully.
I have several .yxdb files that I’ve been appending to daily from a SQL Server table in order to extend the length of time that data is retained.
They’re massive tables, but I may only need one or two rows.
I had hoped to decrease the time it takes to get data from them by running a query on them (or a dynamic query/input) as opposed to using a filter or joining on an existing data set which would have equal values that would produce the same result as a filter.
Essentially, the input of .yxdb would have the option of inputting the full table or a SQL query just like a data connection.
It would be great if I could save a tool and its configuration for future use. I'm sure this could be done now with a macro, and that's possibly how I'll do it.
The scenario is that we have numerous data inputs that require some finessing to import. Because the field headers are not in the first row, they aren't recognized accurately as comma delimited. But some of our fields also contain commas within the field, so the solution is to add a multi-field formula to replace things like ", Inc." with " Inc." and ", Ltd." with " Ltd."
But on occasion I find new text strings that need to be added to the formula, like ", Inc" (without the period) for instance.
I use this replacement technique with several workflows, so it would be great to change it just one time and have the other instances updated automatically.
Maybe in addition to our Favorites toolbar we could have a Custom or User toolbar where we could save tools along with their configuration.
I have a very large geospatial point dataset (~950GB) . When I do a spatial match on this dataset to a small polygon, the entire large geospatial point dataset has to be read into the tool so that the geospatial query can be performed. I suspect that the geospatial query could be significantly speed up of the geospatial data could be indexed (referenced) to a grid (or multiple grids) so that the geoquery could identify the general area of overlap, then extract the data for just that area before performing the precise geoquery. I believe Oracle used (uses) this method of storing and referencing geospatial data.
As I understand SFTP support is planned to be included in the next release (10.5). Is there plans to support PKI based authentication also?
This would be handy as lots of companies are moving files around with 3rd parties and sometimes internally also and to automate these processes would be very helpful. Also, some company policies would prevent using only Username/Password for authentication.
The Field Summary tool is a very useful addition for quickly creating data dictionaries and analysing data sets. However it ignores Boolean data types and seems to raise a strange Conversion Error about 'DATETIMEDIFF1: "" is not a valid DateTime' - with no indication it doesn't like Boolean field types. (Note I'm guessing this error is about the Boolean data types as there's no other indication of an issue and actual DateTime fields are making it through the tool problem free.)
Using the Field Summary tool will actually give the wrong message about the contents of files with many fields as it just ignores those of a data type it doesn't like.
The only way to get a view on all fields in the table is using the Field Info tool, which is also very useful, however it should be unnecessary to 'left join' (in the SQL sense) between Field Info and Field Summary to get a reliable overview of the file being analysed.
Therefore can the Field Summary tool be altered to at least acknowledge the existence of all data types in the file?
I found what I think is a bug. Usually the bug maker is me, but on this occasion I really think that it could be Alteryx (version 10.1.6.60263). Maybe we could add a category for posts as: Is this a bug? Currently, the idea labels allow for a "BUG". But is bug reporting really part of New Ideas?
I'm going to report my findings to email@example.com.
For those interested in what I'm observing:
Try creating a INTERFACE using an ERROR MESSAGE tool. Once you've got a formula and an error message, check the ANNOTATION. Do you see one on the canvas and do you see it in the configuration? Try putting a brief annotation into the Annotation box. I believe that the Annotation should appear in the annotation box as it does with other tools. Check the canvas and see what happens. Here's what mine looks like: