Using Dataiku
- hello, I initally had a devlopped plugin available in a specific Dataiku instance. We had to change the dataiku instance . Is there an easy way to reload or import this previous plugin in this new ins…
Last answer by TurribeachYou need to import the plugin again.
Last answer by TurribeachYou need to import the plugin again.
- How can we effectively replicate this issue in order to conduct thorough testing during the version upgrade process? Specifically, what steps should we follow to ensure that the issue is accurately re…Last answer by Turribeach
It's not really your job to test bug fixes, that's Dataiku's job. Trying to replicate bugs that don't affect you doesn't sound like a reasonable strategy. Dataiku usually does 2 releases a month with dozens of bug fixes in every release the majority of which you are unlikely to be impacted by. In terms of functional testing I would only recommend testing fixes that you are actually impacted by in which case you should know well what are the conditions for the bug to be reproduced. The rest of your focus should be on testing your flows/scenearios/jobs run well in the new version.
Last answer by TurribeachIt's not really your job to test bug fixes, that's Dataiku's job. Trying to replicate bugs that don't affect you doesn't sound like a reasonable strategy. Dataiku usually does 2 releases a month with dozens of bug fixes in every release the majority of which you are unlikely to be impacted by. In terms of functional testing I would only recommend testing fixes that you are actually impacted by in which case you should know well what are the conditions for the bug to be reproduced. The rest of your focus should be on testing your flows/scenearios/jobs run well in the new version.
- How to replicate: Using windows10, download the latest Dataiku DSS on-premise version (13.2.3). Create a New project, upload any dataset with a "target" column having binary value. Click the dataset -…Last answer by
Last answer by LucOBrienI'm on 13.1.2 on premise and imagine that it is a Windows based env.
This is the ML basics project I started to try to recreate the issue:
I'll try it with the Churn example you gave and reply here again. It very well may be a bug between versions.
Edit:
I used the same sample dataset/project and even named my outputs the exact same and it ran successfully:
I'm wondering if someone from @Dataiku can look into this? @Alexandru (you were in a recent thread I looked at)?
- Hi I'm new to Dataiku. I'm using Dataiku 13.2.2. Currently I'm going through Core Designer Certificate from Dataiku academy. I have generated 3 calculated field using prepared recipes. But the calcula…Solution bySolution by Turribeach
After step 9 (Wrote the formula for calculated field) you need to click on Apply on the top right. Then once you click on Run you need wait for the job to succeed. After that you can explore the output dataset and you should see your calculated columns.
- Hello Community, i have a Streamlit app on Code studio which reads data from the Datasets. i am trying to write back comments from Streamlt to Dataiku dataset. When i try to use the below code df_writ…Last answer by
Last answer by JBaruathanks a lot. i just uploaded a blank csv and then used the resultant Dataset. it works now.
- Last answer byLast answer by Puchen
Hi, I'm having exactly the same issue:
https://community.dataiku.com/discussion/44483/when-evaluating-a-deployed-random-forest-model/ - For updating project tags I've tried the below mentioned options. awb_eng_project_metadata = awb_eng_project.get_metadata() awb_eng_project_metadata['tags'] = [] awb_eng_project.set_metadata(awb_eng_p…Last answer by
- Solution bySolution by Turribeach
It looks like you are asking about some Alteryx functionality. If that’s the case you need to specify exactly what you want to do in Dataiku as different Alteryx features will translate to different Dataiku features.
- In Alteryx I was able to build a zero records check that made it possible for me to have a "no data" entry for a particular path in my workflow. This is used in reporting to show that there were no it…Last answer byLast answer by jfurda_HMK77
I think that I figured it all out, but it seems like so many extra steps and . Here's how I did it.
I have numerical conditions (0-9) that get grouped into 4 categories.
- I joined an editable dataset that contains 10 rows (integers 0-9) to my data using a Right Join.
- This created a row for each of the conditions even if there wasn't a corresponding condition value in the main dataset.
- I used a prepare recipe formula to check for undefined values in a field (Full Name) and if true, populated them with "No Data".
- This, when the categories are combined, could result in one or more "no data" fields that need to be removed before the output step.
- I used a window recipe to remove duplicates before splitting the data into the categories.
- After the split into groups, I did a sort recipe to compute the row count.
- I then used another prepare recipe formula to filter out rows where Full_Name == "No Data" && Row_Count != 1
- I joined an editable dataset that contains 10 rows (integers 0-9) to my data using a Right Join.
