Published By
Created On
5 Nov 2021 14:40:17 UTC
Transaction ID
Cost
Safe for Work
Free
Yes
Anatomy of a Data Analytics MVP (PyCon APAC 2014)
Speaker: Ken Hu
Behind the glamour of mobile apps, business SaaS products have been steadily building itself into a billion dollar industry. With companies such as Salesforce spearheading the adaption of business SaaS products in corporate, startups are rising to take on the SME market.
With hot topics such as "big data" and "machine learning" flying around, it may seem overwhelming to put an idea into action. I will walk through the process of building an data analytics minimum viable product (MVP). Attendees will have hands on experiences with the free and/or affordable technology and services that can get their MVP up and running.
This talk assumes attendees understand the basic of Python. Basic understanding of MapReduce and NoSQL databases will be big pluses as well.
About the speaker
I am a data scientist, software developer, and aspiring entrepreneur. I'm especially interested in machine learning, information retrieval, text analytics, and cloud technology. My weapon of choice is Python.
I previously founded a social data analysis company and developed its core technology. I grew the startup into a profitable entity, without any outside funding.
...
https://www.youtube.com/watch?v=BJkWl5RiCAg
Behind the glamour of mobile apps, business SaaS products have been steadily building itself into a billion dollar industry. With companies such as Salesforce spearheading the adaption of business SaaS products in corporate, startups are rising to take on the SME market.
With hot topics such as "big data" and "machine learning" flying around, it may seem overwhelming to put an idea into action. I will walk through the process of building an data analytics minimum viable product (MVP). Attendees will have hands on experiences with the free and/or affordable technology and services that can get their MVP up and running.
This talk assumes attendees understand the basic of Python. Basic understanding of MapReduce and NoSQL databases will be big pluses as well.
About the speaker
I am a data scientist, software developer, and aspiring entrepreneur. I'm especially interested in machine learning, information retrieval, text analytics, and cloud technology. My weapon of choice is Python.
I previously founded a social data analysis company and developed its core technology. I grew the startup into a profitable entity, without any outside funding.
...
https://www.youtube.com/watch?v=BJkWl5RiCAg
Author
Content Type
Unspecified
video/mp4
Language
English
More from the publisher
Controlling
VIDEO
PLAY
play-digital-signal-processing-with
PyCon APAC 2022｜一般演講 Talks｜國泰金控 Cathay Financial Holdings / 美光科技 Micron 冠名贊助
✏️ 共筆 Note：https://hackmd.io/@pycontw/SyyTLTmyi
?? Slido：https://app.sli.do/event/dQGjkmjd7W2cAb1LEUtFGX
? 投影片 Slides：https://drive.google.com/file/d/1QBlbwnfirm-59eNpEyZnh5BRw17FI5oe/view?usp=sharing
? 語言 Language：中文演講/英文投影片 Chinese talk w. English slides
? 層級 Level：中階 Intermediate
? 分類 Category：嵌入式系統 Embedded Systems
? 摘要 Abstract ?
- 高階的音響喇叭非常昂貴，它們聽起來和一般的喇叭就是不一樣，但到底有何不同？這個差異有辦法數據化/視覺化嗎？
- 市面上的耳機這麼多款，到底哪一副耳機的音色你會最喜歡呢？
- 喜歡寫程式的你，是否想過如何用 Python 來控制並優化你音響喇叭的響應曲線呢？
- 我們有辦法讓一副耳機 去模仿任一款耳機的音色表現嗎？
- 為了控制音訊的產出，高階的設備上 有很多按鍵與旋鈕。音訊處理與產出的過程中，可以用 Python 程式 來取代這些按鍵旋鈕，並且做到自動化嗎？
- 如果我們不需要佈線 就可以控制遠端設備上的音訊產出，設備與場地的佈署也就更加容易。音訊設備的控制，可以 無線化/雲端化 嗎? 有興趣的話，進來聽聽吧！
? 說明 Description ?
- 好的音響聽起來就是不一樣，但是也非常貴。如果要提升自己家裡音響的表現，使用 DSP (Digital Signal Processing) 通常是最便宜與快速的方法了。
- 市面上最常見的 DSP 設備 應該就是等化器了，但是雖然經過等化器的音色改變了，和理想的音色表現到底還有多少差異，往往取決於專業耳朵的判斷，這也就是一般人要改善自己音響的最大門檻。
- 如果我們可以使用 Python 和一些便宜的設備，來測量與視覺化不同音響的 頻率響應曲線，應該就可以大幅度地降低這個門檻，幫助人們把錢花在刀口上。
- 在將音響的頻率響應 數據化/視覺化 之後，我們還需要能透過 Python 來產生所需要的 DSP 效果，讓我們音響的表現趨近理想。這時候 SigmaDSP (https://youtu.be/tBn7lo7RdkM) 和 SigmaStudio (https://youtu.be/PPmkCVz5LAs) 就是最方便有效的工具了。透過視覺化的圖形介面，使用者可以輕鬆地拖拉功能方塊，就可以達成需要的 DSP 效果。
- 而我們可以用 Python 來控制 SigmaDSP 的行為，也就可以動態地改變整個音訊的產出結果，讓音響的音色 可以藉由程式來自動地最佳化。
- 同時，我們更可以透過 TCP/IP channel 來控制遠端的 SigmaDSP，也就是說，透過 Python 程式，我們可以控制一群 DSP 設備，不管這些設備是在網路上的什麼地方。也正因為設備可以連網，所以可以方便地開放存取，資源因而得以分享共用，為後續開啟了更多的可能性。
- GitHub repo (https://github.com/Wei1234c/DRC)
- Code example (https://github.com/Wei1234c/DRC/blob/master/notebooks/Digital%20Room%20Correction.ipynb)
- More code examples (*.ipynb) (https://github.com/Wei1234c/DRC/tree/master/notebooks)
? 講者介紹 About Speaker - Wei Lin ?
策略規劃 是專長，
寫程式 是興趣，
對知識的渴望 與 領悟的喜悅 是我永恆的動力。
#pycontw #pyconapac2022 #python #sigmadsp #sigmastudio #dsp #embeddedsystems
Follow “PyCon Taiwan”
⭐️ Official Website: https://tw.pycon.org
⭐️ Facebook: https://www.facebook.com/pycontw
⭐️ Instagram: https://www.instagram.com/pycontw
⭐️ Twitter: https://twitter.com/PyConTW
⭐️ LinkedIn: https://www.linkedin.com/company/pycontw
⭐️ Blogger: https://pycontw.blogspot.com
...
https://www.youtube.com/watch?v=Yc0hUaKKuIw
Transaction
Created
1 year ago
Content Type
Language
video/mp4
zh
Controlling
VIDEO
QUEIC
queick-a-simple-job-queue-system-for
Day 1, 10:05-10:35
Abstract
In this talk, I introduce the concept of a job queue system named Queick that I individually developed, its internal architecture, and comparison with other job queue systems. Queick is a simple job queue system for Python. The main features are; 1. only std libraries are used 2. in-memory data store 3. retry when network connection available Queick stands on the idea of "Deadly Simple", thus, it is made of only standard libraries. This talk includes 4 topics. First, the overview of Queick and its application. Second, the general architecture of a job-queue system. Third, the design and the internal implementation of Queick. In the end, I will explain my OSS development experience. You will get the following knowledge through my presentation; general features and architecture of a job-queue system and things to be aware of when multi-thread programming is needed by using Python. https://github.com/asmsuechan/queick
Description
Queick is published at https://github.com/asmsuechan/queick. Also, there is a post at https://dev.to (https://dev.to/asmsuechan/queick-a-simple-job-queue-system-for-python-1jhi).
I will give my audience not only Queick specific information like my effort and my hardships, but also general information about job-queue system. Actually, the main point of my presentation is a comprehension of job-queue system through my explanation of Queick architecture.
Slides not uploaded by the speaker.
HackMD: https://hackmd.io/@pycontw/2021/%2F%40pycontw%2FSkVaByKMF
Speaker: Ryota SUENAGA
asmsuechan. A software developer working for M3, Inc. in Tokyo, Japan. 实际上，我会说一点儿中文。
...
https://www.youtube.com/watch?v=reLHrYEj6AA
Transaction
Created
1 year ago
Content Type
Language
video/mp4
English
Controlling
VIDEO
TIME
time-series-prediction-implement-on
PyCon Taiwan 2016｜一般演講 Talks
? 摘要 Abstract ?
Time series prediction has become one of the most popular field for applying in the real world. Because there are various models to forecasting the future data, how to choose a suitable model has become a significant issue for every companies who want to join the data driven trend. In this talk, we are going to share our experience and result of the implementation of time series forecasting models. The topic will include the following points:
1. How to choose a suitable model for variety datasets,
2. Why did we choose the current models (ARIMA+SVR, SdA),
3. How to implement the models on python,
4. What problems did we face when we are implementing the model.
? 關於講者 About Speaker - 古宣佑 Hsuanyo ?
目前於 Soocii 任職後端工程師，對機器學習演算法有興趣，工作之餘，會拿工作上面臨的問題，當作練習的題目。之前曾經接觸過時間序列預測的模型開發，在學時期則是專注於自然語言處理 (Sentiment Analysis) 的相關研究。
? 關於講者 About Speaker - Trudie ?
目前就讀於台灣大學資訊管理碩士，研究推薦系統與社群網絡，於物聯網分析公司擔任資料分析師，興趣在於開發資料分析應用。
#python #pycontw #pycontw2016
Follow “PyCon Taiwan”
⭐️ Official Website: https://tw.pycon.org
⭐️ Facebook: https://www.facebook.com/pycontw
⭐️ Instagram: https://www.instagram.com/pycontw
⭐️ Twitter: https://twitter.com/PyConTW
⭐️ LinkedIn: https://www.linkedin.com/company/pycontw
⭐️ Blogger: https://pycontw.blogspot.com
...
https://www.youtube.com/watch?v=ivkrU4n_ZRg
Transaction
Created
1 year ago
Content Type
Language
video/mp4
English
Controlling
VIDEO
90
90-分鐘帶你用-brownie-上架-nft
PyCon APAC 2022｜專業課程 Tutorials｜國泰金控 Cathay Financial Holdings / 美光科技 Micron 冠名贊助
✏️ 共筆 Note：https://hackmd.io/@pycontw/Hy1oU67Jo
?? Slido：https://app.sli.do/event/bxN7J26qBedGW6q2AJYYxX
? 投影片 Slides：https://drive.google.com/file/d/1I9gqjGpXYceR71NcrytJYseZl_4-0Cc6/view?usp=sharing
? 語言 Language：中文演講/英文投影片 Chinese talk w. English slides
? 層級 Level：入門 Novice
? 分類 Category：專案建置工具 Project Tooling
? 摘要 Abstract ?
本專業課程適合想了解 Web3 合約開發與部署或想轉職 Web3 工程師的 Python 開發者。 Brownie 是一用 Python 開發、測試並部署 Solidity 智能合約到以太坊區塊鏈（EVM）的框架。本議程將讓聽眾在 90 分鐘內跟著講者一起，從零開始建立一個 Brownie 開發環境，並且部署智能合約，上架自己的 NFT 到 Opensea 測試網。 除此之外，本議程也會帶聽眾認識 Vyper，一個基於 Python 語法的智能合約開發語言。 此課程新手友善，將從最基本的環境建置開始，會提供智能合約部分的程式碼，並帶你導讀合約。講者實作的成果最終版程式，也會預先釋出，讓卡關的朋友可以快速參考。
? 說明 Description ?
小抄筆記請點我：https://hackmd.io/@EasonC/ryxIsskjq
使用框架：Brownie
開發語言：Python, Vyper, Solidity
開發環境：VSCode
課程大綱：
- 基礎的區塊鏈知識
- 基礎智能合約知識
- 智能合約開發語言介紹與比較：Solidity 與 Vyper
- NFT 合約介紹
- 開發前的環境建置
- Metamask 建立
- 領取 Rinkeby 測試網 Token 做 Funding
- 建立自己的 Infura 區塊鏈結點
- 開發環境建置
- 範例程式碼下載與執行
- 會提供一完整版程式碼供參與者 Clone 下來當作參考
- 跟不上進度時可以複製貼上
- 成功後，會開始開發
- 開始開發
- 建立 Brownie 專案
- 專案架構講解
- ERC721 程式碼之複製貼上與導讀
- 將 NFT 發行到本地網路
- 將 NFT 發行到 Rinkeby 網路
- 測試腳本撰寫
- Bonus 時間：運用所學打造自己的專屬 NFT
- 問題 Q&A 時間
? 講者介紹 About Speaker - 陳怡升 EasonC13 ?
我是 Eason，現就讀於台師大學習科學學士班大四，並於區塊鏈新創 Cohart 擔任研發工程師。我有豐富的區塊鏈開發經驗，而 Brownie 是我第一個接觸的 Solidity 的開發框架，很適合熟悉 Python 的開發者入坑，並藉此了解區塊鏈合約的開發邏輯，之後觸類旁通，也可以快速熟悉其他框架如 Hardhat 或 Truffle。希望能透過這場議程，讓對區塊鏈感興趣的你，不僅能更了解區塊鏈，還能開發自己的去中心化應用。更多關於我請見 https://eason.tw
#pycontw #pyconapac2022 #python #brownie #nft #solidity
Follow “PyCon Taiwan”
⭐️ Official Website: https://tw.pycon.org
⭐️ Facebook: https://www.facebook.com/pycontw
⭐️ Instagram: https://www.instagram.com/pycontw
⭐️ Twitter: https://twitter.com/PyConTW
⭐️ LinkedIn: https://www.linkedin.com/company/pycontw
⭐️ Blogger: https://pycontw.blogspot.com
...
https://www.youtube.com/watch?v=Y7Xo6ObOQiA
Transaction
Created
1 year ago
Content Type
Language
video/mp4
zh
Controlling
VIDEO
TENSO
tensorflow-python-fault-detection-system
Day 1, R0 17:20–17:50
Based on our innovative technological capabilities, SK techx is committed to making our clients’ everyday lives happy and convenient with ICT by providing optimized goods and services for each stage of life.
These services consist of a Web server and a back-end service. Web services use Apache, Nginx, and services made in Java use Tomcat. There are, of course, services built with Python and the Go language.
Back-end services store various types of logs for their purpose.
There are many kinds of log, and the form is also different. Many developers and Open Sources have tried to interpret and process these formats.
I thought that I would automatically interpret these log formats and detect failures. I think TensorFlow is an alternative. First, I will describe the Fault Detection System architecture using TensorFlow. Second, I will introduce the use of the data generated step by step. Finally, I'll show you how to apply it to a service and how to detect it.
The speaker did not upload his slides.
...
https://www.youtube.com/watch?v=Z8zxKMQD9r0
Transaction
Created
1 year ago
Content Type
Language
video/mp4
English
Controlling
VIDEO
WHAT
what-is-async,-how-does-it-work,-and
Speaker: A. Jesse Jiryu Davis
Python's asynchronous frameworks, like asyncio, Tornado, and Twisted, are increasingly important for writing high-performance web applications. Even if you're an experienced web programmer, you may lack a rigorous understanding of how these frameworks work and when to use them. Let's see how asyncio's event loop works, and learn how to efficiently handle very large numbers of concurrent connections.
Attendees will learn when to use async and when not to. They'll learn about asyncio, a major new feature in the Python 3.4 standard library: what it is, how it works, and how to use it for high-concurrency web programming.
About the speaker
Author of Motor, an async MongoDB driver for Tornado, and of Toro, a library of locks and queues for Tornado coroutines. Author of GreenletProfiler. Contributor to Python, PyMongo, MongoDB, Yappi, Tornado, and Tulip.
...
https://www.youtube.com/watch?v=kdzL3r-yJZY
Transaction
Created
1 year ago
Content Type
Language
video/mp4
English
Controlling
VIDEO
SIMPL
simplearray-between-python-and-c++-yung
PyCon Taiwan 2023｜Talk 演講｜Day 1, R0 13:45–14:15
? 說明 Description ?
Contiguous buffer is what makes code run fast. With the regular and compact layout, data can go through the memory hierarchy in a quick way to the processor. Numpy makes use of it to achieve the runtime almost as fast as C.
However, not all algorithms can be easily maintained or designed by using array operations. A more general ways is to use the contiguous buffer as an interface. The dynamic typing system of Python makes it easy to use but hard to optimize. The typed, fixed-shape buffer allows efficient data sharing between the fast code written in C++ and the Python application. In the end, the system is as easy to use as but much faster than Python. The runtime is the same as C++.
In this talk, I will show how to design a simple array system in C++ and make it available in Python. It may be made as a general-purpose library, or embedded for ad hoc optimization.
? 投影片 Slides：https://www.slideshare.net/YungYuChen/simplearray-between-python-and-cpdf
? 講者介紹 About Speaker - Yung-Yu Chen ?
HPC software developer.
Follow “PyCon Taiwan”
⭐️ Official Website: https://tw.pycon.org
⭐️ Facebook: https://www.facebook.com/pycontw
⭐️ Instagram: https://www.instagram.com/pycontw
⭐️ Twitter: https://twitter.com/PyConTW
⭐️ LinkedIn: https://www.linkedin.com/company/pycontw
⭐️ Blogger: https://conf.python.tw/
...
https://www.youtube.com/watch?v=Uk0gtAZ-lao
Transaction
Created
5 months ago
Content Type
Language
video/mp4
English
Controlling
VIDEO
實
實現超自動化的驗證碼辨識
PyCon APAC 2022｜一般演講 Talks｜國泰金控 Cathay Financial Holdings / 美光科技 Micron 冠名贊助
✏️ 共筆 Note：https://hackmd.io/@pycontw/rk0o8aQ1s
?? Slido：https://app.sli.do/event/vnvojAmCGJ3BuYiqE1viLY
? 投影片 Slides：https://docs.google.com/presentation/d/1WsdQoRojBKlfOrDNppgbCmHK1DFC5gOh/edit#slide=id.p1
? 語言 Language：中文演講/英文投影片 Chinese talk w. English slides
? 層級 Level：中階 Intermediate
? 分類 Category：應用 Application
? 摘要 Abstract ?
本演講著重於讓聽眾了解自己如何實現自動辨識驗證碼模型的整體流程，我們是如何進行蒐集文字驗證碼的正確答案以確保獲得足夠的資料，並提供講者的訓練模型方式與經驗。結合 RPA (Robotic Process Automation，機器人流程自動化) 與 AI (Artificial Intelligence，人工智能) 兩個領域，我們可以實現網路驗證碼辨識超自動化 (Hyperautomation) 任務。
? 說明 Description ?
1. 為何需要破解驗證碼：(約 1 分鐘)
由於目前所擔任的職務是負責 RPA 相關任務，常常會收到使用者需要到某個網頁上查詢或點擊按鈕完成高工時且重複性高的例行工作任務。但有礙於部分網站在登入的時候常常會有文字驗證碼需要輸入，導致使用上有斷點，使用者必須人工登打。為了提升使用者滿意度體驗，我們嘗試破解驗證碼，結合深度學習訓練 OCR (Optical Character Recognition，光學字元辨識) 模型，實現超自動化任務。
- RPA：機器人流程自動化，取自 Robotic Process Automation 的縮寫，是指透過軟體技術的應用，達到自動協助辦公室工作者完成日常工作的解決方案。
- OCR：光學字元辨識，取自 Optical Character Recognition 的縮寫，是指對文字資料的圖像檔案進行分析辨識處理，取得文字及版面資訊的過程。
- 超自動化：超自動化是傳統業務流程自動化的延伸，超越了單一流程的範圍。透過將 AI 工具與 RPA 相互結合，超級自動化幾乎可為企業使用者實現所有重複性任務的自動化。
2. 千奇百怪的驗證碼：(約 1 分鐘)
本節跟各位聽眾展示各政府機關等等奇型怪狀的文字驗證碼，其中我們看到有些驗證碼相當歪斜，這會導致模型成效辨識不彰的原因之ㄧ。
3. 想像的做法：(約 1 分鐘)
有些網站很有趣，雖然有圖形驗證碼，但其實會附上答案的連結；有些是會提供音檔，而音檔的答案會附在 script 腳本。以上這些是比較少數的案例，但大多的時候，我們是沒有辦法直接取得驗證碼的答案，就必須透過 OCR 模型辨識取得答案。在還沒有自己訓練模型以前，過去會使用鼎鼎大名的 tesseract OCR 或是語音破解。我們也嘗試使用 tesseract ocr (https://github.com/tesseract-ocr/) 模型去跑搜集的驗證碼一萬張成效，平均辨識成功成效僅為 2.71%，因此在想完成自動登入網站的時候，常常會失敗。
4. 如何進行資料蒐集：(約 8 分鐘)
驗證碼搜集的關鍵在於大量包含正確答案的圖片，有越多的資料才有辦法訓練出一個好模型。透過以下方式搜集約 20 萬張驗證碼圖片和正確答案。蒐集到的答案的方式，統一命名為「答案_uuid.txt」避免檔名重複。本節展示這次模型抓了哪些網站的驗證碼及資料張數。此外，若爬到的驗證碼不確定大小寫，不會用此資料。
為蒐集大量的驗證碼圖片，本節將介紹 6 種有趣的方法來抓驗證碼資料或產製的方法。
5. 模型：(約 5 分鐘)
這次使用的 OCR 模型名稱為 ClovaOCR (https://github.com/clovaai/deep-text-recognition-benchmark)，參照 GITHUB README 即可自行訓練深度學習模型。以下附上訓練策略及成效
- 訓練環境使用 GPU
- 訓練集/驗證集 為 8 : 2 → 16萬：4萬
- 訓練 iteration：30 萬
- 最後 Training Accuracy： 0.99
- 訓練時間： 約 41.23 小時
- 比較 tesseract OCR，新的模型 ClovaOCR 則是在驗證集有 93.49% 的成效，每一張辨識時間僅需 0.1713 秒。
最後也與大家說明如何將 ClovaOCR 將模型推斷重構，符合驗證碼單張推斷的辨識方式。
? 講者介紹 About Speaker - 何信賢 ?
擔任 RPA 工程師與全端工程師，喜歡研究爬蟲和開發網站的服務。 歡迎來找我打羽球或是交流寫程式的心得:)
#pycontw #pyconapac2022 #python #clovaocr #tesseract #ocr #rpa
Follow “PyCon Taiwan”
⭐️ Official Website: https://tw.pycon.org
⭐️ Facebook: https://www.facebook.com/pycontw
⭐️ Instagram: https://www.instagram.com/pycontw
⭐️ Twitter: https://twitter.com/PyConTW
⭐️ LinkedIn: https://www.linkedin.com/company/pycontw
⭐️ Blogger: https://pycontw.blogspot.com
...
https://www.youtube.com/watch?v=oD-FT_33yW4
Transaction
Created
1 year ago
Content Type
Language
video/mp4
zh
Controlling
VIDEO
PRACT
practicing-statistics-in-python
Day 2, R1 10:50–11:35
在一個產品改動實驗裡，控制組與實驗組的平均分別是 169.61 與 169.88，這樣就說明了實驗組表現得比較好嗎？還是差異是來自隨機誤差呢？在這場演講，我們會介紹如何使用假設檢定對實驗結果進行判斷，而且是以 Python 的 SciPy 函數庫，最少僅需要一行程式碼，來介紹包含 p-value、α、β、effect size、sample size 等不可或缺的假設檢定必備知識。
Slides: https://speakerdeck.com/mosky/hypothesis-testing-with-python
Speaker: Mosky Liu
Mosky 是個熱愛 open source 精神的 Python 工程師，也是 Pinkoi 的 Python Charmer，工作時和同事一起打造能夠買到獨特禮物的設計品購物平台。
自從寫下人生第一支程式後，就難以忘懷以敘述為磚、邏輯為泥，堆砌出腦中藍圖的成就感，也熱愛分享自己所學，業餘時是 Python 課程講師，偶爾講點資料科學。也是數場國內外研討會的講者，包含臺灣的 PyCon、COSCUP、TEDxNTUST，以及在日本、新加坡、香港、韓國、馬來西亞等地的 PyCon。More: mosky.tw 。
...
https://www.youtube.com/watch?v=SMa9K0JxX6Q
Transaction
Created
1 year ago
Content Type
Language
video/mp4
zh