Sebastian Dziadzio

python-katas. Short exercises in Python.

ivr-synsets. Using Polish wordnet to build expanded synsets for an IVR system.

idsprites. Easily generate continual learning benchmarks.

disco. Disentangled Continual Learning

twitter-sentiment. Sentiment analysis of tweets

data-science-cheatsheet. Common patterns I use, mostly for data visualisation.

db2arff. Generates an Atribute-Relation File from data collected with the AWARE framework and stored in a MySQL database.

movable-type. Playing around with texts from Project Gutenberg.

collocations. Building dictionaries of verb-noun collocations for automatic speech recognition of Polish.

hate-tweet. Detecting hate speech in tweets using bag-of-trick models and bi-LSTM networks.

pocket-stats. An analysis of articles in my Pocket

mock-reduce. Practicing MapReduce programming model on small datasets using a mock MR Python library.

helm. Holistic Evaluation of Language Models (HELM), a framework to increase the transparency of language models (https://arxiv.org/abs/2211.09110). This framework is also used to evaluate text-to-image models in Holistic Evaluation of Text-to-Image Models (HEIM) (https://arxiv.org/abs/2311.04287).

dualprompt-pytorch. PyTorch Implementation of DualPrompt: Complementary Prompting for Rehearsal-free Continual Learning @ ECCV22

openai-api-playground. Playing with the OpenAI API

opencompass. OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.

wild-llamas. Merging finetuned language models in the wild.

l2p-pytorch. PyTorch Implementation of Learning to Prompt (L2P) for Continual Learning @ CVPR22

model-soups. Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time

seguidora. DoRA with a tracking oracle.

onebench. Sample-Level Benchmarking Over Open-Ended Capabilities

hvae. Exploring hierarchical VAEs for generative replay.

thesis-scripts. Miscellaneous scripts and utilities for my master's thesis.

python-pocket. A Python wrapper around the Pocket API.

pokemon-recognition. Using deep learning to recognise Pokemons

configs. My configs for vim, tmux, zsh etc.

appearance-flow. A deep learning framework for synthesizing novel views of objects and scenes

klaster_sim. Shell

cv. My résumé

homomorphic-estimation. Estimation of the variable map of noise out of SENSE magnetic resonance data using a homomorphic approach.

thesis. Master's Thesis

mammoth. An Extendible (General) Continual Learning Framework based on Pytorch - official codebase of Dark Experience for General Continual Learning

project-euler. Solution to problems from Project Euler.

pydata-warsaw-2017. Slides from PyDataWarsaw 2017

rolling-stone. 60 years of Bob's Songs

UnsupervisedGeometryAwareRepresentationLearning. Unsupervised Geometry-Aware Representation Learning for 3D Human Pose Estimation (ECCV 2018)

learning-deep-learning. Learning Keras and TensorFlow by building simple models

PyKinect2. Wrapper to expose Kinect for Windows v2 API in Python

image_search. Python Library to download images and metadata from popular search engines.

sebastiandziadzio.github.io. Personal website

nvae. An unofficial toy implementation for NVAE

clrs. Implementing algorithms from the Introduction to Algorithms (CLRS).

Multiview2Novelview. An official TensorFlow implementation of "Multi-view to Novel view: Synthesizing novel views with Self-Learned Confidence" (ECCV 2018) by Shao-Hua Sun, Minyoung Huh, Yuan-Hong Liao, Ning Zhang, and Joseph J. Lim

stl-workshop. Examples and exercises from an STL workshop.

rprogramming. Code from various R programming courses.

functional-programming. Haskell

nerfies. This is the code for Deformable Neural Radiance Fields, a.k.a. Nerfies.

telemetry-server. Server for the Mozilla Telemetry project

redi-python-intro. HTML

algorithm-design. Exercises from "Algorithms: Design and Analysis", an online course on Coursera.