Skip to content

Necati Demir

Data Science & Machine Learning Newsletter # 99

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group Ulti­mate Guide To Sta­tis­ti­cal Sig­nif­i­cance Tests in R Imple­ment and inter­pret the com­monly used sta­tis­ti­cal sig­nif­i­cance tests in R, the pur­pose, when to use and how to inter­pret the result. Word2Vec (skip-gram model) The skip-gram neural net­work model is actu­ally sur­pris­ingly sim­ple in its most basic form. Train a sim­ple neural net­work with a sin­gle hid­den layer to per­form Devamını Oku […]

Data Science & Machine Learning Newsletter # 98

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group CAN (Cre­ative Adver­sar­ial Net­work) - Explained GANs (Gen­er­a­tive Adver­sar­ial Net­works), a type of Deep Learn­ing net­works, have been very suc­cess­ful in cre­at­ing non-procedural con­tent. This work explores the pos­si­bil­ity of machine gen­er­ated cre­ative con­tent. 109 Com­monly Asked Data Sci­ence Inter­view Ques­tions For a data sci­ence inter­view, an inter­viewer will ask ques­tions Devamını Oku […]

Data Science & Machine Learning Newsletter # 97

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group How to Apply Machine Learn­ing to Event Pro­cess­ing How do you com­bine his­tor­i­cal Big Data with machine learn­ing for real-time ana­lyt­ics? An approach is out­lined with dif­fer­ent soft­ware ven­dors, busi­ness use cases, and best prac­tices. J.P.Morgan’s mas­sive guide to machine learn­ing and big data jobs in finance J.P. Morgan’s quan­ti­ta­tive invest­ing and deriv­a­tives strat­egy team, Devamını Oku […]

Data Science & Machine Learning Newsletter # 96

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group Python Plot­ting for Exploratory Analy­sis What method can be used to detect sea­son­al­ity in data? A really good way to find peri­od­ic­ity in any reg­u­lar series of data is to inspect its power spec­trum after remov­ing any over­all trend. Word2vec with tens of bil­lions of items, what could pos­si­bly go wrong? Project related to text sum­ma­riza­tion on Github: https://github.com/miso-belica/sumy https://github.com/abisee/pointer-generator https://github.com/LazoCoder/Article-Summarizer https://github.com/hengluchang/newsum https://github.com/sriniiyer/codenn https://github.com/davidadamojr/TextRank An Devamını Oku […]

Data Science & Machine Learning Newsletter # 95

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group 7 Tech­niques to Han­dle Imbal­anced Data This blog post intro­duces seven tech­niques that are com­monly applied in domains like intru­sion detec­tion or real-time bid­ding, because the datasets are often extremely imbal­anced. Bayesian Deep Learn­ing with Edward (and a trick using Dropout) Ele­gant N-gram Gen­er­a­tion in Python A quick few snip­pets of code – solv­ing how to Devamını Oku […]

Data Science & Machine Learning Newsletter # 94

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group Accu­rate, Large Mini­batch SGD: Train­ing Ima­geNet in 1 Hour Deep learn­ing thrives with large neural net­works and large datasets. How­ever, larger net­works and larger datasets result in longer train­ing times that impede research and devel­op­ment progress. Dis­trib­uted syn­chro­nous SGD offers a poten­tial solu­tion to this prob­lem by divid­ing SGD mini­batches over a pool of par­al­lel work­ers. … Time Devamını Oku […]

Data Science & Machine Learning Newsletter # 93

  You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group Fea­ture Engi­neer­ing: Data scientist’s Secret Sauce ! It is very tempt­ing for  data sci­ence prac­ti­tion­ers to opt for the best known  algo­rithms for a given problem.However It’s not the algo­rithm alone , which can pro­vide the best solu­tion  ; Model built on care­fully engi­neered and selected fea­tures can pro­vide far bet­ter results. pix2code: Gen­er­at­ing Code from a Graph­i­cal Devamını Oku […]

Data Science & Machine Learning Newsletter # 92

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group Bayesian machine learn­ing So you know the Bayes rule. How does it relate to machine learn­ing? It can be quite dif­fi­cult to grasp how the puz­zle pieces fit together — we know it took us a while. This arti­cle is an intro­duc­tion we wish we had back then. Machine Learn­ing Tech­niques for Pre­dic­tive Main­te­nance Pre­dic­tive main­te­nance pre­dicts fail­ure, and the actions could include Devamını Oku […]

Data Science & Machine Learning Newsletter # 91

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group Google releases dataset of 50M vec­tor draw­ings, open sources Sketch-RNN imple­men­ta­tion Blog Post Paper Github Repo Github Repo Google Offers Cloud-Based TPU Ser­vice for Train­ing and Deploy­ing Deep Learn­ing Mod­els We’re excited to announce that our second-generation Ten­sor Pro­cess­ing Units (TPUs) are com­ing to Google Cloud to accel­er­ate a wide range of machine learn­ing Devamını Oku […]

Data Science & Machine Learning Newsletter # 90

You want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group Sen­ti­ment analy­sis on Twit­ter using word2vec and keras word2vec is a group of Deep Learn­ing mod­els devel­oped by Google with the aim of cap­tur­ing the con­text of words while at the same time propos­ing a very effi­cient way of pre­pro­cess­ing raw text data. This post dis­c­cusses to do sen­ti­ment analy­ses with word2vec. Fully Con­vo­lu­tional Instance-aware Seman­tic Seg­men­ta­tion FCIS Devamını Oku […]