Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
cliveseldon
@cliveseldon
If you wish to call the endpoints in different circumstances you could use http://docs.seldon.io/advanced-recommender-config.html#recommendation-variants
However, I think its easier to simply have two clients and thus separate API requests as needed
Paolo Squadrito
@Rocknpools
Can i use the same mysql datasource (in the first case)?
for example "Client1" and "Client2" share the same mysql db “ClientDB"
ps. is there a way to force the use of only last recent action?
cliveseldon
@cliveseldon
yes you can use same ClientDB, they would be two different databases inside it, client1 and client2
to specify the number of recent actions the semantic vector recomender uses you can set the option in the configuration io.seldon.algorithm.general.numrecentactionstouse
JayeshLunkad
@JayeshLunkad
Hello All, i was trying seldon on aws, i had setup seldon on it. The prediction example is running fine . i was trying tensorflow-deep-mnist-example.i had started the webservice and web app . but its giving 500 internal server error in web. whats wrong in it ?
Maximophone
@Maximophone
Hi Jayesh, are you following the simple or advanced tutorial? Will try reproducing the steps for you. Could you also tell me which version of the webapp and of seldon you are using? Thanks
Alex Housley
@ahousley
@JayeshLunkad thanks for your question, please see the above response from @Maximophone
cliveseldon
@cliveseldon
We have updated the version of the web app to 1.1 to fix a bug. http://docs.seldon.io/tensorflow-deep-mnist-webapp.html can you try?
Paolo Squadrito
@Rocknpools
Hi guys. I'm trying to connect to the mysql instance of the kubernetes cluster from a client on my pc. The cluster is correctly deployed on Azure in LoadBalancing mode so basically i have to activate a rule to forward the connection request to mysql. So my question is this: is mysql mapped on a different port then the standard one?
cliveseldon
@cliveseldon
It should be the standard port. You would need to expose that port to the outside world to connect from outside the cluster. So you would need to do that via kubectl or change mysql.json in kubernetes/conf
Paolo Squadrito
@Rocknpools
thanks! I have another question. I'm using the variable EDITOR=nano but when i use seldon-cli attr it run a vim session which actually doesen't work and i don't know why. Any suggestion?
Paolo Squadrito
@Rocknpools
I tried seldon-cli attr --action edit --client-name motorsquare --json /path/to/json/attr.json as well but a vim session start and i get a strange error like :Vim: Warning: Output is not to a terminal Vim: Warning: Input is not from a terminal
Paolo Squadrito
@Rocknpools
@cliveseldon
Michal Trmac
@michaltrmac
Hello All, is there someone who know how to set up "Most popular recommender"?
Shreyas Srinivasan
@shreyas87
hi, new to seldon server. seldon up is taking a long time to setup the mysql container. is this normal?
Shreyas Srinivasan
@shreyas87
nvm, assigned more RAM to the virtual machine and that solved it.
Alex
@alextorar_twitter
if i have my own models and other models made by h2o ai and tensorflow, , can i use seldon ai for deploying them like restapis?
akashkatakam
@akashkatakam
AME READY STATUS RESTARTS AGE
hello-minikube-3015430129-ltgb3 1/1 Running 0 1h
influxdb-grafana-vb207 2/2 Running 0 10m
memcached1-z528t 1/1 Running 0 10m
memcached2-c2xzq 1/1 Running 0 10m
mysql 0/1 Pending 0 10m
seldon-control 1/1 Running 0 10m
zookeeper-1 1/1 Running 0 10m
zookeeper-2 1/1 Running 0 10m
zookeeper-3 1/1 Running 0 10m
Waiting for pods to be running as found 1 in non-running state
what can be the reason for this
do i need to have my sql installed on my system?
zack bayhan
@zacbayhan_twitter
Hello everyone hope this message finds you well, was wondering if anyone has ran into ./seldon-up process getting stuck on "waiting for zookeeper to be ready as status check returned
also is anyone currently using supergiant's ui for kubernetes?
hzmbeyond80
@hzmbeyond80
hi, i started seldon server on a local server. can anyone point me to how to access that server via restful api? all the examples were given using seldon-cli
hzmbeyond80
@hzmbeyond80
is there anybody in the room?
Alan Justino da Silva
@alanjds
Hi.
I am evaluating seldon. Had read the docs and found microservices for prediction and for recomendation
Is there microservices for other stuff, like ETL ??
Alan Justino da Silva
@alanjds
I mean, transformation-only services
cliveseldon
@cliveseldon
The simplest way for prediction services is to simply create a scikit learn pipeline as described in the docs http://docs.seldon.io/prediction-pipeline.html
Paolo Squadrito
@Rocknpools
@cliveseldon @gsunner Hi guys, could someone give me some quick hints on how to setup the dimension mapping for a given user?https://github.com/SeldonIO/seldon-server/issues/46#issuecomment-287009290
cliveseldon
@cliveseldon
I updated the github issues with some hints to help on dimension mapping
Mayur911
@Mayur911
Hey Guys,
i am new to seldon and wanted to make my own item similarity model, but while adding schema to my newly created client, it opens up a vim editor with default schema and doesn't allow me to use arrow keys or spacebar or anything, also some error is displayed saying the file you are trying to edit is being used (opened) some where else... can someone help me on this ?
rameshandroid
@rameshandroid
how to deploy server code
??
geoHeil
@geoHeil
@cliveseldon is Seldon still only supporting python 2?
Naga Hari Krishna
@Naga_Hari_Krish_twitter
@cliveseldon Hi Guys,
i have setup of seldon server. could someone help me to how to access that server via restful api? all the examples were given using seldon-cli.
cliveseldon
@cliveseldon
An endpoint will be exposed as discussed at http://docs.seldon.io/install.html#endpoint
You should be able to fire REST calls to it
@geoHeil Yes, at present the python parts of the code is only python 2
Naga Hari Krishna
@Naga_Hari_Krish_twitter
@cliveseldon Thanks for reply,
I have done all the steps, how can we find the host to call rest api?
cliveseldon
@cliveseldon
It depends how you are running kubernetes. If its in the Cloud then a loadbalancer should be exposed on an IP address shown in "kubectl get service" otherwise if using minikube then it will be a nodePort exposed on the local minikube ip found by "minikube ip".
Edoardo Benissimo
@EdoardoB

Hi guys,
I installed Kubernetes on AWS using the official guidance.
Now, I have a node master (m3.medium) and two workers (t2.medium) on which I need to run Seldon.
I'm using the official guidance: http://docs.seldon.io/install.html
But when i run the seldon-up.sh on Kubernetes, I obtained the following error during the creation of containers seldon-control-3388397084-pvr8k and influxdb-grafana-3089120915-fzxt4 :

Multi-Attach error for volume “pvc-d97313ee-93e1-11e7-ae85-02f1ec9e7a70” Volume is already exclusively attached to one node and can’t be attached to another Unable to mount volumes for pod “influxdb-grafana-3089120915-fzxt4_default(e21ad2fc-93e1-11e7-ae85-02f1ec9e7a70)“: timeout expired waiting for volumes to attach/mount for pod “default”/“influxdb-grafana-3089120915-fzxt4". list of unattached/unmounted volumes=[storage] Error syncing pod

How could I solve this issue?

Edoardo Benissimo
@EdoardoB
Ok, I resolved this issue!
yaserrahimi
@yaserrahimi
hi
is there a video aboate how set and run seldon on local host
cliveseldon
@cliveseldon
No not at the moment. Its best you follow http://docs.seldon.io/install.html