The students were fortunate to attain yet another e-GL by incredibly talented Mr Saurabh Shahane on “Ensembles and Kaggle” on 6th March 2021. Mr Saurabh Shahane is a Data Scientist at CKM Sports Management Ltd, Vancouver, British Columbia, Canada. He is also working as a Deep Learning mentor at Technocolabs, Mumbai. He has recently become the Kaggle Notebook Master as he is the epitome of knowledge regarding Machine Learning, Deep learning, AI, Data visualization, Statistical Modelling and Data Analysis. During his early career, he had worked as a Technical Backend Support at MKCL and machine learning Intern at Algonauts Technologies. He is a certified AI engineer by IBM and also TensorFlow Developer by Deeplearning.ai. Mr Saurabh Shahane has done his honours in Deep Learning by National Research University Higher School of Economics, Moscow, Russia. His recent achievements are Kaggle Notebooks Master – (rank – 163 / 1,60,278) and Kaggle Dataset Master – (position – 28/33,882).
The speaker started the session by presenting the tools that emphasize the growth and development of everything in generic. He addressed the critical concept such as stacking, bagging, boosting etc., used in the Machine Learning algorithm to get effective accuracy by using the ensemble model. He further discussed how ensemble techniques are used to create more accurate solutions than any other methods. He has also addressed the students about the significance of base-learners models and super-learner or meta-learner, based on the prediction used in stacking and using those predictions, to derive new forecasts.
The speaker has also shared his insights on how to split the datasets into training and testing sets. He also showed the various techniques such as k-fold cross-validation, random forest to get the most productive predictions, which depends on the datasets. The speaker also encouraged the students to practice as much machine learning algorithm by using various techniques such as ensembles techniques to be familiar with the data environment to create the best solutions. The speaker also explained the features and categories of “Kaggle” and why it mattered. Kaggle is the largest data scientist community globally, with almost more than two million followers, mostly belongs to Data Scientist. It is a tool that can help us grow in data analytics.
Towards the end, Mr. Shahane concluded the session with Q&As where he had cleared all the students’ queries. Overall, it was indeed an interactive and insightful session.