пропозиції роботи
Łukasz Nalewajko
Computer vision, Image processing, Machine learning , Deep learning, Data science
Lublin,
lubelskie
Dołącz do GoldenLine i obejrzyj pełny profil tego użytkownika
Zobacz pełny profil
Umiejętności
C++
Digital Image Processing
Git
Matlab
Matplotlib
Microsoft Excel
Microsoft Word
OpenCV
Python (podstawy)
scikit.learn
Tensorflow
Seaborn
Keras
Języki
angielski
biegły
francuski
podstawowy
Doświadczenie zawodowe
Junior Python developer
Whiteaster sp. z o.o.
Warehouse operative
Extra Personnel Ltd
Tkmaxx warehouse operative, Hams Hall, UK.
- Manual handling
- Working in a team environment
- Working in a safe manner
- Processing paperwork - goods inwards, picking lists, full pallet lists, stock checks, customer related
paperwork
- Work within a physically demanding environment
- Operating of Mechanical Handling Equipment
- Manual handling
- Working in a team environment
- Working in a safe manner
- Processing paperwork - goods inwards, picking lists, full pallet lists, stock checks, customer related
paperwork
- Work within a physically demanding environment
- Operating of Mechanical Handling Equipment
Production worker
Extra Personnel Ltd
Plastic Omnium, Coleshill, UK. Land Rover plastic interior doors production. Manual deburring and inspection.
Purchasing Specialist
OR-SAT ZPCHr Sp. z o.o.
Automotive plant. Poland. Suppliers sourcing, assessment and nomination, complaints managing, cost breakdown analysis.
Sales Specialist
Automotive plant. Poland. Quality assurance, laboratory testing of raw
materials. Dimensions checking with caliper based on technical drawings.
materials. Dimensions checking with caliper based on technical drawings.
Projekty
Kaggle: Human Protein Atlas Image Classification
Classify subcellular protein patterns in human cells.
Classification, image data. Implementation Python(fastai). CNN model was created.
Model based on kernel:
https://www.kaggle.com/zhugds/resnet34-with-rgby-fast-ai-fork
Top 36% finish(788 out of 2236).
Classification, image data. Implementation Python(fastai). CNN model was created.
Model based on kernel:
https://www.kaggle.com/zhugds/resnet34-with-rgby-fast-ai-fork
Top 36% finish(788 out of 2236).
Kaggle: Airbus Ship Detection Challenge
Image data, object detection, object segmentation. Implementation Python(fastai)
CNN model was created. ResNet34 + U-Net architecture.
Model based on kernel https://www.kaggle.com/iafoss/unet34-dice-0-87
Top 42% finish(373 out of 894)
CNN model was created. ResNet34 + U-Net architecture.
Model based on kernel https://www.kaggle.com/iafoss/unet34-dice-0-87
Top 42% finish(373 out of 894)
Kaggle: TGS Salt Identification Challenge
Geology, image data. Image segmentation. Implementation Python(fastai)
CNN model was created. ResNet34 + U-Net architecture.
Model based on kernel https://www.kaggle.com/lukaszn/fork-of-fork-of-tgs-fastai-resnet34-unet-v1?scriptVersionId=6350847
Top 37% finish(1184 out of 3234)
CNN model was created. ResNet34 + U-Net architecture.
Model based on kernel https://www.kaggle.com/lukaszn/fork-of-fork-of-tgs-fastai-resnet34-unet-v1?scriptVersionId=6350847
Top 37% finish(1184 out of 3234)
Kaggle: Data Science Bowl 2018
Image segmentation. CNN model was created based on U-Net architecture. Implementation Python(Keras).
Solution based on kernel https://www.kaggle.com/sjoly123/kerasunet-281, with manual train/test split for data in the first stage of competition. Data augmentation added.
Top 16% finish (561 out of 3634).
Solution based on kernel https://www.kaggle.com/sjoly123/kerasunet-281, with manual train/test split for data in the first stage of competition. Data augmentation added.
Top 16% finish (561 out of 3634).
Kaggle: Porto Seguro’s Safe Driver Prediction
Tabular data, binary classification. Implementation Python(XGBoost).
XGBoost model was created.
Solution based on https://www.kaggle.com/sudosudoohio/stratified-kfold-xgboost-eda-tutorial-0-281
Top 49% finish(2509/5169)
XGBoost model was created.
Solution based on https://www.kaggle.com/sudosudoohio/stratified-kfold-xgboost-eda-tutorial-0-281
Top 49% finish(2509/5169)
Kaggle: Data Science Bowl 2017
Healthcare, image data, binary classification. 3D CT scans. Implementation Python(Keras).
3D CNN model was created with channels permutation.
Solution based on kernel https://www.kaggle.com/zfturbo/keras-vs-cancer.
Model not classified on leaderboard. Scored 0.67300 on private dataset, it is around 283 place out of 1972( Top 15%)
3D CNN model was created with channels permutation.
Solution based on kernel https://www.kaggle.com/zfturbo/keras-vs-cancer.
Model not classified on leaderboard. Scored 0.67300 on private dataset, it is around 283 place out of 1972( Top 15%)
Edukacja
Introduction to PySpark, średnie
DataCamp
Extreme Gradient Boosting with XGBoost, średnie
DataCamp
Data Types for Data Science Course, średnie
DataCamp
Introduction to Git for Data Science Course, średnie
DataCamp
Joining Data in PostgreSQL Course, średnie
DataCamp
Introduction to Data Visualization with Python Course, średnie
DataCamp
Importing Data in Python (Part 2) Course, średnie
DataCamp
Importing Data in Python (Part 1) Course, średnie
DataCamp
Cleaning Data in Python Course, średnie
DataCamp
Merging DataFrames with pandas Course, średnie
DataCamp
Manipulating DataFrames with pandas Course, średnie
DataCamp
pandas Foundations Course, średnie
DataCamp
Python Data Science Toolbox (Part 1) Course, średnie
DataCamp
Machine Learning with the Experts: School Budgets Course, średnie
DataCamp
Natural Language Processing Fundamentals in Python Course, średnie
DataCamp
Unsupervised Learning in Python Course, średnie
DataCamp
Statistical Thinking in Python (Part 2) Course, średnie
DataCamp
Statistical Thinking in Python (Part 1) Course, średnie
DataCamp
Deep Learning in Python Course, średnie
DataCamp
Supervised Learning with scikit-learn Course, średnie
DataCamp
Python Data Science Toolbox (Part 2) Course, średnie
DataCamp
Intermediate Python for Data Science Course, średnie
DataCamp
Intro to Python for Data Science Course, średnie
DataCamp
Intro to SQL for Data Science Course, średnie
DataCamp
Deep Learning, studia podyplomowe
Udacity.com on-line education
C++ Certified Professional Programmer, średnie
www.cppinstitute.org
C++ Certified Associate Programmer, średnie
www.cppinstitute.org
Fundamentals of Digital Image and Video Processing www.coursera.org , studia podyplomowe
Northwestern University
Introduction a la programmation orientee objet ( en C++ ) www.coursera.org , studia podyplomowe
Ecole Polytechnique Federale de Lausanne
6th International Workshop on Biomedical Image Registration, studia podyplomowe
University College London
Discrete Inference and Learning in Artificial Vision, studia podyplomowe
CentraleSupélec
C++ For C Programmers www.coursera.org, studia podyplomowe
University of California, Santa Cruz
Initiation a la programmation ( en C++ ) www.coursera.org, studia podyplomowe
Ecole Polytechnique Federale de Lausanne
2nd Biomedical Image Analysis Summer School : Modalities, Methodologies & Clinical Research, studia podyplomowe
The Center of Visual Computing of Ecole Centrale de Paris & Ecole des Ponts-ParisTech
Digital Signal Processing (www.coursera.com on-line), studia podyplomowe
Ecole Polytechnique Federale de Lausanne
Calculus: Single Variable (www.coursera.com on-line), studia podyplomowe
University of Pennsylvania(on-line)
Image and video processing: From Mars to Hollywood with a stop at the hospital (www.coursera.com on-line), studia podyplomowe
Duke University (on-line)
Neural Networks for Machine Learning(www.coursera.com on-line course), studia podyplomowe
University of Toronto (on-line)
Probabilistic Graphical Models(www.coursera.com on-line course), studia podyplomowe
Stanford University (on-line)
Machine Learning(www.coursera.org on-line course) , studia podyplomowe
Stanford University(on-line)
ST101: Introduction to Statistics: Making Decisions Based on Data , studia podyplomowe
Udacity.com on-line education
6.002x Circuits and Electronics on-line course, studia podyplomowe
Massachusetts Institute of Technology
Przetwarzanie i analiza obrazów biomedycznych, studia podyplomowe
Politechnika Łódzka
Fizyka Techniczna, magisterskie
Politechnika Wrocławska
Specjalizacje
IT - Rozwój oprogramowania
Programista Python
Zainteresowania
- algorytmy przetwarzania i analiza obrazów biomedycznych
- sport (siatkówka, koszykówka)
- taniec (salsa)
- sport (siatkówka, koszykówka)
- taniec (salsa)
Inne
International English Language Testing System (IELTS) - Academic.
Band score 7.0 (12.03.2016).
TRF no. 1510017607NALL294A.
Band score 7.0 (12.03.2016).
TRF no. 1510017607NALL294A.
