Nikita Schneider

Published in Geek Culture

·Pinned

A Simple Guide To Reinforcement Learning With The Super Mario Bros. Environment

Theory Let’s say we want to design an algorithm that will be able to complete the first level of the Super Mario Bros. game. How will we do that? The goal of this game is quite simple — get the flag at the rightmost end of the level as fast as…

Deep Learning

10 min read

A Simple Guide To Reinforcement Learning With The Super Mario Bros. Environment

Published in Geek Culture

·Aug 8, 2021

A Simple Guide To Java Native Interface (JNI) Using Native Maven Plugin

Sometimes it’s necessary to call C/C++ code from Java, e.g., for GPGPU calculations using CUDA/OpenCL or to render complex 3D graphics using Vulkan/OpenGL. In such cases, the easiest way to invoke the C++ functions is to use Java Native Interface (JNI). …

Programming

5 min read

A Simple Guide To Java Native Interface (JNI) Using Native Maven Plugin

Published in Geek Culture

·Apr 4, 2021

Fine-Tune EleutherAI GPT-Neo to Generate Netflix Movie Descriptions in Only 47 Lines of Code

Recently, EleutherAI released their GPT-3-like model GPT-Neo, and a few days ago, it was released as a part of the Hugging Face framework. At the time of writing, this model is available only at the master branch of the transformers repository, so you need to install it like this: pip…

Deep Learning

4 min read

Fine-Tune EleutherAI GPT-Neo to Generate Netflix Movie Descriptions in Only 47 Lines of Code

Published in Geek Culture

·Mar 26, 2021

Speech Recognition Using CRNN, CTC Loss, DeepSpeech Beam Search Decoder, and KenLM Scorer

Theory Today, three of the most popular end-to-end ASR (Automatic Speech Recognition) models are Jasper, Wave2Letter+, and Deep Speech 2. Now they are available as a part of the OpenSeq2Seq toolkit made by Nvidia. …

Deep Learning

5 min read

Speech Recognition Using CRNN, CTC Loss, Beam Search Decoder and KenLM Scorer

Published in The Startup

·Dec 16, 2020

Multi-Digit Sequence Recognition With CRNN and CTC Loss Using PyTorch Framework

Theory An Optical Character Recognition (OCR) task is quite an old problem dated back to the 1970s when the first omni-font OCR technology has been developed. The complexity of this task comes from many natural features of texts: In some alphabets (Arabic, for example, especially in the cursive form) letters are…

Deep Learning

9 min read

Multi-Digit Sequence Recognition With CRNN and CTC Loss Using PyTorch Framework

Published in The Startup

·Nov 9, 2020

Number Hand Gestures Recognition Using TensorFlow.js

Theory As the demand for data-driven products grows, the data science community has been rapidly developing solutions that allow us to create and apply all the recent revolutionary advances in artificial intelligence across multiple platforms. In the early years of the so-called AI era, it was very common to have a…

Data Science

7 min read

Number Hand Gestures Recognition Using TensorFlow.js

Published in The Startup

·Oct 31, 2020

Fine Tuning BERT for Text Classification and Question Answering Using TensorFlow Framework

Theory Google BERT (Bidirectional Encoder Representations from Transformers) and other transformer-based models further improved the state of the art on eleven natural language processing tasks under broad categories of single text classification (e.g., sentiment analysis), text pair classification (e.g., natural language inference), question answering (like SQuAD 1.1) and text tagging (e.g…

Deep Learning

7 min read

Fine Tuning BERT for Text Classification and Question Answering Using TensorFlow Framework

Published in The Startup

·Jun 19, 2020

How To Design Seq2Seq Chatbot Using Keras Framework

Some theory Well, let’s assume that we have a dialog corpus that could be divided into pairs of questions and answers. And we need some kind of neural network that will predict the next word based on the previous ones. We also know that the input of neural network must be a…

Data Science

9 min read

How To Design Seq2Seq Chatbot Using Keras Framework

Published in The Startup

·Jun 7, 2020

Weather Forecasting Using Multilayer Recurrent Neural Network

Some theory Well, actually, there are plenty of useful resources like this or this, that explained in details how GRU/LSTM architectures work, all the math behind them and so on, but I think that all explanations I’ve seen before are somewhat misleading. Many articles show pictures like these:

Data Sceince

7 min read

Weather Forecasting Using Multilayer Recurrent Neural Network

Published in The Startup

·Jun 3, 2020

Image Generation Using Keras Framework

Some theory One of the easiest ways to generate images of decent quality is to use Deep Convolutional Generative Adversarial Network (DCGAN) architecture, invented by Ian Goodfellow in 2014. This architecture consists of two networks — generator and discriminator. The generator network is able to take random noise and map it into…

Data Science

8 min read