Hi, my name is Aleksandr Drozd

protoNN: developing neural models made easy

Transparent and elastic scheduling of DNN training jobs on modern HPC systems.
Monitoring and visualizing model parameters and computational performance statistics.
Semi-automatic hyperparameter tuning/optimization and architecture search.
A user-defined interactive interface to drive the framework/ design process, not bound to any particular framework
Scaling the functionality and performance

and lot's of other tools for NLP applications with emphasis on reproducibility

some work on measuring performance of deep learning models

and some work on predicting the performance of various hardware platforms

CFDG-inspired cairo-based pythonic generative art tool

cf.py (Source)

	`from contextfree.contextfree import *`
	`@check_limits`
	`def branch():`
	`line(0, 1)`
	`with translate(0, 0.9):`
	`with scale(0.7 + rnd(0.3)):`
	`with rotate(-0.4 + rnd(0.5)):`
	`branch()`
	`with rotate(0.4 + rnd(0.5)):`
	`branch()`
	`init(canvas_size=(300, 300))`
	`with translate(0, -1):`
	`with scale(0.6):`
	`branch()`
	`display_ipython()`

I like taking pictures ^_^