Archive

2023  1

July  1

[Notes] MaxViT: Multi-Axis Vision Transformer

July 16, 2023 · Ceshine Lee

2022  4

May  1

[Notes] PolyLoss: A Polynomial Expansion Perspective of Classification Loss Functions

May 15, 2022 · Ceshine Lee

March  1

[Notes] Understanding Visual Attention Network

March 14, 2022 · Ceshine Lee

January  2

[Notes] Understanding ConvNeXt

January 28, 2022 · Ceshine Lee

Use MPIRE to Parallelize PostgreSQL Queries

January 7, 2022 · Ceshine Lee

2021  12

August  1

[Notes] (Ir)Reproducible Machine Learning: A Case Study

August 27, 2021 · Ceshine Lee

July  2

[Notes] Understanding XCiT - Part 2

July 25, 2021 · Ceshine Lee

[Notes] Understanding XCiT - Part 1

July 24, 2021 · Ceshine Lee

June  1

How to Create a Documentation Website for Your Python Package

June 13, 2021 · Ceshine Lee

May  1

Text Analysis using Julia

May 1, 2021 · Ceshine Lee

April  2

How to Reduce the Loading Time of Julia Scripts

April 18, 2021 · Ceshine Lee

[Notes] Gradient Checkpointing with BERT

April 4, 2021 · Ceshine Lee

March  2

[Paper] Adafactor: Adaptive Learning Rates with Sublinear Memory Cost

March 18, 2021 · Ceshine Lee

Mistake I Made that Crippled My Streamlit App

March 14, 2021 · Ceshine Lee

February  2

A Case Study of fastcore @patch_to

February 19, 2021 · Ceshine Lee

[Paper] Rethinking Cooperative Rationalization: Introspective Extraction and Complement Control

February 14, 2021 · Ceshine Lee

January  1

Reducing the SentencePiece Vocabulary Size of Pretrained NLP Models

January 18, 2021 · Ceshine Lee

2020  22

December  4

[Kaggle] Google Research Football 2020

December 28, 2020 · Ceshine Lee

[PyTorch Lightning] Log Training Losses when Accumulating Gradients

December 22, 2020 · Ceshine Lee

Generating Synthetic Tabular Data Using GAN

December 14, 2020 · Ceshine Lee

[Paper] Are We Really Making Much Progress?

December 4, 2020 · Ceshine Lee

November  2

Weird Behavior in the FiveThirtyEight 2020 Election Model

November 23, 2020 · Ceshine Lee

The Book Of Why: The New Science of Cause and Effect

November 14, 2020 · Ceshine Lee

October  3

Automatic Testing Your SQLite Database with Great Expectations

October 17, 2020 · Ceshine Lee

[Tensorflow] Training CV Models on TPU without Using Cloud Storage

October 11, 2020 · Ceshine Lee

Replicate Conda Environment in Docker

October 7, 2020 · Ceshine Lee

September  1

[Paper] Please Stop Permuting Features

September 8, 2020 · Ceshine Lee

August  2

[Paper] Language-agnostic BERT Sentence Embedding

August 19, 2020 · Ceshine Lee

[Competition] Jigsaw Multilingual Toxic Comment Classification

August 5, 2020 · Ceshine Lee

July  3

[Paper] Training Question Answering Models From Synthetic Data

July 23, 2020 · Ceshine Lee

[Tip] TorchScript Supports Half Precision

July 11, 2020 · Ceshine Lee

Self-Supervised Domain Adaptation

July 6, 2020 · Ceshine Lee

June  2

Using Julia to Do Whole Word Masking

June 28, 2020 · Ceshine Lee

[Failure Report] Distill Fine-tuned Transformers into Recurrent Neural Networks

June 16, 2020 · Ceshine Lee

May  1

Deploying EfficientNet Model using TorchServe

May 4, 2020 · Ceshine Lee

April  2

Tensorflow Profiler with Custom Training Loop

April 24, 2020 · Ceshine Lee

Monitor Python Script Cron Jobs using Telegram

April 10, 2020 · Ceshine Lee

March  1

Clutter-free Interactive Charts in R using Plotly

March 31, 2020 · Ceshine Lee

February  1

TensorFlow 2.1 with TPU in Practice

February 13, 2020 · Ceshine Lee

2019  23

December  2

Create a Customized Text Annotation Tool in Two Days - Part 2

December 17, 2019 · Ceshine Lee

Create a Customized Text Annotation Tool in Two Days - Part 1

December 16, 2019 · Ceshine Lee

November  2

Fine-tuning BERT for Similarity Search

November 28, 2019 · Ceshine Lee

[Notes] “Statistical Inference Enables Bad Science; Statistical Thinking Enables Good Science”

November 9, 2019 · Ceshine Lee

October  1

Pro Tip: Use Shutdown Script Detect Preemption on GCP

October 25, 2019 · Ceshine Lee

September  1

Zero Shot Cross-Lingual Transfer with Multilingual BERT

September 24, 2019 · Ceshine Lee

August  3

More Memory-Efficient Swish Activation Function

August 22, 2019 · Ceshine Lee

Customizing Spacy Sentence Segmentation

August 14, 2019 · Ceshine Lee

[Notes] Jigsaw Unintended Bias in Toxicity Classification

August 4, 2019 · Ceshine Lee

July  1

[Notes] iMet Collection 2019 - FGVC6 (Part 1)

July 16, 2019 · Ceshine Lee

June  3

Dealing with Synthetic Data

June 25, 2019 · Ceshine Lee

Smaller Docker Image using Multi-Stage Build

June 21, 2019 · Ceshine Lee

Mixed Precision Training on Tesla T4 and P100

June 13, 2019 · Ceshine Lee

May  1

[Notes] SHAP Values

May 23, 2019 · Ceshine Lee

April  3

Detecting Chinese Characters in Unicode Strings

April 24, 2019 · Ceshine Lee

A First Look at Plotly Express

April 9, 2019 · Ceshine Lee

Custom Image Augmentation with Keras

April 4, 2019 · Ceshine Lee

March  2

UMAP on RAPIDS (15x Speedup)

March 30, 2019 · Ceshine Lee

Use NVIDIA Apex for Easy Mixed Precision Training in PyTorch

March 26, 2019 · Ceshine Lee

February  2

Multilingual Similarity Search Using Pretrained Bidirectional LSTM Encoder

February 15, 2019 · Ceshine Lee

News Topic Similarity Measure using Pretrained BERT Model

February 10, 2019 · Ceshine Lee

January  2

Playing with rstudio/gt R Package

January 22, 2019 · Ceshine Lee

More Portable, Reproducible R Development Environment

January 3, 2019 · Ceshine Lee

2018  10

December  1

Use TextRank to Extract Most Important Sentences in Article

December 7, 2018 · Ceshine Lee

November  2

Implementing Beam Search - Part 2

November 7, 2018 · Ceshine Lee

Implementing Beam Search - Part 1

November 5, 2018 · Ceshine Lee

October  1

[Notes] Neural Language Models with PyTorch

October 13, 2018 · Ceshine Lee

September  1

Use Visual Studio Code To Develop Python Programs

September 25, 2018 · Ceshine Lee

August  1

Prepare Deep-Learning-Ready VMs on Google Cloud Platform

August 10, 2018 · Ceshine Lee

July  2

Quantile Regression — Part 2

July 16, 2018 · Ceshine Lee

Quantile Regression — Part 1

July 12, 2018 · Ceshine Lee

March  1

[Review] Kaggle Toxic Comment Classification Challenge

March 24, 2018 · Ceshine Lee

February  1

Analyzing Tweets with R

February 27, 2018 · Ceshine Lee

2017  4

October  1

Feature Importance Measures for Tree Models — Part I

October 28, 2017 · Ceshine Lee

July  3

[Learning Note] Single Shot MultiBox Detector with Pytorch — Part 3

July 27, 2017 · Ceshine Lee

[Learning Note] Single Shot MultiBox Detector with Pytorch — Part 2

July 26, 2017 · Ceshine Lee

[Learning Note] Single Shot MultiBox Detector with PyTorch — Part 1

July 24, 2017 · Ceshine Lee

2016  1

July  1

Beijing PM2.5 Concentration History 2015-2016

July 13, 2016 · Ceshine Lee

2015  9

August  2

First Step of Web Scraping in Go

August 29, 2015 · Ceshine Lee

Plotly Example: Deaths Caused By Cancer in Taiwan

August 13, 2015 · Ceshine Lee

July  5

Random Sampling Data with Header

July 31, 2015 · Ceshine Lee

Migrated the Blog from Pelican to Hugo

July 28, 2015 · Ceshine Lee

Bayesian Logistic Regression using PyMC3

July 11, 2015 · Ceshine Lee

Change Sources of Ubuntu in a Docker image

July 7, 2015 · Ceshine Lee

Docker: Remove All Untagged Images

July 5, 2015 · Ceshine Lee

January  2

Random Sampling at the Command Line

January 22, 2015 · Ceshine Lee

Implement FTRL-Proximal Algorithm in Go - Part 2

January 2, 2015 · Ceshine Lee

2014  5

December  1

Implement FTRL-Proximal Algorithm in Go - Part 1

December 9, 2014 · Ceshine Lee

November  1

The Power of PyPy

November 29, 2014 · Ceshine Lee

April  1

Tip for using iPython Notebooks in virtualenv

April 29, 2014 · Ceshine Lee

March  1

A simple script to automate MySQLdump backups

March 5, 2014 · Ceshine Lee

February  1

Shortcuts for some common statistical functions

February 25, 2014 · Ceshine Lee

2013  1

October  1

Dicussing the zen of python

October 7, 2013 · Ceshine Lee