# Daily Note 14/08/2020 ###### tags: `Daily Notes` , `Acumos` ## Name : Christofel Rio Goenawan ## University : Bandung Institute of Technology (ITB) --- ## Schedule: 1. Study Detailed Works of Acumos AI Marketplace from White Paper. 2. Study Background of Acumos AI Releases from White Paper. 3. Try to Solve Connection Issue in Jenkins Installation for Acumos AI. ## Outcome : 1. Explained Detailed Works of Acumos AI Marketplace from White Paper. 2. Explained Background of Acumos AI Releases from White Paper. 3. Solve Admin Permission for Acumos AI Installer Issue but Got New Issue in Jenkins's Set Up. ## Further Plan : - Continue to deploy Acumos AIO in NTUST server - Study more detailed about AIO Installation in Kubernets --- ## Daily Log ### 1.Study Detailed Works of Acumos AI Marketplace from White Paper.<mark>(11.00)</mark> - Study more detail explanation in [White Paper](https://cache.techmahindra.com/static/img/pdf/throught-leadership/thought-leadership-how-acumos.pdf) and other sources. ### 2.Study Background of Acumos AI Releases from White Paper. <mark>(12.00)</mark> - Study more detail explanation in [White Paper](https://cache.techmahindra.com/static/img/pdf/throught-leadership/thought-leadership-how-acumos.pdf) and other sources. ### 3.Try to Solve Connection Issue in Jenkins Installation for Acumos AI.<mark>(15.00)</mark> - Continue to try to deploy Acumos AIO in NTUST server using Prep- Deploy Process based one previous [study notes](https://hackmd.io/@christofel04/TEEP_Daily_Notes_10_7_2020). - Try to solve Connection Issue in Jenkins Installation for Acumos AI based one previous [study notes](https://hackmd.io/@christofel04/TEEP_Daily_Notes_13_8_2020). --- ## Report ### 1. Acumos AI Marketplace > - In this note Writer use [White Paper](https://cache.techmahindra.com/static/img/pdf/throught-leadership/thought-leadership-how-acumos.pdf) as reference. >- Acumos AI Marketplace can be seen [here](https://marketplace.acumos.org/#/home). From White Paper , Acumos AI Marketplace background can be seen as below. - Acumos leveraged the community development concept not just to develop its own code but also to create ***Market Place for Models***. - Borrowing the **app store** concept , Acumos provides **a marketplace where academics and commercial data scientist can sell specialized ML models and customized to address specific industrial use cases**. - Company can experiment with the crowdsourced models on their own datasets and maybe pay to use them in production when they like the models. - By minimizing the customization and contextualization needed to use off- the - shelf pre- trained models, users can **get insights more quickly and economically**. #### 1. Acumos AI Marketplace Interfaces The interface of Acumos AI Marketplace home site can be seen as below. ![](https://i.imgur.com/paQDD8d.png) The page of Acumos AI Marketplace model example can be seen as below. ![](https://i.imgur.com/Daxwbsv.png) #### 2. What is Inside Acumos AI Marketplace ? - Some models are available in the marketplace that are generic enough to be re- trained for a specifict context. - These models cover areas such as below. - **Image Classification** - **Face Detection** - **Face Privacy Filtering** - **Image Mood Classification** - etc #### 3. Issue in Acumos AI Marketplace - While the basic mathematical principles behind AI are not subject to copyright , **models that have been trained on a particular type of dara can be copyrighted**. - Michael Lanzetta notes as below. :::info ***Currently the big players are sharing their findings ( new neuron types , etc ) but we expect that to become rarer as new neuronal structures become valuable intellectual property***. ::: - So Acumos AI Marketplace **could provide an opportunity for AI Scientist to market the wares and for business users to tap into the expertise of AI Specialists**. #### 4. Development in Acumos AI Marketplace - Acumos AI initiator such as AT&T , Amdocs , etc plans to **contribute more models to Acumos Marketplace over time , enabling anyone to download , use and customized them**. - Wider **variaety of models to choose in Acumos AI Marketplace**. As Michael Lanzetta explained as below. :::info ***Since it sometimes takes days or weeks to train a model , picking the right model structure, optimizations techniques, learning rates , and other hyperparameters an make difference between success and failure***. ::: --- ### 2. Acumos AI Releases >In this note Writer use [White Paper](https://cache.techmahindra.com/static/img/pdf/throught-leadership/thought-leadership-how-acumos.pdf) as study sources. Currently Acumos AI have two release as below. 1. **Athena Releases** 2. **Boreas Releases** #### 1. Athena Releases :::info Release Name: **Athena** Release Version: **1.0.0** Release Date: **7 November 2018** ::: With Athena releases, users can now **deploy models on their own hardware , including servers and virtual machines in a private Kubernetes environment or into a public or private clodu infrastructutes** for functions such as **testing and training**. Operation wise , we can now **shares models privately within your tea , and with specific companies** , as well as share them publicly in the marketplace. We can also **set up role- based permissions within our organization for acessing and using Acumos AI Modles**. ##### 2. Boreas Releases > Because White Paper made before Boreas's releases, Writer use [reference](https://www.lightreading.com/artificial-intelligence-machine-learning/acumos-ai-project-adds-smarts-with-boreas/a/d-id/752359) for Study Sources. :::info Release Name: **Boreas** Release Version: **1.0.0** Release Date: **5 June 2019** ::: Boreas Releases introduce more convenient model training as well as data extraction pipelines to make models more flexible. Boreas include **updates to assist closed- source model developers** , including **secure and reliable licensing** components to provide **execution control and performance feeedback across the community**. Boreas also include **integration with Linux Foundation's ONAP and Akraino projects** and demonstrate some **5G Use Cases**. --- ### 3. Try to Solve Connection Issue in Jenkins Installation >This deployment is continuation from [yesterday's notes]( https://hackmd.io/@christofel04/TEEP_Daily_Notes_14_8_2020 ). In this note Writer use [Previous Notes](https://hackmd.io/@christofel04/TEEP_Daily_Notes_10_7_2020) as study sources. After solving Open Shift Cluster issues , next Writer set up Jenkins services for Acumos AI in NTUST Server. :::danger In Writer's installation the installation stop when try to log in to server as below. ![](https://i.imgur.com/YNhb8eR.png) ::: :::info **Next Writer Will Continue to Solve the Jenkins Issue in NTUST Server** ::: --- ## Reference 1. https://github.com/nokia/ONAP-VESPA 2. https://docs.o-ran-sc.org/projects/o-ran-sc-ric-plt-vespamgr/en/latest/overview.html 3. https://wiki.o-ran-sc.org/display/RICP/RIC+Alarm+System