�ݺ�ߣshows by User: SetuChokshi

�ݺ�ߣshows by User: SetuChokshi / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: SetuChokshi / Sat, 19 Jan 2019 01:40:21 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: SetuChokshi Build vs Buy: Ensuring maximum ROI from AI /slideshow/build-vs-buy-ensuring-maximum-roi-from-ai/128441105 buildvsbuy-190119014021
Whether it consists of building internal data science capability, integrating AI into lines of business, purchasing new tools, or collaborating with external partners - the range of choices can often be a road-block to getting AI projects off the ground in the first place. The presentation will explore what are the benefits and disadvantages of build vs buy and how to avoid AI for PR based activities that will waste time and money and potentially produce negative publicity.]]>
Whether it consists of building internal data science capability, integrating AI into lines of business, purchasing new tools, or collaborating with external partners - the range of choices can often be a road-block to getting AI projects off the ground in the first place. The presentation will explore what are the benefits and disadvantages of build vs buy and how to avoid AI for PR based activities that will waste time and money and potentially produce negative publicity.]]> Sat, 19 Jan 2019 01:40:21 GMT /slideshow/build-vs-buy-ensuring-maximum-roi-from-ai/128441105 SetuChokshi@slideshare.net(SetuChokshi) Build vs Buy: Ensuring maximum ROI from AI SetuChokshi Whether it consists of building internal data science capability, integrating AI into lines of business, purchasing new tools, or collaborating with external partners - the range of choices can often be a road-block to getting AI projects off the ground in the first place. The presentation will explore what are the benefits and disadvantages of build vs buy and how to avoid AI for PR based activities that will waste time and money and potentially produce negative publicity. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/buildvsbuy-190119014021-thumbnail.jpg?width=120&height=120&fit=bounds" /> Whether it consists of building internal data science capability, integrating AI into lines of business, purchasing new tools, or collaborating with external partners - the range of choices can often be a road-block to getting AI projects off the ground in the first place. The presentation will explore what are the benefits and disadvantages of build vs buy and how to avoid AI for PR based activities that will waste time and money and potentially produce negative publicity.

Build vs Buy: Ensuring maximum ROI from AI from Setu Chokshi

]]> 237 1 https://cdn.slidesharecdn.com/ss_thumbnails/buildvsbuy-190119014021-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 AI for AI: Building state of the art models /slideshow/ai-for-ai-building-state-of-the-art-models/128441103 aiforaifinal-190119014019
Azure Cognitive Services and AutoML (and NNI) are among the offerings that allows a developer with limited knowledge of AI to build state of the art models with ease and speed. In this demo we will learn to build a real time iOS image classifier, training a model to answer Trivia questions and train ensemble models for regression. ]]>
Azure Cognitive Services and AutoML (and NNI) are among the offerings that allows a developer with limited knowledge of AI to build state of the art models with ease and speed. In this demo we will learn to build a real time iOS image classifier, training a model to answer Trivia questions and train ensemble models for regression. ]]> Sat, 19 Jan 2019 01:40:19 GMT /slideshow/ai-for-ai-building-state-of-the-art-models/128441103 SetuChokshi@slideshare.net(SetuChokshi) AI for AI: Building state of the art models SetuChokshi Azure Cognitive Services and AutoML (and NNI) are among the offerings that allows a developer with limited knowledge of AI to build state of the art models with ease and speed. In this demo we will learn to build a real time iOS image classifier, training a model to answer Trivia questions and train ensemble models for regression. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/aiforaifinal-190119014019-thumbnail.jpg?width=120&height=120&fit=bounds" /> Azure Cognitive Services and AutoML (and NNI) are among the offerings that allows a developer with limited knowledge of AI to build state of the art models with ease and speed. In this demo we will learn to build a real time iOS image classifier, training a model to answer Trivia questions and train ensemble models for regression.

AI for AI: Building state of the art models from Setu Chokshi

]]> 178 1 https://cdn.slidesharecdn.com/ss_thumbnails/aiforaifinal-190119014019-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Microsoft Introduction to Automated Machine Learning /slideshow/microsoft-introduction-to-automated-machine-learning/126008425 globalaibootcampautomatedml-181216114355
A gentle introduction to Microsoft's AutoML SDK package. This presentation introduces the concept of why automated machine learning has an important place in any data scientists tool box. Auto ML SDK allows you to to build and run machine learning workflows with the Azure Machine Learning service. You can interact with the service in any Python environment, including Jupyter Notebooks or your favourite Python IDE. The demos included in the presentation are making use of the Azure Notebooks. ]]>
A gentle introduction to Microsoft's AutoML SDK package. This presentation introduces the concept of why automated machine learning has an important place in any data scientists tool box. Auto ML SDK allows you to to build and run machine learning workflows with the Azure Machine Learning service. You can interact with the service in any Python environment, including Jupyter Notebooks or your favourite Python IDE. The demos included in the presentation are making use of the Azure Notebooks. ]]> Sun, 16 Dec 2018 11:43:55 GMT /slideshow/microsoft-introduction-to-automated-machine-learning/126008425 SetuChokshi@slideshare.net(SetuChokshi) Microsoft Introduction to Automated Machine Learning SetuChokshi A gentle introduction to Microsoft's AutoML SDK package. This presentation introduces the concept of why automated machine learning has an important place in any data scientists tool box. Auto ML SDK allows you to to build and run machine learning workflows with the Azure Machine Learning service. You can interact with the service in any Python environment, including Jupyter Notebooks or your favourite Python IDE. The demos included in the presentation are making use of the Azure Notebooks. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/globalaibootcampautomatedml-181216114355-thumbnail.jpg?width=120&height=120&fit=bounds" /> A gentle introduction to Microsoft's AutoML SDK package. This presentation introduces the concept of why automated machine learning has an important place in any data scientists tool box. Auto ML SDK allows you to to build and run machine learning workflows with the Azure Machine Learning service. You can interact with the service in any Python environment, including Jupyter Notebooks or your favourite Python IDE. The demos included in the presentation are making use of the Azure Notebooks.

Microsoft Introduction to Automated Machine Learning from Setu Chokshi

]]> 383 4 https://cdn.slidesharecdn.com/ss_thumbnails/globalaibootcampautomatedml-181216114355-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 2018 Global Azure Bootcamp Azure Machine Learning for neural networks /slideshow/2018-global-azure-bootcamp-azure-machine-learning-for-neural-networks-110784821/110784821 2018gabazuremachinelearninghandsonlab-neuralnetworks-180821040504
This was the introduction session done for the 2018 Global Azure Bootcamp to get the users started with neural networks on Azure Machine Learning Studio. This gives them the initial introduction on how to develop and write the neural networks. We started with writing LeNet architecture on Azure Machine Learning studio to identify handwritten digits and then moved on to cats and dogs. This was also the presented in the first workshop of my meetup Microsoft Ai, ML Community which can be reached here https://www.meetup.com/Microsoft-AI-ML-Community/]]>
This was the introduction session done for the 2018 Global Azure Bootcamp to get the users started with neural networks on Azure Machine Learning Studio. This gives them the initial introduction on how to develop and write the neural networks. We started with writing LeNet architecture on Azure Machine Learning studio to identify handwritten digits and then moved on to cats and dogs. This was also the presented in the first workshop of my meetup Microsoft Ai, ML Community which can be reached here https://www.meetup.com/Microsoft-AI-ML-Community/]]> Tue, 21 Aug 2018 04:05:04 GMT /slideshow/2018-global-azure-bootcamp-azure-machine-learning-for-neural-networks-110784821/110784821 SetuChokshi@slideshare.net(SetuChokshi) 2018 Global Azure Bootcamp Azure Machine Learning for neural networks SetuChokshi This was the introduction session done for the 2018 Global Azure Bootcamp to get the users started with neural networks on Azure Machine Learning Studio. This gives them the initial introduction on how to develop and write the neural networks. We started with writing LeNet architecture on Azure Machine Learning studio to identify handwritten digits and then moved on to cats and dogs. This was also the presented in the first workshop of my meetup Microsoft Ai, ML Community which can be reached here https://www.meetup.com/Microsoft-AI-ML-Community/ <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2018gabazuremachinelearninghandsonlab-neuralnetworks-180821040504-thumbnail.jpg?width=120&height=120&fit=bounds" /> This was the introduction session done for the 2018 Global Azure Bootcamp to get the users started with neural networks on Azure Machine Learning Studio. This gives them the initial introduction on how to develop and write the neural networks. We started with writing LeNet architecture on Azure Machine Learning studio to identify handwritten digits and then moved on to cats and dogs. This was also the presented in the first workshop of my meetup Microsoft Ai, ML Community which can be reached here https://www.meetup.com/Microsoft-AI-ML-Community/

2018 Global Azure Bootcamp Azure Machine Learning for neural networks from Setu Chokshi

]]> 137 2 https://cdn.slidesharecdn.com/ss_thumbnails/2018gabazuremachinelearninghandsonlab-neuralnetworks-180821040504-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Azure machine learning 101 Parts 1 & 2 - Classification Algorithms /slideshow/azure-machine-learning-101-parts-1-classification-algorithms/81219958 azuremachinelearning101-part12-171026025623
This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017 and 25th October 2017. These were the slides that were presented in the session. We covered the Decision Tree and Support Vector Machine (SVM) based algorithms: 1. Two-Class /Multiclasss Decision Forest 2. Two-Class /Multiclasss Decision Jungle 3. Two-Class Boosted Decision Tree 4. Two-Class Support Vector Machine (SVM) 5. Two-Class Locally Deep Support Vector Machine (LDSVM) We also looked at how to look at Feature Importance using "Permutation Feature Importance" module We also looked at "Partition and Sample" module and how to use it together with "Cross Validate Model" module. ]]>
This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017 and 25th October 2017. These were the slides that were presented in the session. We covered the Decision Tree and Support Vector Machine (SVM) based algorithms: 1. Two-Class /Multiclasss Decision Forest 2. Two-Class /Multiclasss Decision Jungle 3. Two-Class Boosted Decision Tree 4. Two-Class Support Vector Machine (SVM) 5. Two-Class Locally Deep Support Vector Machine (LDSVM) We also looked at how to look at Feature Importance using "Permutation Feature Importance" module We also looked at "Partition and Sample" module and how to use it together with "Cross Validate Model" module. ]]> Thu, 26 Oct 2017 02:56:23 GMT /slideshow/azure-machine-learning-101-parts-1-classification-algorithms/81219958 SetuChokshi@slideshare.net(SetuChokshi) Azure machine learning 101 Parts 1 & 2 - Classification Algorithms SetuChokshi This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017 and 25th October 2017. These were the slides that were presented in the session. We covered the Decision Tree and Support Vector Machine (SVM) based algorithms: 1. Two-Class /Multiclasss Decision Forest 2. Two-Class /Multiclasss Decision Jungle 3. Two-Class Boosted Decision Tree 4. Two-Class Support Vector Machine (SVM) 5. Two-Class Locally Deep Support Vector Machine (LDSVM) We also looked at how to look at Feature Importance using "Permutation Feature Importance" module We also looked at "Partition and Sample" module and how to use it together with "Cross Validate Model" module. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/azuremachinelearning101-part12-171026025623-thumbnail.jpg?width=120&height=120&fit=bounds" /> This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017 and 25th October 2017. These were the slides that were presented in the session. We covered the Decision Tree and Support Vector Machine (SVM) based algorithms: 1. Two-Class /Multiclasss Decision Forest 2. Two-Class /Multiclasss Decision Jungle 3. Two-Class Boosted Decision Tree 4. Two-Class Support Vector Machine (SVM) 5. Two-Class Locally Deep Support Vector Machine (LDSVM) We also looked at how to look at Feature Importance using "Permutation Feature Importance" module We also looked at "Partition and Sample" module and how to use it together with "Cross Validate Model" module.

Azure machine learning 101 Parts 1 & 2 - Classification Algorithms from Setu Chokshi

]]> 543 1 https://cdn.slidesharecdn.com/ss_thumbnails/azuremachinelearning101-part12-171026025623-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Azure machine learning 101 - Part 1 /slideshow/azure-machine-learning-101-part-1/79753724 azuremachinelearning101-part1-170914050333
This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017. These were the slides that were presented in the session. The excel file is available separately and can be downloaded from here (https://1drv.ms/x/s!AmjwdE_MMESksHOrlLXP400eHb3p). We covered on how decision trees are created and extended the concept to building Random Forest algorithm. ]]>
This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017. These were the slides that were presented in the session. The excel file is available separately and can be downloaded from here (https://1drv.ms/x/s!AmjwdE_MMESksHOrlLXP400eHb3p). We covered on how decision trees are created and extended the concept to building Random Forest algorithm. ]]> Thu, 14 Sep 2017 05:03:33 GMT /slideshow/azure-machine-learning-101-part-1/79753724 SetuChokshi@slideshare.net(SetuChokshi) Azure machine learning 101 - Part 1 SetuChokshi This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017. These were the slides that were presented in the session. The excel file is available separately and can be downloaded from here (https://1drv.ms/x/s!AmjwdE_MMESksHOrlLXP400eHb3p). We covered on how decision trees are created and extended the concept to building Random Forest algorithm. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/azuremachinelearning101-part1-170914050333-thumbnail.jpg?width=120&height=120&fit=bounds" /> This was the first introductory course on getting started with Azure Machine Learning that was held by Microsoft User Group on 13th September 2017. These were the slides that were presented in the session. The excel file is available separately and can be downloaded from here (https://1drv.ms/x/s!AmjwdE_MMESksHOrlLXP400eHb3p). We covered on how decision trees are created and extended the concept to building Random Forest algorithm.

Azure machine learning 101 - Part 1 from Setu Chokshi

]]> 217 2 https://cdn.slidesharecdn.com/ss_thumbnails/azuremachinelearning101-part1-170914050333-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Analysis on the US Consumer Expenditure /slideshow/analysis-on-the-us-consumer-expenditure/78589017 pivotalcedataexercisev2-170805101609
Given the US Consumer Expenses (Expenditure) dataset for 1996-‐2000 containing 12,000 rows and 220 columns. The objective of the analysis was to propose one way of using the data employing one of the following methods: regression, classification or clustering. This presentation shows my approach and methodology I have also shared the insights from the model and how it could be presented to a senior level manager with(out) the technical details. ]]>
Given the US Consumer Expenses (Expenditure) dataset for 1996-‐2000 containing 12,000 rows and 220 columns. The objective of the analysis was to propose one way of using the data employing one of the following methods: regression, classification or clustering. This presentation shows my approach and methodology I have also shared the insights from the model and how it could be presented to a senior level manager with(out) the technical details. ]]> Sat, 05 Aug 2017 10:16:09 GMT /slideshow/analysis-on-the-us-consumer-expenditure/78589017 SetuChokshi@slideshare.net(SetuChokshi) Analysis on the US Consumer Expenditure SetuChokshi Given the US Consumer Expenses (Expenditure) dataset for 1996-‐2000 containing 12,000 rows and 220 columns. The objective of the analysis was to propose one way of using the data employing one of the following methods: regression, classification or clustering. This presentation shows my approach and methodology I have also shared the insights from the model and how it could be presented to a senior level manager with(out) the technical details. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/pivotalcedataexercisev2-170805101609-thumbnail.jpg?width=120&height=120&fit=bounds" /> Given the US Consumer Expenses (Expenditure) dataset for 1996-‐2000 containing 12,000 rows and 220 columns. The objective of the analysis was to propose one way of using the data employing one of the following methods: regression, classification or clustering. This presentation shows my approach and methodology I have also shared the insights from the model and how it could be presented to a senior level manager with(out) the technical details.

Analysis on the US Consumer Expenditure from Setu Chokshi

]]> 252 2 https://cdn.slidesharecdn.com/ss_thumbnails/pivotalcedataexercisev2-170805101609-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Azure Boot Camp 2017 getting started with azure machine learning /slideshow/azure-boot-camp-2017-getting-started-with-azure-machine-learning/75285050 abc2017gettingstartedwithazuremachinelearning-170421161911
This presentation was done at the Azure Boot Camp 2017 in Singapore held on 23rd April 2017. This presentation covers an general introduction to machine learning and Azure Machine Learning Platform. To introduce the various features of the platform we make use of the Matchbox Recommender to build a collaborative filtering on the MovieLens dataset. We also obtain the state of art Normalized Discounted Cumulative Gain (NDCG) of 0.92 without supplementing any additional user data or movie tags. We just use the movie ratings as the base. I also show how to build an intuition for the recommendation systems using an Excel Worksheet and show how the latent factors in the recommendation systems are built and what the algorithm does to obtain an optimal solution. The link to the Excel Worksheet will be made available after the session is over. ]]>
This presentation was done at the Azure Boot Camp 2017 in Singapore held on 23rd April 2017. This presentation covers an general introduction to machine learning and Azure Machine Learning Platform. To introduce the various features of the platform we make use of the Matchbox Recommender to build a collaborative filtering on the MovieLens dataset. We also obtain the state of art Normalized Discounted Cumulative Gain (NDCG) of 0.92 without supplementing any additional user data or movie tags. We just use the movie ratings as the base. I also show how to build an intuition for the recommendation systems using an Excel Worksheet and show how the latent factors in the recommendation systems are built and what the algorithm does to obtain an optimal solution. The link to the Excel Worksheet will be made available after the session is over. ]]> Fri, 21 Apr 2017 16:19:10 GMT /slideshow/azure-boot-camp-2017-getting-started-with-azure-machine-learning/75285050 SetuChokshi@slideshare.net(SetuChokshi) Azure Boot Camp 2017 getting started with azure machine learning SetuChokshi This presentation was done at the Azure Boot Camp 2017 in Singapore held on 23rd April 2017. This presentation covers an general introduction to machine learning and Azure Machine Learning Platform. To introduce the various features of the platform we make use of the Matchbox Recommender to build a collaborative filtering on the MovieLens dataset. We also obtain the state of art Normalized Discounted Cumulative Gain (NDCG) of 0.92 without supplementing any additional user data or movie tags. We just use the movie ratings as the base. I also show how to build an intuition for the recommendation systems using an Excel Worksheet and show how the latent factors in the recommendation systems are built and what the algorithm does to obtain an optimal solution. The link to the Excel Worksheet will be made available after the session is over. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/abc2017gettingstartedwithazuremachinelearning-170421161911-thumbnail.jpg?width=120&height=120&fit=bounds" /> This presentation was done at the Azure Boot Camp 2017 in Singapore held on 23rd April 2017. This presentation covers an general introduction to machine learning and Azure Machine Learning Platform. To introduce the various features of the platform we make use of the Matchbox Recommender to build a collaborative filtering on the MovieLens dataset. We also obtain the state of art Normalized Discounted Cumulative Gain (NDCG) of 0.92 without supplementing any additional user data or movie tags. We just use the movie ratings as the base. I also show how to build an intuition for the recommendation systems using an Excel Worksheet and show how the latent factors in the recommendation systems are built and what the algorithm does to obtain an optimal solution. The link to the Excel Worksheet will be made available after the session is over.

Azure Boot Camp 2017 getting started with azure machine learning from Setu Chokshi

]]> 244 3 https://cdn.slidesharecdn.com/ss_thumbnails/abc2017gettingstartedwithazuremachinelearning-170421161911-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Machine Learning 101 /slideshow/machine-learning-101-74017810/74017810 ctumachinelearning101-170331011251
This was the presentation for the Microsoft Community Technology Update of 2016. The idea was to introduce to people the concept of Machine Learning and its easy to get started if you are keen. My objective was also to communicate how some of the algorithms work and they require no more than basic understanding of Math to get going, sometimes not even that. The algorithms we covered were, Support Vector Machines (SVM), Decision Tree using R2D3 and Neural Networks for classification. We used the Tensorflow Playground to help understand the Neural Network and Deep Learning concepts. I gave an analogy of how Machine Learning process is like making a smoothie where your algorithm is a recipe, your data are your ingredients, your computer is your blender and your smoothie is the model that you developed. I used the same example to convey the concept of Training Validation and Testing. Coverage of Type 1 and Type 2 errors together with the metrics of Recall and Precision was covered as well. Finally I closed the session with what are some good resources to get started with Machine Learning for all skill levels. There are references to websites, courses, kaggle competition, podcasts, cheat sheets and books. ]]>
This was the presentation for the Microsoft Community Technology Update of 2016. The idea was to introduce to people the concept of Machine Learning and its easy to get started if you are keen. My objective was also to communicate how some of the algorithms work and they require no more than basic understanding of Math to get going, sometimes not even that. The algorithms we covered were, Support Vector Machines (SVM), Decision Tree using R2D3 and Neural Networks for classification. We used the Tensorflow Playground to help understand the Neural Network and Deep Learning concepts. I gave an analogy of how Machine Learning process is like making a smoothie where your algorithm is a recipe, your data are your ingredients, your computer is your blender and your smoothie is the model that you developed. I used the same example to convey the concept of Training Validation and Testing. Coverage of Type 1 and Type 2 errors together with the metrics of Recall and Precision was covered as well. Finally I closed the session with what are some good resources to get started with Machine Learning for all skill levels. There are references to websites, courses, kaggle competition, podcasts, cheat sheets and books. ]]> Fri, 31 Mar 2017 01:12:50 GMT /slideshow/machine-learning-101-74017810/74017810 SetuChokshi@slideshare.net(SetuChokshi) Machine Learning 101 SetuChokshi This was the presentation for the Microsoft Community Technology Update of 2016. The idea was to introduce to people the concept of Machine Learning and its easy to get started if you are keen. My objective was also to communicate how some of the algorithms work and they require no more than basic understanding of Math to get going, sometimes not even that. The algorithms we covered were, Support Vector Machines (SVM), Decision Tree using R2D3 and Neural Networks for classification. We used the Tensorflow Playground to help understand the Neural Network and Deep Learning concepts. I gave an analogy of how Machine Learning process is like making a smoothie where your algorithm is a recipe, your data are your ingredients, your computer is your blender and your smoothie is the model that you developed. I used the same example to convey the concept of Training Validation and Testing. Coverage of Type 1 and Type 2 errors together with the metrics of Recall and Precision was covered as well. Finally I closed the session with what are some good resources to get started with Machine Learning for all skill levels. There are references to websites, courses, kaggle competition, podcasts, cheat sheets and books. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/ctumachinelearning101-170331011251-thumbnail.jpg?width=120&height=120&fit=bounds" /> This was the presentation for the Microsoft Community Technology Update of 2016. The idea was to introduce to people the concept of Machine Learning and its easy to get started if you are keen. My objective was also to communicate how some of the algorithms work and they require no more than basic understanding of Math to get going, sometimes not even that. The algorithms we covered were, Support Vector Machines (SVM), Decision Tree using R2D3 and Neural Networks for classification. We used the Tensorflow Playground to help understand the Neural Network and Deep Learning concepts. I gave an analogy of how Machine Learning process is like making a smoothie where your algorithm is a recipe, your data are your ingredients, your computer is your blender and your smoothie is the model that you developed. I used the same example to convey the concept of Training Validation and Testing. Coverage of Type 1 and Type 2 errors together with the metrics of Recall and Precision was covered as well. Finally I closed the session with what are some good resources to get started with Machine Learning for all skill levels. There are references to websites, courses, kaggle competition, podcasts, cheat sheets and books.

Machine Learning 101 from Setu Chokshi

]]> 2684 5 https://cdn.slidesharecdn.com/ss_thumbnails/ctumachinelearning101-170331011251-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Time series predictions using LSTMs /slideshow/time-series-predictions-using-lstms/73922883 timeseriespredictionsusinglstm-170330024535
This was a presentation done for the Techspace of IoT Asia 2017 oon 30th March 2017. This is an introductory session to introduce the concept of Long Short-Term Memory (LSTMs) for the prediction in Time Series. I also shared the Keras code to work out a simple Sin Wave example and a Household power consumption data to use for the predictions. The links for the code can be found in the presentation. ]]>
This was a presentation done for the Techspace of IoT Asia 2017 oon 30th March 2017. This is an introductory session to introduce the concept of Long Short-Term Memory (LSTMs) for the prediction in Time Series. I also shared the Keras code to work out a simple Sin Wave example and a Household power consumption data to use for the predictions. The links for the code can be found in the presentation. ]]> Thu, 30 Mar 2017 02:45:35 GMT /slideshow/time-series-predictions-using-lstms/73922883 SetuChokshi@slideshare.net(SetuChokshi) Time series predictions using LSTMs SetuChokshi This was a presentation done for the Techspace of IoT Asia 2017 oon 30th March 2017. This is an introductory session to introduce the concept of Long Short-Term Memory (LSTMs) for the prediction in Time Series. I also shared the Keras code to work out a simple Sin Wave example and a Household power consumption data to use for the predictions. The links for the code can be found in the presentation. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/timeseriespredictionsusinglstm-170330024535-thumbnail.jpg?width=120&height=120&fit=bounds" /> This was a presentation done for the Techspace of IoT Asia 2017 oon 30th March 2017. This is an introductory session to introduce the concept of Long Short-Term Memory (LSTMs) for the prediction in Time Series. I also shared the Keras code to work out a simple Sin Wave example and a Household power consumption data to use for the predictions. The links for the code can be found in the presentation.

Time series predictions using LSTMs from Setu Chokshi

]]> 2356 4 https://cdn.slidesharecdn.com/ss_thumbnails/timeseriespredictionsusinglstm-170330024535-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 https://cdn.slidesharecdn.com/profile-photo-SetuChokshi-48x48.jpg?cb=1712134160 Hello, my name is Setu Chokshi WHO I AM: I’m a senior technical leader, innovator and specialist in machine learning and artificial intelligence. I'm also a leader who has gained the respect of my team through active listening and delegating tasks aligned to talents. MY BACKGROUND has occurred organically as technical triumphs have led to greater opportunities. I’ve been fortunate to have worked with industry behemoths—General Electric and Nielsen. PROUD MOMENTS: ✔ Participated in a 1300-question survey for 25,000 people—one of the largest ever completed. ✔ Patents: an image quality recognition methodology that out-performed state-of-the-art methodologies; a process for reducing bi... http://books.google.com/books?id=EtiyACS3Z8EC&pg=PA858&lpg=PA858&dq=Setu+Chokshi&source=bl&ots=FyfB4Szn2E&sig=ToIspOHrsxM31IB206g_bGzoE7c&hl=en&ei=mSOoSbWkKcyJngef7YjWDw&sa=X&oi=book_result&resnum=11&ct=result https://cdn.slidesharecdn.com/ss_thumbnails/buildvsbuy-190119014021-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/build-vs-buy-ensuring-maximum-roi-from-ai/128441105 Build vs Buy: Ensuring... https://cdn.slidesharecdn.com/ss_thumbnails/aiforaifinal-190119014019-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/ai-for-ai-building-state-of-the-art-models/128441103 AI for AI: Building st... https://cdn.slidesharecdn.com/ss_thumbnails/globalaibootcampautomatedml-181216114355-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/microsoft-introduction-to-automated-machine-learning/126008425 Microsoft Introduction...