[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15334100#comment-15334100
]
yuhao yang commented on SPARK-15930:
[~GayathriMurali] Adding the information value would be an
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15332588#comment-15332588
]
Gayathri Murali commented on SPARK-15930:
-
[~yuhaoyan] If you havent already started working on
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15332559#comment-15332559
]
Joseph K. Bradley commented on SPARK-15930:
---
This seems like a reasonable field to add to
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329564#comment-15329564
]
John Aherne commented on SPARK-15930:
-
Oh.. Sorry, you are correct, I had my statistical terms mixed
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329558#comment-15329558
]
Sean Owen commented on SPARK-15930:
---
Confidence just needs the support of two item sets, not the size
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329559#comment-15329559
]
John Aherne commented on SPARK-15930:
-
I don't necessarily know the site of the dataset being
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329548#comment-15329548
]
John Aherne commented on SPARK-15930:
-
Exactly. The number of rows in the training dataset is used to
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329087#comment-15329087
]
Sean Owen commented on SPARK-15930:
---
Yes, you already have that information, but I can see wanting that
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329074#comment-15329074
]
Jeff Zhang commented on SPARK-15930:
I see, I guess you are trying to get the total number of
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329072#comment-15329072
]
yuhao yang commented on SPARK-15930:
We need the size of input dataset to calculate relative
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329063#comment-15329063
]
Sean Owen commented on SPARK-15930:
---
The sum of all frequencies from freqItemSets is not the total size
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329059#comment-15329059
]
Jeff Zhang commented on SPARK-15930:
Don't we can get the count from freqItemsets in FPGrowthModel ?
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15329025#comment-15329025
]
Sean Owen commented on SPARK-15930:
---
What's the use case for this? the FP growth output is logically
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15328802#comment-15328802
]
yuhao yang commented on SPARK-15930:
That looks reasonable. +1.
[~John Aherne] I would wait for one
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15328786#comment-15328786
]
John Aherne commented on SPARK-15930:
-
In your example, the row count would be 4.
> Add Row count
[
https://issues.apache.org/jira/browse/SPARK-15930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15328759#comment-15328759
]
yuhao yang commented on SPARK-15930:
|| items|| freq||
|[27]|5|
|
16 matches
Mail list logo