Python Bytes: #210 Analyzing Kickstarter Campaigns with Python

<p>The <strong><a href="https://www.youtube.com/watch?v=WKAeXtLv5-k">live stream recording on YouTube</a></strong>.</p> <p>Special guest: <strong><a href="https://twitter.com/kjaymiller">Jay Miller</a></strong></p> <p>Sponsored by us! Support our work through:</p> <ul> <li>Our <a href="https://training.talkpython.fm/"><strong>courses at Talk Python Training</strong></a></li> <li><a href="https://testandcode.com/"><strong>Test &amp; Code</strong></a> Podcast</li> <li><a href="https://www.patreon.com/pythonbytes"><strong>Patreon Supporters</strong></a></li> </ul> <p><strong>Brian #1:</strong> <a href="https://medium.com/swlh/kickstarter-projects-do-they-succeed-f4a789010585"><strong>Analyzing Kickstarter Campaigns with Python Data Science Tools</strong></a></p> <ul> <li>Article title: “Kickstarter Projects — Do They Succeed?”</li> <li>Aditya Patkar</li> <li>Using a Kaggle dataset of 378,661-ish projects up to 2018.</li> <li>Looks at using pandas data frames to explore the data.</li> <li>Using <code>.describe()</code> data frame method to learn a lot.</li> <li>Uses matplotlib and seaborn to analyze the data further.</li> <li>Odd statement that I’m not sure is straight faced or a really dry joke: “The data from 1970 seems to be bad or insignificant data.”</li> <li>Examples of using heat maps, line graphs, bar charts, to look at different aspects.</li> <li>Some results: <ul> <li>35.64% of projects are successful (meaning goal hit)</li> <li>tech asks for the most for goals, and has the highest average per backer.</li> <li>Comics has the lowest pledged amount per backer average.</li> </ul></li> <li>Nice that you can use the techniques to ask your own questions of the data.</li> </ul> <p><strong>Michael #2:</strong> <a href="https://towardsdatascience.com/beyond-cuda-gpu-accelerated-python-for-machine-learning-in-cross-vendor-graphics-cards-made-simple-6cc828a45cc3"><strong>GPU Accelerated Python for Machine Learning on Cross-Vendor Graphics Cards</strong></a></p> <ul> <li>Building machine learning algorithms using the Vulkan Kompute Python Framework</li> <li>When you hear “CUDA”, that means Nvidia 🙂</li> <li>Uses <a href="https://github.com/EthicalML/vulkan-kompute">Vulkan Kompute</a> framework</li> <li>A large number of high profile (and new) machine learning frameworks such as Google’s Tensorflow, Facebook’s Pytorch, Tencent’s NCNN, Alibaba’s MNN —between others — have been adopting Vulkan as their core cross-vendor GPU computing SDK.</li> <li>As you can imagine, the Vulkan SDK provides very low-level C / C++ access to GPUs, which allows for very specialized optimizations.</li> <li>The main disadvantage is the verbosity involved, requiring 500–2000+ lines of C++ code to only get the base boilerplate required to even start writing the application logic.</li> <li><a href="https://github.com/EthicalML/vulkan-kompute#vulkan-kompute"><strong>The Kompute Python package</strong></a> is built on top of the Vulkan SDK through optimized C++ bindings, which exposes Vulkan’s core computing capabilities. Kompute is the Python <a href="https://en.wikipedia.org/wiki/General-purpose_computing_on_graphics_processing_units">GPGPU framework.</a></li> <li>The main article talks through a couple of numerical computation examples.</li> </ul> <p><strong>Jay</strong> <strong>#3:</strong> <a href="https://www.adafruit.com/product/4116">Adafruit PyPortal - CircuitPython Powered Internet Display</a></p> <ul> <li>Gift for the tinkering pythonista</li> <li>CircuitPy</li> <li>Use it to make plenty of cool things</li> <li>Screen/speaker/Light Sensor Built-In</li> </ul> <p><strong>Brian</strong> <strong>#4:</strong> <a href="https://pabloinsente.github.io/intro-linear-algebra"><strong>Introduction to Linear Algebra for Applied Machine Learning with Python</strong></a></p> <ul> <li>Pablo Caceres</li> <li>Intended as a reference and not a comprehensive review.</li> <li>Still, I very much appreciate it.</li> <li>Includes links to both free and paid resources to thoroughly learn linear algebra</li> <li>Covers <ul> <li>sets, ordered pairs, relations, functions, </li> <li>vectors</li> <li>matrices</li> <li>linear and affine mappings</li> <li>matrix decomposition</li> </ul></li> <li>Uses numpy, pandas, and altair for examples</li> <li>Quick (but useful) explanations of concepts, along with how to represent and do it with numpy</li> <li>I’m really just getting into it, but I’m enjoying it and this is the right level of handholding I needed.</li> </ul> <p><strong>Michael</strong> <strong>#5:</strong> <a href="https://deepnote.com/"><strong>How many notebook frameworks? Many, and now +1 with Deepnote</strong></a></p> <ul> <li>Deepnote is a new kind of data science notebook. Jupyter-compatible with real-time collaboration and running in the cloud. </li> <li>Free for individuals, paid for teams and companies</li> <li>Real time collaboration is a key feature</li> <li>Built in versioning coming</li> <li>Code review in the notebook coming</li> <li>“View” your variables as a whole environment</li> <li>Better — real — autocomplete</li> <li>Dashboards coming too</li> </ul> <p><strong>Jay</strong> <strong>#6:</strong> <a href="https://imagekit.io"><strong>imagekit.io</strong></a></p> <ul> <li>image cdn</li> <li>started using imagekit on my own website and noticed faster load times</li> <li><a href="https://imagekit.io/features/image-transformation">allows for some responsive “fanciness”</a> <ul> <li>Add Blurs</li> <li>Smart Cropping</li> </ul></li> <li><a href="https://github.com/imagekit-developer/imagekit-python">Python API</a> or URL-Schema</li> </ul> <p><strong>Extras</strong></p> <p><strong>Michael:</strong></p> <ul> <li>The Apple M1 mac mini wait continues. :)</li> <li>Talk Python To Me, pro edition</li> <li>PSF Fundraiser for the month of December: https://ift.tt/37yCUJF> </ul> <p><strong>Jay:</strong></p> <ul> <li><a href="https://www.youtube.com/channel/UC7z5VlhDHnorjUm6oW5dXcw">Elastic Community YouTube Channel</a> <ul> <li>Just posted my lightning talk on looking at open data from the government.</li> <li>Upcoming interview on one of our newest clients - Eland which is python client to create pandas-like dataframes with elasticsearch datastores.</li> </ul></li> <li>My Podcast <a href="http://podcast.productivityintech.com">The PIT Show</a> weekly insights from me on my developer journey and interviews with amazing folks in the tech space.</li> <li>Elastic Blog - Just posted my first Elastic Blog post <a href="https://www.elastic.co/blog/elastic-contributor-program-how-to-create-a-video-tutorial"><strong>Elastic Contributor Program: How to create a video tutorial</strong></a></li> </ul> <p><strong>Joke:</strong></p> <p>via <a href="https://twitter.com/Spirix3/status/1330611989891207168">twitter.com/Spirix3/status/1330611989891207168</a></p> <ul> <li>Q: why can't SQL and NoSQL Developers date one other?</li> <li>A: because they don't agree on relationships.</li> </ul>

from Planet Python
via read more