[cperezln]

I cannot, right now I am working remotly and I have no access to the cluster.

I have two questions on that.

• How much is “too much” data? I am sending by Kafka a 32 items dictionary each second, which after the pipeline will be appended as a new row to the table (with 32 columns).
• Is there any way on increasing the memory from Insights Enterpise itself?

Kind regards,

Cristian

[cperezln]

Why is not a valid query for qsql endpoint?

 

Thanks,

Cristian

[cperezln]

Hi ,

yes, I have correctly installed every requirement. Nevertheless, now appears to be working.

I have a couple of questions about the actual need to use ML-Toolkit, because now I am planning on using it in another machine, and so I need to configure everything:

– I am using it and working with Python 3.7, is there any recommendation of using another Python version?

– I am actually using a regular q license (not insights), I assume there is no need of using an insights license?

– I am working with:

• pandas 1.1.5
• matplotlib 3.5.3
• numpy 1.21.5
• scipy 1.7.3
• scikit-learn 0.23.0

Is there any recommendation about libraries versions, not strictly specified in requirements.txt?

 

Kind Regards,

Cristian

[cperezln]

I have solved the problem so far. To use the Q kernel with the Insights license, you need to use Python 3.7. I was using Python 3.8.

[cperezln]

Hi David,

I have jupyterq 1.1.13 version right now.

My license is kc.lic.

The flags I have in my license are:

• “insights.lib.embedq”
• “insights.lib.pykx”
• “insights.lib.sql”
• “insights.app.rt”
• “insights.lib.qlog”
• “insights.lib.kurl”
• “insights.lib.objstore”
• “insights.lib.bigquery”
• “insights.lib.restserver”

QLIC is actually pointing to  /home/tian/anaconda3/envs/jupyterqenv/qlic and QHOME to /home/tian/anaconda3/envs/jupyterqenv/q.

 

If you need any further information let me know.