2019-08-28 05:58:44 UTC - chris: @chetanm I changed the config in the application.conf. (Memory: max 512m => 2G and activation max 1048576 => 2097152)
After then, I reinstall the openwhisk by using docker compose again. But, nothing change. it shows the same error message. https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566971924000100?thread_ts=1566876260.021000&cid=C3TPCAQG1 ---- 2019-08-28 05:59:38 UTC - chris: The path is `/incubator-openwhisk-devtools/docker-compose/openwhisk-src/common/scala/src/main/resources/application.conf` https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566971978000300?thread_ts=1566876260.021000&cid=C3TPCAQG1 ---- 2019-08-28 07:55:39 UTC - Pepi Paraskevoulakou: Hey guys anyone’s who deployed his own ml model in python in whisk? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566978939001300 ---- 2019-08-28 07:56:47 UTC - Nikolai Starodubtcev: @Dave Grove thanks, I missed the isolation based on network policy. will take a look at it https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566979007001400?thread_ts=1566626075.016100&cid=C3TPCAQG1 ---- 2019-08-28 07:59:31 UTC - Nikolai Starodubtcev: If I get the point right what you need to do is simply to put your trained model in container and extend it in OpenWhisk? If so, I suggest to build a docker image with the code in it, using this documentation: <https://github.com/apache/openwhisk/blob/master/docs/actions-docker.md> Please, correct me if I'm wrong, and ml has much differences which I can see from my perspective https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566979171001600?thread_ts=1566911381.022900&cid=C3TPCAQG1 ---- 2019-08-28 08:02:48 UTC - Pepi Paraskevoulakou: Yes this is what I want to do but as I be read from many articles building a new docker image from scratch isn’t a simple job to do due to the fact that I must follow specific instructions which referred in the above mentioned article that you provided. So can I take a prefix image and change the code for example? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566979368005000?thread_ts=1566911381.022900&cid=C3TPCAQG1 ---- 2019-08-28 08:24:03 UTC - James Thomas: @Pepi Paraskevoulakou have you gone through the blog posts in the thread above? those do talk explicitly about this use-case… https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566980643005600?thread_ts=1566980643.005600&cid=C3TPCAQG1 ---- 2019-08-28 08:27:46 UTC - Pepi Paraskevoulakou: I saw your blog posts how to install libraries or how to use pre trained models like tensorflow but my use case is to build my own code and my own model which I will train it with my own dataset https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566980866007600?thread_ts=1566980643.005600&cid=C3TPCAQG1 ---- 2019-08-28 09:25:42 UTC - Pepi Paraskevoulakou: The article also Says that I have to write a manifest file and other things, maybe this is difficult ? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1566984342008700?thread_ts=1566911381.022900&cid=C3TPCAQG1 ----
