Imej profil aymar73
@aymar73
Penanda Canada Montreal, Canada
Ahli semenjak 4 Mac 2018
0 Cadangan

aymar73

Dalam Talian Luar Talian
I am a Machine learning engineer with deep expertise in Statistics and Python programming. I analyse structured and unstructured data and I am on my ease performing supervised, and unsupervised learning on images, text and so on... Depending on the size of data, I can perform cloud computing (AWS). I often perform my job under Jupyter Notebook where it is easy to code, visualize and document my projects. My projects are available under request.
$47 USD/hr
1 ulasan
1.4
  • 100%Kerja Diselesai
  • 100%Mengikut Bajet
  • 100%Tepat Pada Masa
  • N/AKadar Ulang Upah

Portfolio

Ulasan Terkini

  • Imej Veljko B. Project 16472431 has been deleted €45.00 EUR

    “It was a pleasure to work with Aymar Thierry. He is very professional and knowledgeable and I would definitely recommend to everyone to work with him. He is always at available, provides quick feedback, very dedicated to resolve the problem and his work is of a high quality.”

Pengalaman

Senior Credit risk Analyst

Jan 2016

•Predicting trends in regard to key risk issues such as delinquency, downgrading to impairment, provision on credit loss with SAS EM, SAS EG, Decision Trees and Logistic Regression •Classifying customers based on their rate of default using decision trees via SAS EM and SAS EG •Quantifying the statistical performance of commercial credit risk indicators proposed by external vendor such as EQUIFAX

BI Analyst

Jan 2014 - Jan 2016 (2 years)

•Generated and compiled data from multiple data sources using various business intelligence tools and information management systems •Performed calculations and analysis to fulfill requirements including but not limited to performance reporting - Worked with Business Analysts to refine understanding of project requirements •Delivered scalable data processing and analytics services.

Statistical Analyst

Oct 2008 - Jul 2013 (4 years)

•Wrangled, extracted, transformed, and loaded data from various databases, formats, and data sources •Used exploratory data analysis techniques to identify meaningful relationships, patterns, or trends from complex data sets. •Troubleshooted issues with data import and export routines and with the resulting datasets. •Designed and developped code to implement automated approaches for data handling and quality review tasks.

Kelayakkan

Machine Learning Engineering (2018)

Udacity

Becoming a machine learning engineer and applying predictive models to massive data sets in fields like education, finance, healthcare or robotics. Projects covered supervised, unsupervised, reinforcement learning, and some of the algorithms I used in my projetcs was SVM, random forest, Adaboost, boosting. The Tools used were python, tensorflow, keras, Ipython Notebook, and also cloud computing under amazon web service.

Deep Learning (2018)

Udacity

Learning the foundation topics in the field of deep learning to solve complex problems. I have developed Neural Networks projects using tensorflow and keras, Deep Convolutional Neural nets using Keras under AWS, Recurrent Neural nets, LSTM projects and I am on the verge to finish my first Generative Adversarial Nets project

Introduction to hadoop and Mapreduce: How to process big data (2014)

Udacity

Learnt the fundamental of MapReduce and Apache Hadoop to start making sense of big data in real world. Courses included for example Mapreduce design pattern (filtering pattern, inverted index, combiners) and codes are executed in python.

Natural Language Processing (2019)

Udacity

Penerbitan

https://review-api.udacity.com/api/v1/submissions/1000748/archive

Used a deep convolutional neural networks with 4 hidden layers, 2 fully connected dense layers each with a ReLU activation function, dropout and maxpooling with stride layers, Adam optimization with a 0.001 learning rate and a 0% decay rate with EarlyStopping, to perform a 98.6% in accuracy rate from the training data and a 93.9% accuracy rate from the validation data in building an algorithm that automatically identifies if a remotely sensed target is a ship or iceberg.

Pengesahan

  • Facebook Disambung
  • Freelancer Keutamaan
  • Bayaran Disahkan
  • Telefon Disahkan
  • Identiti Disahkan
  • E-mel Disahkan

Kemahiran Teratas Saya

Cari Freelancer yang serupa