Quantcast
Channel: Pentaho Community Forums
Viewing all 16689 articles
Browse latest View live
↧

Multi Value list displays id when value is selected

July 23, 2014, 1:48 pm
$
0
0
I created a canned report that is using a query for a multi value list parameter. When users select the value from the list in the user console, it displays the ID at the bottom of the list. It doesn't do that in the Report Designer though. The query is selecting an ID and FundName.

Values of the parameter are:
Value Type = Object
Display Type = Multi Value List
Query = the name of my query
Value = ID
Display Name = FundName

How can I fix this so that it doesn't at the ID to the bottom of the list when it's selected?

Thanks,
Wendy
↧
Search

Job executor issue

July 23, 2014, 3:46 pm
$
0
0
I have a main transformation in which I am calling a job using the job executor

I am returning only one column with a single row from the job back to the main transformation..

In the job executor result rows tab , i have given the field name with the data type of the field returning (which is a number field ).


I am facing 2 issues when I run

1) getting data type errors like "he 'Integer' data type passed from the jobs result rows does not correspond to the specified 'Number' data type. Make sure you are passing rows with the expected layout"

2) The target step i gave in the result row / execution results / result files tab is not persistant . the field name box gets empty every time and I have to manually delete the hop everytime and drag to make this work ..

Anyone faced similar kind of issues ? I am using pdi-ce-5.0.1-stable
↧
↧

Where can I download biserver manual installation?

July 23, 2014, 4:34 pm
$
0
0
Hello everyone,

I'm a newbie to Pentaho and I wonder where I can download a biserver manual ce 5.1 distribution .zip because when I go to the Pentaho Customer Support Portal, as described on the documentation, it is requiring me a login.

Additionally, I think the documentation is not right about creating a folder called .pentaho since you cannot start a folder with a dot on Windows. Maybe it works just with Linux.

Thanks in advance,
Ronald
↧

Edit Pentaho Mondrian Models Inline in your Browser

July 23, 2014, 5:00 pm
↧

Weka-Parallel step by step installation

July 23, 2014, 7:10 pm
$
0
0
Can anyone give me step by step installation on Weka-Parallel.. How do i install it. How do I set it up etc. I am a beginner 99% is over my head right now :( Thank you
↧
↧

How to run multiple pentaho application on same server.

July 17, 2014, 11:33 pm
$
0
0
How to run multiple pentaho application on same server with MySQL database.
↧

Kettle Java Web Start could't start

July 18, 2014, 12:11 am
$
0
0
I have download source code of kettle 4.4.0 stable from http://source.pentaho.org/svnkettleroot/archive/Kettle/tags/4.4.0-stable/.when I access to the browser, just after opening the Kettle splash screen, the app closes without errors. Does I using a wrong JNLP file? I found the JNLP file in source code has't modify since 2008.If it is a problem about JNLP file,how should I modify it? Thanks a lot.
↧

ThresholdSelector and CostSensitiveClassifier on updatable classifiers / stream

July 18, 2014, 1:09 am
$
0
0
Hi,

Do you know if ThresholdSelector and/or CostSensitiveClassifier are compatible with updatable classifiers learning on streams ? My goal is to use them with weka.classifiers.meta.MOA to focus learning on a specific class and minimize FN.

Thanks a lot!
↧
Search

Suggestion in time series analysis

July 18, 2014, 1:19 am
$
0
0
Hi all,
I need some help in Weka time series forecasting. I have some data divided in file each of them with 2 attributes and some of them with few instances (about 20) and some of about 60 with monthly period.
I want run the time forecasting series for 12 months looking for the best algorithm, but I need some help to how validate the models. My idea is:
1) File with few instaces (i.e n instances): I run weka in training files with different algorithms and compare the looking RMSE (MAPE MAE )and choose according these? Or there is another way?
2) File with more instances, May I have to work like above, or run with hold-out (i.e 80%) and check how is the error in prediction instances. Or I have to remove m instances run the modules and check which work better with n-m data?
Thanks
Antonio
↧
↧

Wrong result - BTable with Subtotals and ParallelPeriod

July 18, 2014, 2:14 am
$
0
0
Hello All,

Following MDX is generated by BTable Analyzer (we are using Pivot and Subtotals feature of the BTable) which returns wrong result, i.e, product repeat for each category instead of returning only the products in a category.

Code:
with
 set [Category_Category_Set] as '{[Category].[Category].Members}'
 set [Product_Product_Set] as '{[Product].[Product].Members}'
 set [Measures_Set] as '{[Measures].[Amount], [Measures].[DifferenceLastYear]}'
 set [Time_Year_Set] as 'Filter({[Time].[Year].Members}, ([Time].[Year].CurrentMember.Name = "2013" OR [Time].[Year].CurrentMember.Name =  "2014"))'
 set [Time_Month_Set] as 'Filter(Filter({[Time].[Month].Members}, ([Time].[Month].CurrentMember.Name = "6")), ((Exists(Ancestor([Time].CurrentMember, [Time].[Year]), [Time_Year_Set]).Count > 0))))'
 member [Category].[BT_TOTAL] as 'Aggregate([Category_Category_Set])'
 member [Product].[BT_TOTAL] as 'Aggregate([Product_Product_Set])'

select NON EMPTY Crossjoin(Descendants([Time_Month_Set], [Time].[Month], SELF), [Measures_Set]) on COLUMNS,
NON EMPTY
Union(
  Crossjoin([Category_Category_Set], Union([Product_Product_Set], [Product].[BT_TOTAL])),
  Crossjoin([Category].[BT_TOTAL], [Product].[BT_TOTAL])) on ROWS
from [Sales]
Formula for calculated measure 'DifferenceLastYear' is:

Code:
([Measures].[Amount]/(ParallelPeriod([Time].[Year],1,[Time].CurrentMember),[Measures].[Amount])) -1
If we remove the DifferenceLastYear measure from the MDX then result is fine (no product repetition) and if we use some other formula for the calculated measure (for e.g. just - [Measures].[Amount] * 10 )then the result is also fine with the above MDX. If we do not use Subtotal feature then the result is fine and the MDX generated is:

Code:
with
 set [Category_Category_Set] as '{[Category].[Category].Members}'
 set [Product_Product_Set] as '{[Product].[Product].Members}'
 set [Measures_Set] as '{[Measures].[Amount], [Measures].[DifferenceLastYear]}'
 set  [Time_Year_Set] as 'Filter({[Time].[Year].Members},  ([Time].[Year].CurrentMember.Name = "2013" OR  [Time].[Year].CurrentMember.Name =  "2014"))'
 set [Time_Month_Set]  as 'Filter(Filter({[Time].[Month].Members},  ([Time].[Month].CurrentMember.Name = "6")),  ((Exists(Ancestor([Time].CurrentMember, [Time].[Year]),  [Time_Year_Set]).Count > 0))))'
 member [Category].[BT_TOTAL] as 'Aggregate([Category_Category_Set])'
 member [Product].[BT_TOTAL] as 'Aggregate([Product_Product_Set])'

select NON EMPTY Crossjoin(Descendants([Time_Month_Set], [Time].[Month], SELF), [Measures_Set]) on COLUMNS,
NON EMPTY
Union(
 CrossJoin([Category_Category_Set],  [Product_Product_Set]),
 Crossjoin([Category].[BT_TOTAL], [Product].[BT_TOTAL])
) on ROWS
from [Sales]
We could not pinpoint where the problem is.

Can someone please shed some light on the issue.

Thanks,
↧

RSS Feed to file

July 18, 2014, 4:06 am
$
0
0
Hi Guys,

We have an RSS feed that needs to be saved into a text file..(simple and is working) Feed only consists of two fields."title" = date and "description"
I sit with a problem where the feed gets constantly updated, but I don't want any duplicate data in my text file and I obviously don't want to miss any lines.

Any ideas how to achieve that?
Please see attached transformation file.

Thanks

RSS_Sample.ktr
Attached Files
↧

Set MySQL session variables during transformation

July 18, 2014, 4:38 am
$
0
0
Hey guys,

is it possible to set a system variable of the MySQL server dynamically during a transformation in kettle? I tried the following:

Step 'Execute SQL Statements': SET SESSION group_concat_max_len = ?;

The question mark is replaced by an argument of the step before. But even if I place an absolute value there instead of the question mark, it's still not working. The following steps, e.g. a table input step, will ignore the modified session variables so the default system values are used.

Setting the system variable globally on the whole connection (and not only for the session) is no option in my case. Using PDI 5.0.7, MySQL 5.1.73.

Many thanks,
tba
↧

Filtered Classifier: Naive Bayes Updateable + StringToWord Vector

July 18, 2014, 5:48 am
$
0
0
Hi, I need to execute StringToWord Vector with Naive Bayes Updateable.
First, I want to save the .model with the FilteredClassifier trained. Then open it, classify and update it with a new instance.
Could you help me with the java code to do this?
Thank you so much.
Andrea
↧
↧

JavaScript step error handling

July 18, 2014, 6:17 am
$
0
0
Hello,
I'm having some issues on error handling.
I'm trying to do a fine-grained validation on data input and I've written some javascript code to do it.

Code:
// validation

function isBlank(p) {
        if ( typeof p == "object"){
                if ( p )
                        return false;
                else
                        return true;
        }
          return !(!!p && !(p.trim().length === 0));
}


try{
        if ( isBlank(NUMERO_PRATICA))
                throw {
            message: "NUMERO_PRATICA should not be empty",
            nr_errors: 1,
            field: "value",
            errcode: "ERR:002"
        }
       


        trans_Status = CONTINUE_TRANSFORMATION;
       
}
catch(e){
        Alert("yo");
        _step_.putError(getInputRowMeta(), row, e.nr_errors, e.message, e.field, e.errcode);
    trans_Status = SKIP_TRANSFORMATION;
}
Now, this is my Transformation:

error.jpg

To sum it up: I select some rows from a table (and they're coming up just fine), do the validation you can see in the code up and if it's wrong throw an error.

Ideally, I'd like the Transformation work like this:
Every record you find faulty, you put it in the log and update the selected record (so that another service will manage it) and keep working the other (hopefully/ correct records.

Is it possible?

I've read everything I could online on pentaho error handling but I'm facing a wall:
The errors goes correctly on the log BUT the step "convert Json" also goes in error since it's trying to convert null values.
Also, how can I put the error message I'm throwing up in the javascript code?

Thanks in advance,
Massimo.
Attached Images
↧

Pentaho CDE book

July 18, 2014, 6:46 am
$
0
0
Hi,

Could you please provide the soft copy of Pentaho cde book that will help us to creating dashboards.else let us know the book name

Thanks
Bharath
↧
Search

Help on Crosstab group Title Header visible property

July 23, 2014, 8:04 pm
$
0
0
I have a crosstab table, I want to show/hide the columns depending on the query.
I found out that crossstab-row-group > Title header > Label > visible property does not respond to Expresssions like =FALSE().
However, the crossstab-row-group > Group header > String field > visible property works ok. Is this a bug?
↧

SparkL Server Side Parameter

July 23, 2014, 9:12 pm
$
0
0
Hello experts

Im trying out the SparkL plugin architect, so far its great, i just cant manage to get the server side parameters like cpk.plugin.id,cpk.solution.system.dir,cpk.session.username for a transformation endpoint, custom parameters works just fine,im just wondering if this is cause im using pentaho 4.8 build 18 version, ive seen in a tutorial this works for pentaho 5

Thanks in advance
↧
↧

Horribly slow performance of Dimension Lookup/Update Step

July 23, 2014, 9:30 pm
$
0
0
I have a Dimension Lookup/Update Step that keeps performing very slow (100 records per sec) and eventually crashes, out of memory.

For testing purposes I removed 10 of the 11 fields I had in the Fields tab, and just left one Integer field.

Performance is improved to 1,580 records per sec... an over 15x improvement, and no Out Of Memory crash.

Is this a known bug? Or something I may be doing wrong?
↧

Wants read postgres OID images in pentaho data integration

July 24, 2014, 12:35 am
$
0
0
Hello All,

I have case in which i want to access postgres OID images files. I want to migrate that images in other database BLOB. Any one have solution help me.


Thanks & Regards
Nitesh Lohar
↧

Recursive Parent / Children problem

July 24, 2014, 2:18 am
$
0
0
Hi eveyone !

I am searching for a while now to create a simple recursive traitment from a CSV file that I have.

Here is my setup:

1) I have a CSV file that contains a list of sections, each having multiple children. The relation is set thanks to a field "ParentLevel" that indicates, for every children, which parent it has.
The child has its own field "CurrentLevel" that can appear in another entry as a parent.

That way, it creates a tree with unique parents having multiple children.

Its template is the following :
ParentLevel ; CurrentLevel ; Name
What I am trying to achieve is the following :

I am trying to recreate the entire tree from the CSV file, with the following transform (can't upload picture so i will explain)

I read the csv file, use a "Closure Generator" to get the distance between every parent and child. From there, I sort on CurrentLevel and filter the rows on ParentLevel that are equals to "ROOT".
From here, I can clone the rows by the distance the closure generator gave me (minus 1 with a modified java value), it gives me the following ouput :

-----------------------------------
ParentLevel = ROOT
CurrentLevel = 997715
Distance = 3
CloneNumber = 0
-----------------------------------
ParentLevel = ROOT
CurrentLevel = 997715
Distance = 3
CloneNumber = 1
-----------------------------------
ParentLevel = ROOT
CurrentLevel = 997715
Distance = 3
CloneNumber = 2
-----------------------------------

What I need to do, is to get the levels between ROOT and 997715, like this :
-----------------------------------
CurrentLevel = 997715
ParentLevel = 2103361
-----------------------------------
CurrentLevel = 2103361
ParentLevel = 123
-----------------------------------
CurrentLevel = 123
ParentLevel = ROOT
-----------------------------------



I tried to filter on the "distance" value and decrement it, get the parent from the child and loop, but pentaho will not allow it. (Loops are not allowed!)

I am using Pentaho Data Integration 5.1.0.

If anyone has a clue of how to do it, I'm all ears! In the mean time, I'll try to work something out and let you know if I find anything.

Thank you for your time !

Cheers,
François.
↧
Viewing all 16689 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>