{"id":2186,"date":"2020-01-10T00:20:42","date_gmt":"2020-01-10T00:20:42","guid":{"rendered":"http:\/\/kusuaks7\/?p=1791"},"modified":"2024-01-29T11:51:31","modified_gmt":"2024-01-29T11:51:31","slug":"competitive-advantage-how-to-put-machine-learning-models-into-production","status":"publish","type":"post","link":"https:\/\/www.experfy.com\/blog\/ai-ml\/competitive-advantage-how-to-put-machine-learning-models-into-production\/","title":{"rendered":"Competitive advantage: how to put machine learning models into production"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\nMost CTOs highlight machine learning as the technology that will disrupt their industry and lead to new innovations. But, how can they effectively put machine learning models into production?\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tMachine learning is a race. Those companies that can put machine learning models into production, on a large scale, first, will gain a huge advantage over their competitors and billions in potential revenue. But, there is a huge issue with the usability of machine learning \u2014 there is a significant\u00a0challenge around putting machine learning models into production at scale.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tOrganisations can create incredibly complex machine learning models, but it\u2019s problematic to take huge datasets, apply them to different iterations of ML models and then deploy those successful iterations into production.\n
\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

\"Machine learning versus AI, and putting data science models into production\"<\/h4><\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Machine learning is becoming the phrase that data scientists hide from CVs, putting a data science model into production is the biggest data challenge, and companies are still not getting it. We spoke to a data expert on the state of data science, and why machine learning is a more appropriate phrase than AI. Read here<\/a><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t

The machine learning landscape: where are we?<\/strong><\/h3><\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tWhen it comes to working out complex correlations \u2014 software development \u2014\u00a0\u201cno one in the world is talented enough to figure these out, there\u2019s too much noise in the data,\u201d explained\u00a0Eero Laaksonen<\/a>,\u00a0CEO at\u00a0Valohai<\/a>\u00a0\u2014 the machine learning platform.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\nThis is where ML and deep learning comes into play; it acts as the bridge between the start point and endpoint. ML, now, builds the function and the outcome is the model. \u201cThis is different from software development because the developers just write the function,\u201d continued Laaksonen. \u201cWith ML, it\u2019s combining the code with the data to define the model.\u201d\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tToday, the working methods with ML are very similar to what was happening in the 1990s with software development \u2014 developers are under increased pressure to deploy successful ML algorithms into production faster.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

\n

\u201cML models are still an art form\u201d \u2014\u00a0Eero Laaksonen<\/h3>\n<\/blockquote><\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Why is it difficult to put ML models into production?<\/strong><\/h3><\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThere are a number of reasons why it\u2019s difficult to put machine learning models into production\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t1. Experiment reproducibility:<\/strong>\u00a0the same combination of code and data (what the ML is trained on) can\u2019t be reproduced easily.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t2. Regulatory compliance:<\/strong>\u00a0\u201cML can\u2019t work in the wild west,\u201d said Laaksonen. \u201cOrganisations and regulators need to figure out laws around automated decision-making, something that is more reliable from the human perspective.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\n\u201cEurope has been proactive in this with GDPR, and it is a strike in the right direction, but it\u2019s difficult to bake regulation into ML production. Banks have to be able to explain automated decision they made six months ago \u2014 it\u2019s the output of data and models.\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cIf the ML model is running in production, currently there is no way to determine what caused it to come to that decision, and this needs to change. You need traceability and, therefore, version control in ML is very important.\u201d\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t3. Fast on-boarding of teams:\u00a0<\/strong>Organisations want to grow their developer and ML teams, but have disparate datasets and lots of code. It\u2019s difficult for people to come on board and see projects, whose working on them and identify where the data. \u201cAnd then there\u2019s the issue of hiring data scientists who are a hot commodity,\u201d continued Laaksonen. \u201cYou need the ability to track what they\u2019re doing if they leave, their projects and pipelines etcetera.\u201d\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t4. Quick experiments:\u00a0<\/strong>Speed is key, but with ML, it is very much a trial and error approach. The only way to try more things with ML is to put more hardware on it, which is quite challenging \u2014 every change must be tested with lots of data.\n\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Putting machine learning models into production faster<\/strong><\/h3><\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tValohai, which means Lantern Shark in Finnish, solves these problems as a machine learning platform-as-a-service \u2014 it illuminates deep learning and machine learning.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThe platform saves the datasets that have been used to run different ML models and shows who is accountable, the experimentation cost and what type of data is being used, among other metrics.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tEffectively, the platform fast tracks the trial and error model necessary in machine learning by forking the data pipeline into new models.\n\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\u201cWhen running on the cloud developers can\u2019t monitor how much processing power you\u2019re using. With our platform you can save the outputs and see the results on your data storage. Organisations can test different ML models on saved datasets from different sources and can run all the executions automatically with autoscaling on different cloud hosting providers,\u201d explained Laaksonen.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThe aim is to move from raw data to production faster, and the platform gives organisations the ability to rerun the data pipeline, retrain the model and deploy it.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThe aim is to move from raw data to production faster, and the platform gives organisations the ability to rerun the data pipeline, retrain the model and deploy it.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

\n

\u201cWe\u2019re at the phase where businesses will deploy ML on their own, before realising they need specialists\u201d \u2014 Laaksonen<\/h3>\n<\/blockquote><\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Use cases<\/h3><\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThere are a number of use cases, and Laaksonen referred to one of Valohai\u2019s customers, TwoHat Security, which is\u00a0building a model to stop distribution of child pornography. TwoHat Security is working with Canada\u2019s law enforcement and universities to build a machine vision model to detect sexual abuse material from darknets and other hard to reach places of the internet on top of Valohai\u2019s platform.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tThere are also other applications for predictive maintenance, predicting risks in financial services and with telecommunications for forecasting the location of future towers.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"

Those companies that can put machine learning models into production, on a large scale, first, will gain a huge advantage over their competitors and billions in potential revenue. But, there is a huge issue with the usability of machine learning — there is a significant challenge around putting machine learning models into production at scale. Organisations can create incredibly complex machine learning models, but it’s problematic to take huge datasets, apply them to different iterations of ML models and then deploy those successful iterations into production.<\/p>\n","protected":false},"author":635,"featured_media":3304,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","footnotes":""},"categories":[183],"tags":[92],"ppma_author":[3361],"class_list":["post-2186","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-ml","tag-machine-learning"],"authors":[{"term_id":3361,"user_id":635,"is_guest":0,"slug":"nicholas-ismail","display_name":"Nicholas Ismail","avatar_url":"https:\/\/www.experfy.com\/blog\/wp-content\/uploads\/2020\/04\/medium_7846d3d9-9e57-4968-8d69-64a8c8947899-150x150.jpg","user_url":"https:\/\/www.information-age.com\/","last_name":"Ismail","first_name":"Nicholas","job_title":"","description":"Nicholas Ismail is Editor at Information Age<\/a>.\u00a0 He writes original articles for the magazine and website, interviewing business leaders both on and off-camera while providing analysis for IT and business leaders on the latest technology trends disrupting the industry."}],"_links":{"self":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts\/2186","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/users\/635"}],"replies":[{"embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/comments?post=2186"}],"version-history":[{"count":6,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts\/2186\/revisions"}],"predecessor-version":[{"id":35724,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/posts\/2186\/revisions\/35724"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/media\/3304"}],"wp:attachment":[{"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/media?parent=2186"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/categories?post=2186"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/tags?post=2186"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.experfy.com\/blog\/wp-json\/wp\/v2\/ppma_author?post=2186"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}