Konstantin Lopuhin
lopuhin
Languages
Top Repositories
Kaggle DSTL Satellite Imagery Feature Detection
Transformer language model (GPT-2) with sentencepiece tokenizer
Kaggle Data Science Bowl 2018 dataset fixes
Kaggle script build system template
Kaggle Kuzushiji Recognition: 2nd place solution
Repositories
97Post-training with Tinker
Kaggle DSTL Satellite Imagery Feature Detection
Transformer language model (GPT-2) with sentencepiece tokenizer
Kaggle script build system template
Kaggle Kuzushiji Recognition: 2nd place solution
Kaggle Data Science Bowl 2018 dataset fixes
Various python emacs stuff
A Python library for connecting securely to your Cloud SQL instances.
2nd place solution for Kaggle NOAA Fisheries Steller Sea Lion Population Count
🤗 Transformers: State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.
Teaching a tiny LLM some basic math
Use pyppeteer from a Scrapy spider
AdaGram (adaptive skip-gram) for Python
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Planning work described by grantt diagram, according to given constraints
An integrated model of eye movement behaviour in sentence comprehension. The ACT-R parser of Lewis & Vasishth (2005), extended with eye movement control.
https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/
General purpose GPU compute framework for cross vendor graphics cards (AMD, Qualcomm, NVIDIA & friends). Blazing fast, mobile-enabled, asynchronous and optimized for advanced GPU data processing usecases. Backed by the Linux Foundation.
No description provided.
TensorRT-LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and build TensorRT engines that contain state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT-LLM also contains components to create Python and C++ runtimes that execute those TensorRT engines.
No description provided.
No description provided.
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
WIP
Topic Modelling for Humans
Sense grouping survey
No description provided.
Calculate the number of non-attacking kings placements of n*n kings on 2*n x 2*n board
No description provided.
Sense frequencies and WSD