René's URL Explorer Experiment


Title: 李理的博客

direct link

Domain: fancyerii.github.io

NoneIE=edge
google-site-verificationxBT4GhYoi5qRD5tr338pgPM5OWHHIDR6mNg1a3euekI

Links:

李理的博客https://fancyerii.github.io/
Homehttps://fancyerii.github.io/
Abouthttps://fancyerii.github.io/about/
Tagshttps://fancyerii.github.io/tags/
深度学习理论与实战：提高篇 国内首本免费的深度学习书籍！涵盖听觉、视觉、语言和强化学习四大领域，深入浅出的理论分析和详尽的代码分析。(5/18增加用Cycle GAN实现Image to Image Tanslation；6/5增加机器翻译；6/17增加Policy Gradient)。 全书更新完毕！ 转载请联系作者(fancyerii at gmail dot com)！ https://fancyerii.github.io/2019/03/14/dl-book/
用Cycle GAN实现Image to Image Tanslationhttps://fancyerii.github.io/books/cycle-gan
机器翻译https://fancyerii.github.io/books/mt
Policy Gradienthttps://fancyerii.github.io/books/pg
《C++ High Performance: Master the art of optimizing the functioning of your C++ code , Second Edition》学习 https://fancyerii.github.io/2025/12/30/chp/
Implementing and Optimizing a BPE Tokenizer from Scratch—Part 11: Wrapping C++ Code with Cython This series of articles implements a subtask of Stanford’s CS336 Assignment 1: building an efficient training algorithm for a BPE Tokenizer. Through a series of optimizations, our algorithm’s training time on OpenWebText was reduced from over 10 hours to less than 10 minutes. This series explains these optimizations, including algorithmic improvements, data structure enhancements, parallelization with OpenMP, Cython optimization, and implementing key code in C++ along with its integration via Cython. This is the twelfth and final article in a series on using Cython to wrap the previous C++ code into an extension module for Python. https://fancyerii.github.io/2025/09/25/bpe-trainer-11_en/
动手实现和优化BPE Tokenizer的训练——第11部分：使用cython封装c++代码 本系列文章完成Stanford CS336作业1的一个子任务——实现BPE Tokenizer的高效训练算法。通过一系列优化，我们的算法在OpenWebText上的训练时间从最初的10多个小时优化到小于10分钟。本系列文章解释这一系列优化过程，包括：算法的优化，数据结构的优化，并行(openmp)优化，cython优化，用c++实现关键代码和c++库的cython集成等内容。本文是第十二篇，也是最后一篇，使用cython把之前的c++代码封装成扩展模块供python调用。 https://fancyerii.github.io/2025/09/25/bpe-trainer-11/
Implementing and Optimizing a BPE Tokenizer from Scratch—Part 10: Using Cython and PyPy for Acceleration This series of articles implements a subtask of Stanford’s CS336 Assignment 1: building an efficient training algorithm for a BPE Tokenizer. Through a series of optimizations, our algorithm’s training time on OpenWebText was reduced from over 10 hours to less than 10 minutes. This series explains these optimizations, including algorithmic improvements, data structure enhancements, parallelization with OpenMP, Cython optimization, and implementing key code in C++ along with its integration via Cython. This article, the eleventh in the series, will cover using Cython and PyPy to accelerate Python code. https://fancyerii.github.io/2025/09/24/bpe-trainer-10_en/
动手实现和优化BPE Tokenizer的训练——第10部分：使用cython和pypy来加速 本系列文章完成Stanford CS336作业1的一个子任务——实现BPE Tokenizer的高效训练算法。通过一系列优化，我们的算法在OpenWebText上的训练时间从最初的10多个小时优化到小于10分钟。本系列文章解释这一系列优化过程，包括：算法的优化，数据结构的优化，并行(openmp)优化，cython优化，用c++实现关键代码和c++库的cython集成等内容。本文是第十一篇，使用cython和pypy来加速python代码。 https://fancyerii.github.io/2025/09/24/bpe-trainer-10/
Implementing and Optimizing a BPE Tokenizer from Scratch—Part 9: Using a Heap to Find the Maximum Pair This series of articles implements a subtask of Stanford’s CS336 Assignment 1: building an efficient training algorithm for a BPE Tokenizer. Through a series of optimizations, our algorithm’s training time on OpenWebText was reduced from over 10 hours to less than 10 minutes. This series explains these optimizations, including algorithmic improvements, data structure enhancements, parallelization with OpenMP, Cython optimization, and implementing key code in C++ along with its integration via Cython. This is the tenth article, where we use the heap data structure to replace the process of finding the maximum pair, thereby improving performance. https://fancyerii.github.io/2025/09/21/bpe-trainer-9_en/
动手实现和优化BPE Tokenizer的训练——第9部分：使用堆来寻找最大pair 本系列文章完成Stanford CS336作业1的一个子任务——实现BPE Tokenizer的高效训练算法。通过一系列优化，我们的算法在OpenWebText上的训练时间从最初的10多个小时优化到小于10分钟。本系列文章解释这一系列优化过程，包括：算法的优化，数据结构的优化，并行(openmp)优化，cython优化，用c++实现关键代码和c++库的cython集成等内容。本文是第十篇，使用堆(heap)这个数据结构来替代求最大pair，提升性能。 https://fancyerii.github.io/2025/09/21/bpe-trainer-9/
Implementing and Optimizing a BPE Tokenizer from Scratch—Part 8: Implementing Fine-Grained Updates This series of articles implements a subtask of Stanford’s CS336 Assignment 1: building an efficient training algorithm for a BPE Tokenizer. Through a series of optimizations, our algorithm’s training time on OpenWebText was reduced from over 10 hours to less than 10 minutes. This series explains these optimizations, including algorithmic improvements, data structure enhancements, parallelization with OpenMP, Cython optimization, and implementing key code in C++ along with its integration via Cython. This is the ninth article, which focuses on optimizing the update process for data structures like pair_counts. https://fancyerii.github.io/2025/09/19/bpe-trainer-8_en/
动手实现和优化BPE Tokenizer的训练——第8部分：实现细粒度更新 本系列文章完成Stanford CS336作业1的一个子任务——实现BPE Tokenizer的高效训练算法。通过一系列优化，我们的算法在OpenWebText上的训练时间从最初的10多个小时优化到小于10分钟。本系列文章解释这一系列优化过程，包括：算法的优化，数据结构的优化，并行(openmp)优化，cython优化，用c++实现关键代码和c++库的cython集成等内容。本文是第九篇，优化pair_counts等数据结构的更新过程。 https://fancyerii.github.io/2025/09/19/bpe-trainer-8/
Implementing and Optimizing a BPE Tokenizer from Scratch—Part 7: Using Flat Hash Map instead of std::unordered_map This series of articles implements a subtask of Stanford’s CS336 Assignment 1: building an efficient training algorithm for a BPE Tokenizer. Through a series of optimizations, our algorithm’s training time on OpenWebText was reduced from over 10 hours to less than 10 minutes. This series explains these optimizations, including algorithmic improvements, data structure enhancements, parallelization with OpenMP, Cython optimization, and implementing key code in C++ along with its integration via Cython. This is the eighth article in the series, focusing on using a flat hash map to replace the C++ standard library’s std::unordered_map for improved performance. https://fancyerii.github.io/2025/09/18/bpe-trainer-7_en/
Older Posts →https://fancyerii.github.io/page2
FEATURED TAGShttps://fancyerii.github.io/tags/
人工智能 https://fancyerii.github.io/tags/#人工智能
深度学习 https://fancyerii.github.io/tags/#深度学习
chatbot https://fancyerii.github.io/tags/#chatbot
PyTorch https://fancyerii.github.io/tags/#PyTorch
Java https://fancyerii.github.io/tags/#Java
BERT https://fancyerii.github.io/tags/#BERT
git https://fancyerii.github.io/tags/#git
编程 https://fancyerii.github.io/tags/#编程
OCR https://fancyerii.github.io/tags/#OCR
汪曾祺 https://fancyerii.github.io/tags/#汪曾祺
语音识别 https://fancyerii.github.io/tags/#语音识别
Kaldi https://fancyerii.github.io/tags/#Kaldi
Linux https://fancyerii.github.io/tags/#Linux
XLNet https://fancyerii.github.io/tags/#XLNet
情感分析 https://fancyerii.github.io/tags/#情感分析
sentiment analysis https://fancyerii.github.io/tags/#sentiment analysis
语法纠错 https://fancyerii.github.io/tags/#语法纠错
Transformer https://fancyerii.github.io/tags/#Transformer
Tensorflow https://fancyerii.github.io/tags/#Tensorflow
Huggingface https://fancyerii.github.io/tags/#Huggingface
Ubuntu https://fancyerii.github.io/tags/#Ubuntu
TensorFlow https://fancyerii.github.io/tags/#TensorFlow
深度学习框架 https://fancyerii.github.io/tags/#深度学习框架
Tensor2Tensor https://fancyerii.github.io/tags/#Tensor2Tensor
机器翻译 https://fancyerii.github.io/tags/#机器翻译
微信 https://fancyerii.github.io/tags/#微信
wechat https://fancyerii.github.io/tags/#wechat
automation https://fancyerii.github.io/tags/#automation
selenium https://fancyerii.github.io/tags/#selenium
webdriver https://fancyerii.github.io/tags/#webdriver
pywinauto https://fancyerii.github.io/tags/#pywinauto
CentOS https://fancyerii.github.io/tags/#CentOS
GPU https://fancyerii.github.io/tags/#GPU
Appium https://fancyerii.github.io/tags/#Appium
t2t https://fancyerii.github.io/tags/#t2t
代码阅读 https://fancyerii.github.io/tags/#代码阅读
中英翻译 https://fancyerii.github.io/tags/#中英翻译
公众号 https://fancyerii.github.io/tags/#公众号
爬虫 https://fancyerii.github.io/tags/#爬虫
ocr https://fancyerii.github.io/tags/#ocr
tesseract https://fancyerii.github.io/tags/#tesseract
pytesseract https://fancyerii.github.io/tags/#pytesseract
python https://fancyerii.github.io/tags/#python
默认参数 https://fancyerii.github.io/tags/#默认参数
位置参数 https://fancyerii.github.io/tags/#位置参数
VPN https://fancyerii.github.io/tags/#VPN
JSON https://fancyerii.github.io/tags/#JSON
Jackson https://fancyerii.github.io/tags/#Jackson
huggingface https://fancyerii.github.io/tags/#huggingface
RoPE https://fancyerii.github.io/tags/#RoPE
PagedAttention https://fancyerii.github.io/tags/#PagedAttention
vLLM https://fancyerii.github.io/tags/#vLLM
Pre-training https://fancyerii.github.io/tags/#Pre-training
LLM https://fancyerii.github.io/tags/#LLM
CPT https://fancyerii.github.io/tags/#CPT
weather https://fancyerii.github.io/tags/#weather
forecasting https://fancyerii.github.io/tags/#forecasting
graph neural networks https://fancyerii.github.io/tags/#graph neural networks
qlora https://fancyerii.github.io/tags/#qlora
quantization https://fancyerii.github.io/tags/#quantization
transformers https://fancyerii.github.io/tags/#transformers
cmake https://fancyerii.github.io/tags/#cmake
pip https://fancyerii.github.io/tags/#pip
pipenv https://fancyerii.github.io/tags/#pipenv
conda https://fancyerii.github.io/tags/#conda
padding https://fancyerii.github.io/tags/#padding
vscode https://fancyerii.github.io/tags/#vscode
debug https://fancyerii.github.io/tags/#debug
source code https://fancyerii.github.io/tags/#source code
build https://fancyerii.github.io/tags/#build
deep learning https://fancyerii.github.io/tags/#deep learning
Speech https://fancyerii.github.io/tags/#Speech
ASR https://fancyerii.github.io/tags/#ASR
linux https://fancyerii.github.io/tags/#linux
pytorch https://fancyerii.github.io/tags/#pytorch
c++ https://fancyerii.github.io/tags/#c++
extension https://fancyerii.github.io/tags/#extension
Deep Learning https://fancyerii.github.io/tags/#Deep Learning
DeepSeek https://fancyerii.github.io/tags/#DeepSeek
Attention https://fancyerii.github.io/tags/#Attention
MoE https://fancyerii.github.io/tags/#MoE
cs336 https://fancyerii.github.io/tags/#cs336
bpe tokenizer https://fancyerii.github.io/tags/#bpe tokenizer
ABOUT MEhttps://fancyerii.github.io/about/
https://fancyerii.github.io/feed.xml
https://github.com/fancyerii
Li Lihttp://fancyerii.github.io
https://fancyerii.github.io/feed.xml
https://github.com/fancyerii
https://fancyerii.github.io/site.txt

Viewport: width=device-width, initial-scale=1

URLs of crawlers that visited me.