Developers should be prepared to integrate with and use their deep learning portfolio.  Before we get some scope on their API's and endpoints some background in deep learning (a specialized subset of machine learning concentrated on the "Greedy Layer-Wise Unsupervised Pretraining procedure" - Hinton, 2006 - University of Toronto) and the "Long Short-Term Memory model - Hochreiter and Schmidhuber 1997" will be required.

JIRAs

LOG-500 - Getting issue details... STATUS

LOG-511 - Getting issue details... STATUS

LOG-104 - Getting issue details... STATUS

Design/Theory Preparation

Get the following Nov 2016 MIT book from Amazon by Ian Goodfellow, Yoshua Benglo, Aaron Courville - (one for work, and one for home - as it is usually out of stock).  Review your linear transformation and matrix math to prep.

https://www.amazon.ca/Deep-Learning-Ian-Goodfellow/dp/0262035618/ref=sr_1_1?ie=UTF8&qid=1486501444&sr=8-1&keywords=deep+learning

https://developers.google.com/machine-learning/crash-course/

Installing Tensorflow

follow/verify via https://www.tensorflow.org/install/install_mac

obrien:obrienlabs amdocs$ docker run -it tensorflow/tensorflow bash
Unable to find image 'tensorflow/tensorflow:latest' locally
latest: Pulling from tensorflow/tensorflow
22dc81ace0ea: Pull complete 
1a8b3c87dba3: Pull complete 
91390a1c435a: Pull complete 
07844b14977e: Pull complete 
b78396653dae: Pull complete 
22bb9efa20f2: Pull complete 
e385adcc1f05: Pull complete 
da0eaa434771: Pull complete 
f3ad8e6a231d: Pull complete 
1da6064a2568: Pull complete 
e701d2aeb76d: Pull complete 
cba62e3f7418: Pull complete 
3b523741fd2e: Pull complete 
Digest: sha256:f3b5484e3335d2eb72940a6addde6f714173ff46c8a13c06aaf10915e96bc539
Status: Downloaded newer image for tensorflow/tensorflow:latest
root@4c49aac50ce1:/notebooks# python
Python 2.7.12 (default, Dec  4 2017, 14:50:18) 
[GCC 5.4.0 20160609] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
/usr/local/lib/python2.7/dist-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters
>>> hello = tf.constant('Hello, TensorFlow!')

>>> sess = tf.Session()

2018-04-22 23:57:53.221560: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
>>> print(sess.run(hello))
Hello, TensorFlow!


AIaaS

Open Source

Acumos

https://jenkins.acumos.org/

https://nexus.acumos.org/

https://wiki.acumos.org/display/AC/Acumos+Installation

http://docs.acumos.org/en/latest/

https://jira.acumos.org/browse/ACUMOS-849

meetings https://wiki.acumos.org/pages/viewpage.action?pageId=4653650

ONAP integration with Acumos -  LOG-264 - Getting issue details... STATUS , LOG-411 - Getting issue details... STATUS and https://jira.acumos.org/browse/ACUMOS-900

Docker architecture

ubuntu@ip-172-31-60-228:~$ sudo docker ps
CONTAINER ID        IMAGE                                                 COMMAND                  CREATED             STATUS              PORTS                                                                                         NAMES
afa39db75934        kong:0.11                                             "/docker-entrypoint.…"   About an hour ago   Up About an hour    0.0.0.0:80->8000/tcp, 0.0.0.0:8480->8001/tcp, 0.0.0.0:443->8443/tcp, 0.0.0.0:8443->8444/tcp   acumos_kong_1
952f648ad4cc        nexus3.acumos.org:10004/acumos-portal-fe:1.15.16      "/bin/sh -c 'cd /mav…"   About an hour ago   Up About an hour    0.0.0.0:8085->8085/tcp                                                                        acumos_portal-fe_1
757ceefa7345        nexus3.acumos.org:10004/onboarding-app:1.22.0         "/bin/sh -c 'cd /mav…"   About an hour ago   Up About an hour    0.0.0.0:8090->8090/tcp                                                                        acumos_onboarding-app_1
b355fa99fdc7        nexus3.acumos.org:10004/federation-gateway:1.1.2      "/bin/sh -c 'cd /mav…"   About an hour ago   Up About an hour    9011/tcp, 0.0.0.0:9084->9084/tcp                                                              acumos_federation-gateway_1
614b7b173da2        nexus3.acumos.org:10004/acumos-portal-be:1.15.16      "/bin/sh -c 'cd /mav…"   About an hour ago   Up About an hour    8083/tcp                                                                                      acumos_portal-be_1
e69d0230e872        nexus3.acumos.org:10004/acumos-filebeat:1.0.0         "/usr/local/bin/dock…"   About an hour ago   Up About an hour                                                                                                  acumos_filebeat_1
5d317cd0f8bb        nexus3.acumos.org:10004/acumos-azure-client:1.96.0    "/bin/sh -c 'java -X…"   About an hour ago   Up About an hour    9081/tcp                                                                                      acumos_azure-client_1
d6db55627deb        nexus3.acumos.org:10004/acumos-cms-docker:1.3.2       "/bin/sh -c 'cd /mav…"   About an hour ago   Up About an hour    0.0.0.0:9080->9080/tcp                                                                        acumos_cms_1
91b13dfcdbb9        sonatype/nexus3:3.9.0                                 "sh -c ${SONATYPE_DI…"   About an hour ago   Up About an hour    0.0.0.0:8001-8003->8001-8003/tcp, 0.0.0.0:8081->8081/tcp                                      acumos_nexus_1
158802c95df0        postgres:9.5                                          "docker-entrypoint.s…"   About an hour ago   Up About an hour    0.0.0.0:5432->5432/tcp                                                                        acumos_kong-database_1
f0564e008713        nexus3.acumos.org:10004/ds-compositionengine:0.0.26   "/bin/sh -c 'cd /mav…"   About an hour ago   Up About an hour    8088/tcp                                                                                      acumos_ds-compositionengine_1
3d552ab715c6        nexus3.acumos.org:10004/validation-middleware:1.2.1   "python validation_m…"   About an hour ago   Up About an hour    9604/tcp                                                                                      acumos_validation-middleware_1
674cb9d41af7        nexus3.acumos.org:10004/validation-engine:1.2.2       "python validation_e…"   About an hour ago   Up About an hour    9605/tcp                                                                                      acumos_validation-engine_1
1cdc2b4843b2        nexus3.acumos.org:10004/validation-client:1.2.1       "python validation_c…"   About an hour ago   Up About an hour    9603/tcp                                                                                      acumos_validation-client_1
6f91fb238c95        nexus3.acumos.org:10002/common-dataservice:1.14.3     "/bin/sh -c 'cd /mav…"   About an hour ago   Up About an hour    0.0.0.0:8000->8000/tcp                                                                        acumos_common-data-svc_1



TensorFlow

https://www.tensorflow.org/install/install_mac

OpenAI

https://github.com/openai/kubernetes-ec2-autoscaler

https://openai.com/systems/

http://www.cs.toronto.edu/~fritz/absps/ncfast.pdf

http://people.idsia.ch/~juergen/lstm2003tutorial.pdf

http://dl.acm.org/citation.cfm?id=1246450&CFID=898083621&CFTOKEN=58541063&qualifier=LU1039002

https://deeplearning4j.org/devguide

Commercial

https://aws.amazon.com/machine-learning/amis/

https://aws.amazon.com/mxnet/?&trk=ha_a131L000005CvfAQAS&sc_ichannel=ha&sc_icampaign=pac_q4_site_merch_sign-in_deep_learning&sc_ioutcome=Product_Adoption_Campaigns&sc_iplace=SignInPage

https://www.microsoft.com/cognitive-services/en-us/sign-up


https://deepfield.com/cloud-genome/

  • No labels

1 Comment

  1. Michael O'Brien

    michaelobrien [11:08 AM]
    had a 90 min talk with Nir over AI use cases - Meeting again 2nd next monday after I fill out some more potential use cases around anomaly detection or some sort of predictive analysis - would require us to start hammering the system with the automated distribute robot jobs and write up a backup of elasticsearch so we can save the training data. I also did not have a good answer as to why we don't move to a standard format for opentracing

    https://jira.onap.org/browse/LOG-104

    https://jira.onap.org/browse/LOG-500

    and https://wiki.onap.org/display/DW/Artificial+Intelligence+and+Deep+Learning