Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • 10:24
    bloussou assigned #423
  • 09:06
    HuifangYeo assigned #300
  • 07:42
    review-notebook-app[bot] commented #304
  • Dec 03 12:04
    mtadams007 commented #432
  • Dec 03 11:35
    xpilasav commented #432
  • Dec 03 11:33
    xpilasav closed #432
  • Dec 03 11:33
    xpilasav commented #432
  • Dec 03 11:28
    mtadams007 commented #432
  • Dec 03 10:12
    jnd77 commented #386
  • Dec 03 09:50
    tibdex commented #431
  • Dec 03 09:48
    xpilasav closed #431
  • Dec 03 09:48
    xpilasav commented #431
  • Dec 03 09:47
    xpilasav closed #430
  • Dec 03 09:47
    xpilasav commented #430
  • Dec 03 09:25
    tibdex commented #386
  • Dec 03 09:11
    tibdex assigned #430
  • Dec 03 09:11
    tibdex unlabeled #430
  • Dec 03 09:11
    tibdex unlabeled #430
  • Dec 03 09:10
    tibdex commented #430
  • Dec 03 09:07
    tibdex edited #430
vrotival
@vrotival

Hello is there any way to increase the default timeout when doing operations on the Java server ? I keep getting a lot of these
raise AtotiJavaException(
atoti.exceptions.AtotiJavaException: Write timeout

when trying to do things like
del measures[m]

usually a restart makes it OK so it looks random and there may be some settings at the session level to account for this

Thanks

6 replies
Yunus Talip Erol
@yerol89
Hi! I tried to read a csv file but atoti throw me this error:
AtotiJavaException: Failed to parse the row 0 while inferring types of the CSV. The row [CHN, China, "1, 445, 426, 055", "1, 439, 323, 776", "9, 706, 961 sq_km", 149/sq_km, 0.34%, 18.34%, 1] has 17 fields but the header [iso_code, country, 2021_last_updated, 2020_population, area, density_sq_km, growthrate, world%, rank] has 9 fields.
I think this error caused by the commas between the quotation marks. read_csv() method considers all the commas inside "1, 445, 426, 055" column as a separate column and this falsified Atoti.
I also tried the same operation and same csv on pandas but it work there. So the problem is not with the csv file.
But I know don't how to figure out this situation. Wanted to ask for a help.
Hope you will figure out this problem soon.
3 replies
Yunus Talip Erol
@yerol89
Hi! I have column having binary numerical data like 0 and 1. But as I know(correct me if I am wrong please), I can not use this column as a Hierarchy due its type. Is there a way to convert this atoti.Column object into categorical column and use it as a new atoti Hierarchy?
1 reply
Brieuc Loussouarn
@bloussou
Hello, atoti 0.6.2 has just been released. It brings some bugfixes, new features and small non-breaking changes.
mansourkefrawi
@mansourkefrawi
Hello, I am trying solve a Division by 0 issue when creating a new measure, I tried the python if condition syntax its giving an error. How can I get the expected result ?
7 replies
Yunus Talip Erol
@yerol89

Hello Everyone, as usual I have several questions and want to ask all of them at this single post. Hope you have time to answer them.
QUESTION-1 : We have two features to display on a line graph. One of them has value between the range of 0-100 while other is between 10K-20K. Due to the huge difference between these two features , the feature which has lesser values can not be displayed properly on the graph. I mean the bars which are indicating the lesser value displayed very small on the chart and even can not be noticeable compared to the feature which has greater values. Is there a way to display both of these features on the same chart in a proper way? I want both features to be noticeable on the chart regardless of their scale.

QUESTION-2: Assume we have a bar chart. As you know, we can display the features when we hover on a single bar on a pop-up. I wonder whether is it possible to display some extra features on the pop-up which are not displayed on the chart when I hover on a bar. I mean I want extra features to displayed on pop-up but not on the chart for the sake of simplicity of the graph.

QUESTION-3: As a last question, I have two tables and set a relation between these two tables using join() method and mapped them to each other on mutual features(assume Feature-A from Table-1 and Feature-B from Table-2).
When I try to draw a chart using this two table, I can display all the Hierarchies from both table at the hierarchy tab, but can not do the same thing for the measures.
I want to tell this scenario with an example also. I can display all the Hierarchies from Table-1 and Table-2 from the tabs of Tabel-1 and Table-2. But can only display measures belongs to Table-1 from Table-1 tab ; can only display measures belongs to Table-2 from Table-2 tab. I mean I can display the measures which are only belongs to a specific table.
Is the only way for displaying all the measures merging this two table as a single table? Or there is a way to do this without merging them?

Sorry for this long entry but this are the issues that we couldn't solve today. If there is a way for learning this type of specific questions except the atoti documentation please let me know.

8 replies
mansourkefrawi
@mansourkefrawi
Hello Everyone, I want to add slicers to the web application in order to choose for example specific companies to read their data, i can choose one or more company at the same time, how can this be done? Thanks !!!
9 replies
Alexis Blervaque
@alexis-bvq

Hello !
Is there a way to update a table by loading a dictionary?
For example, I have a table like this:

tradeId (key)
portfolio
businessUnit
subBusinessUnit

Then I would be able to do this:

table.load([
{
"tradeId": "2",
"portfolio": "NewPortfolioName"
},
])

Which would allow to avoid having to reload all fields of the table but just update the one you need for keys

3 replies
Mano251290
@Mano251290
Hi - how to add labels to axis in atotti
3 replies
Mano251290
@Mano251290
Is there a way to remove legends from pie chart
1 reply
Turning out data tricks since 2006!
@thistleknot
'Cube' object has no attribute 'create_static_parameter_hierarchy'
trying to call
cube.create_static_parameter_hierarchy(
"Dates", ret["Date"].tolist(), index_measure="Date Index", store_name="Dates"
)
1 reply
Turning out data tricks since 2006!
@thistleknot
"Open the notebook in JupyterLab with the atoti extension enabled to build this widget." how do I accomplish this? I don't see a reference to a "jupyter labextension install my-extension my-other-extension" in atoti documentation and my jupyterlab notebook isn't actually visualizing anything. I have graphviz installed and in my path https://github.com/thistleknot/Python-Stock/blob/master/code/markowitz/Markowitz-slim.ipynb
4 replies
Alexis Blervaque
@alexis-bvq
Hello how can we know which version of the ui is under atoti ?
I need this information because I need to know if 0.6.2 has AUI 5.0.3 to edit the "smartFiltering.ignoredHierarchies" settings :)
2 replies
ab1956
@ab1956

I am trying to understand the measures Atoti automatically creates. I looked at the training video of Ms. Hetal Kapadia (intro to Atoti) but could not understand firstly how to switch this auto function off, secondly what are data types I can have , thirdly I need a hierarchy of many items classified as measures to be hierarchies.
Any help in this matter would be appreciated.

My data is single CSV files of cash flows by Date and Time Bucket by common attributes.

3 replies
piyuw
@piyuw

Hi HuiFang,

This is the questions I had about the where statements:

First Question

In the below query vs the original query, just to understand the syntax structure:

In the Queries below, you have used where outside of the tt.agg.sum instead of inside, is it the difference between summing and taking the aggregate output Vs Summing and replacing the same row element ? for eg: if there are ten rows, and when you groupby the date, it becomes 5 rows of Dates, using this methodology sums up the value at the 5 row summary level. where as using the below methodology with the where inside of the tt.agg.sum, the output still has 10 rows, but the summary values of the 5 rows have replaced the unique 10 row values ? essentially every row of the 10 rows, has the summary value of the group repeated twice ?

m["Past 3 months"] = tt.where( 
    l["Dates"] != None,
    tt.agg.sum(
        m["Value.SUM"], scope=tt.scope.cumulative(l["Dates"], window=("-3M", None))
    ),
)

VS

m, l = SaleCube.measures, SaleCube.levels
m["NewInvoicedTax"] = tt.agg.sum(
   where( <-- Referencing the Question Above
           m["InvoicedTax.SUM"] == 0,   # if condition
           SaleCube.measures["InvoicedAmount.SUM"]  * 8%,  # then
           SaleCube.measures["InvoicedTax.SUM"]    # else -- you can embed another "where" condition here
   ),
    scope=tt.scope.origin(l["BusinessUnit"], l["ReportDate"], l["CustomerRefListID"], l["TxnID"]),
)

Second Question

m, l = SaleCube.measures, SaleCube.levels
m["NewInvoicedTax"] = tt.agg.sum(
   where(
           m["InvoicedTax.SUM"] == 0,   # if condition
           SaleCube.measures["InvoicedAmount.SUM"]  * 8%,  # then
           SaleCube.measures["InvoicedTax.SUM"]    # else -- you can embed another "where" condition here <-- You said i can embed another where statement here, but does that mean the new where statement goes inside the                                                                                                                                                                                                        old where statement ? like the one below ? 
   ),
    scope=tt.scope.origin(l["BusinessUnit"], l["ReportDate"], l["CustomerRefListID"], l["TxnID"]),
)

I want to know which option is correct one, and moreover whats the difference of having where vs tt.where [ as per example number 3] ? is there a difference or is there no difference ? and whats the differene between option 1 and 2 (if there is a difference at all) ?

My version

1st Option

m, l = SaleCube.measures, SaleCube.levels
m["NewInvoicedTax"] = tt.agg.sum(
   where( m["InvoicedTax.SUM"] == 0,  SaleCube.measures["InvoicedAmount.SUM"]  * 8%  ,
               where( m["InvoicedTax.SUM"] > 2, SaleCube.measures["InvoicedAmount.SUM"]  * 15%   ,
                        where( m["InvoicedTax.SUM"] > 10, SaleCube.measures["InvoicedAmount.SUM"]  * 25%   ,
                             SaleCube.measures["InvoicedTax.SUM"]  # Else statement
                         )
                     )
                ),

    scope=tt.scope.origin(l["BusinessUnit"], l["ReportDate"], l["CustomerRefListID"], l["TxnID"]),
)

2nd Option

m, l = SaleCube.measures, SaleCube.levels
m["NewInvoicedTax"] = tt.agg.sum(
   where( m["InvoicedTax.SUM"] == 0,  SaleCube.measures["InvoicedAmount.SUM"]  * 8%  , SaleCube.measures["InvoicedTax.SUM"] ),
    where( m["InvoicedTax.SUM"] > 2, SaleCube.measures["InvoicedAmount.SUM"]  * 15%   ,SaleCube.measures["InvoicedTax.SUM"]  ),
    where( m["InvoicedTax.SUM"] > 10, SaleCube.measures["InvoicedAmount.SUM"]  * 25%   , SaleCube.measures["InvoicedTax.SUM"] ),


    scope=tt.scope.origin(l["BusinessUnit"], l["ReportDate"], l["CustomerRefListID"], l["TxnID"]),
)

given that the discussion was very helpful, I hope the above question can also help everyone else in the community.

Thanks
piyuw

2 replies
ab1956
@ab1956
i am trying to run the real time risk note book I get the following error
RuntimeError: Cannot load plugin atoti-kafka v0.6.2 because it does not have the same version as atoti (v0.6.1).
2 replies
but I already have atoti 0.6.2 installed Welcome to atoti 0.6.2!
Chenna Reddy
@chenna-reddy
Hello, I'm trying to start with atoti/project-template. When I launch the default URL http://localhost:<port>/ it throws 404. Also As mentioned in a few examples session doesn't have url property (it does have port though). Wondering what I'm missing.
1 reply
Harold LAKEHAL-AYAT
@Harold7665
Hello
How to access to the documentation of a nightly?
2 replies
Yeo Hui Fang
@HuifangYeo

Hi,

is there a way to get current date from atoti? I would like to compute a moving date, e.g. age from DOB, or aging transaction.

2 replies
Yeo Hui Fang
@HuifangYeo
Hi, is it possible to export the underlying data programmatically from atoti?
7 replies
Turning out data tricks since 2006!
@thistleknot
image.png
how can I use cube.query against randomsymbols
2 replies
Harold LAKEHAL-AYAT
@Harold7665
Hello
What is the ActivePivot version used in atoti 0.6.2?
7 replies
Chenna Reddy
@chenna-reddy

Lazy loading Data from Database

I have data from multiple DB tables. Would like to expose each table as a Cube. So end-users can create dashboards using any of these Cubes/Tables.
I'm using session.read_sql at the start of the program to create these cubes.
My question is, "How can I instruct atoti to make these DB fetches lazily?". Another way of saying "Don't load everything at the start of the program, as these tables can be huge it will impact startup time."
Also, I'm seeing lot of following warnings:

DatastoreSelectionSessionHolder : Trying to register an activepivot schema on a datastore that already has data, this can lead to performance issues. Please start and initialize your manager before feeding the datastore with data. Only ignore this warning if you are aware of the consequences.

Wondering if creating all cubes at the start of the program is the right thing to do. Or is there a better way of doing this?
Thanks in advance for ur response!

4 replies
Chenna Reddy
@chenna-reddy

Context Path

Is it possible to host atoti server at different context path? We tried with spring property server.servlet.context-path, but I think its not respected by all modules/server components (e.g. the one responsible for serving static content). Please let me know how I can acheive this.
4 replies
Sangram Deshpande
@SchrodingersSangru
Hello all, Myself sangram. I want to ask that can I use postgresql database with atoti? if yes, then how can I use? Please answer my query.
1 reply
mecoccaro
@mecoccaro
Hi! I'm using the project template with docker, how can I access the localhost in that case?
1 reply
Sangram Deshpande
@SchrodingersSangru
I want to create a 2-3 page application using atoti and also want to add nav bar, how can I do it?
1 reply
Thibault Derousseaux
@tibdex
Hello, atoti 0.6.3 has been released. It brings some new features, bugfixes, and small non-breaking changes.
Thibault Derousseaux
@tibdex
Hi all, if you've ever used our Docker image with atoti and its JupyterLab extension can you please share your use-case in this thread?
4 replies
khaledShabrawy
@khaledShabrawy
Hi
i need your any one support me for videos Discribe the atoti tools to build BI App
1 reply
Marco TĂșlio Carvalho
@MTulio2000
How can I make charts dynamically, just changing the filter on a line chart? I'm using Jupyter notebook to generate charts. I have a list of filter and using a "for" loop I didn't get what I needed. Is there a way to do this?
4 replies
vrotival
@vrotival
Hello When connecting to a session with its URL like xxxx.com:5180, do we have a way to edit the page name ? I we have 5 servers we currently have 5 'atoti' tabs so it's not easy to know what is what ?
Thanks and keep on the amazing work
3 replies
piyuw
@piyuw

Hi Thib, you mentioned the following :

Alternatively, you could set the logging destination to sys.stdout so that the Java logs are interleaved with the Python ones which would help understanding the chronology of your issue.

however, when i did the following :

import sys
config = {"logging": {"destination": sys.stdout}}

where does the new atoti together with python log file get written to on a linux machine? Because when i print out :

print(config)
{'logging': {'destination': <ipykernel.iostream.OutStream object at 0x7f096f70f150>}}

The above is the reply i get.

I was hoping for the location of the new log file in the form of home\username.ipython....

please let me know your thoughts.

thanks
piyuw

13 replies
Chenna Reddy
@chenna-reddy

DB Connection leaks

Hello, we use atoti_sql#load_sql to load data from Oracle Database. But I observed it never releases connections (I see connection count reaching >50 sometimes). As atoti never refetches data on its own (i.e. refresh), it doesn't make sense to keeping connections open. I believe this is a bug. Where can I raise this to get fixed?
2 replies
monapeter
@monapeter
image.png
AtotiJavaException: The partially joined field b from store df3_t is also used as a level in a hierarchy b@df1_t
I'm trying to join two columns d1 & d2 but I get the above error message. Is it possible to create such a circular join?
2 replies
Dhruthi A
@dhruthi:matrix.org
[m]
Atoti JavaException: How to resolve?AtotiJavaException: Error while running command to get file permissions : ExitCodeException exitCode=-1073741515:
1 reply
monapeter
@monapeter
Hello, how can I create a measure calculating the difference between base and scenario in a source simulation? In the examples on github (price simuIation) I only found it done in the mdx. Is this still possible with atoti? I did not manage in the gui . Can you still display and edit the mdx in the browser? Thanks a lot!
2 replies
Harold LAKEHAL-AYAT
@Harold7665
Hello,
Can I do a parameter simulation on a source simulation? I can not find the level "branch". I am using atoti 0.6.0
6 replies
Chenna Reddy
@chenna-reddy

Custom Level Comparator

Hello,
How can I implement Custom Comparator? first_members isn't a good fit for our use case. I see there are just 3 implementations provided (ASCENDING, DESCENDING, and FIRST_MEMBERS).
Also, it would be nice to have some comparator like 'AS_IS', so the provided values are already in some sort of order, don't bother sorting them.
2 replies
Mano251290
@Mano251290
Hi - How to delete a created measure from the cube
2 replies
Yeo Hui Fang
@HuifangYeo

I have the comparators on the trans_date in descending order. How can I do the cumulative aggregation?
Currently it's aggregating in the other order:

image.png

2 replies
Lorena
@glorena06:matrix.org
[m]
Hello, I just started using atoti and I got problem with multidimensional concept. When I enter this : cube.query(m["Quantity.SUM"]) I got this: data type 'string' not understood. Can someone please help me fix this problem. I searched on google but not convenient answer.
1 reply
Harold LAKEHAL-AYAT
@Harold7665
Hello,
With the atoti S3 connector, can I load .gz files to the atoti table?
2 replies