The pace of the blog posts has slowed a bit, but not the newsletter, so here's a new newsletter item. It has to do with the volume of work I'm luckily facing. I've also created an account on Upwork to keep broadening my data science and data engineering skills by doing actual client work. When you're new to the platform, you face some challenges, and that's something I'll write about later.
Until then, here's what's going on out there.
- Upskill to be the Most Sought-After Data Scientist
- Audio Effects Library by Spotify
- The relationship between the Law of Large Numbers & Central Limit Theorem
- Data Engineering Roadmap (831 upvotes; 72 replies)
- Asked a recruiter for a salary range, they responded with a non-answer. (275 upvotes; 133 replies)
- Showing the prediction of a kNN classifier based on the position (279 upvotes; 8 replies)
- Do you use OOP in your daily data science work? (202 upvotes; 159 replies)
- How do you handle 3 hour interviews when you have a job? (199 upvotes; 52 replies)
- 2nd time this week and it's Wednesday (226 upvotes; 16 replies)
- Some of these job postings out there are absolutely hilarious. WTF is this jumbled mess of shit? (145 upvotes; 88 replies)
- Is this group moderated actively? Can it be improved? (79 upvotes; 45 replies)
- I'm a Data Engineer. How do I become a better Software Engineer? (72 upvotes; 37 replies)
- What are some of the best data engineering jokes you have seen? (57 upvotes; 42 replies)
- [D] How OpenAI Sold its Soul for $1 Billion: The company behind GPT-3 and Codex isn’t as open as it claims. (612 upvotes; 110 replies)
- [P] CLIP Guided Diffusion: Generates images from text prompts Web Demo (593 upvotes; 22 replies)
- [P] OpenSource d*ck pic detection model to improve womens online life (456 upvotes; 23 replies)
- [D] Resources for Understanding The Original Transformer Paper (277 upvotes; 38 replies)
- [D] Schmidhuber: The most cited neural networks all build on work done in my labs (240 upvotes; 151 replies)
- Real Time Recognition of Handwritten Math Functions and Predicting their Graphs using Machine Learning (1172 upvotes; 29 replies)
- ML Meme War- This is one of my favourite Machine Learning Memes, what´s yours? (737 upvotes; 13 replies)
- Andrew Ng Courses - All Machine Learning And Deep Learning Courses (260 upvotes; 20 replies)
- A Lexicon of Forbidden Letters Generated by Machine Learning (185 upvotes; 16 replies)
- Bomberland competition: build ML agents for Bomberman (details in comments) (147 upvotes; 5 replies)
- If assumptions can be tested, why are they 'assumed'? (15 upvotes; 15 replies)
- What exactly is a probability distribution? (13 upvotes; 7 replies)
- What are good books to learn about stochastic and mechanistic modeling? (10 upvotes; 3 replies)
- Forecast Pooling/Combinations: Robust weighting methods that account for correlated forecasts? (8 upvotes; 0 replies)
- Statistical Analysis to Compare the Performance of Different Machine Learning Approaches (8 upvotes; 8 replies)
- Introducing Spice.ai - open source, time series AI for developers (13 upvotes; 0 replies)
- You Can Now Deepfake In Real-Time (12 upvotes; 0 replies)
- Make Slow Motion Videos With AI! TimeLens explained: a new model for video frame interpolation published at CVPR2021 (14 upvotes; 1 replies)
- [2108.13976] WarpDrive: Extremely Fast End-to-End Deep Multi-Agent Reinforcement Learning on a GPU (9 upvotes; 2 replies)
- [2106.01487] LLC: Accurate, Multi-purpose Learnt Low-dimensional Binary Codes (2 upvotes; 1 replies)
- How to get more effective at reading ML/AI/NLP papers? (15 upvotes; 4 replies)
- Topics Recommendation for ML Project (10 upvotes; 2 replies)
- Getting started with GraphML (10 upvotes; 1 replies)
- How do I implement predictive maintenance for a software project system? (8 upvotes; 3 replies)
- (PyTorch Lightning) Model Checkpoint seems to save the last epoch, even though I thought I'd set it to save the epoch with the minimum validation loss. Any ideas where my code is in error? (8 upvotes; 3 replies)
- Data Engineering Roadmap (836 upvotes; 72 replies)
- Asked a recruiter for a salary range, they responded with a non-answer. (273 upvotes; 133 replies)
- Showing the prediction of a kNN classifier based on the position (292 upvotes; 8 replies)
- Do you use OOP in your daily data science work? (206 upvotes; 159 replies)
- How do you handle 3 hour interviews when you have a job? (200 upvotes; 52 replies)
- 2nd time this week and it's Wednesday (228 upvotes; 16 replies)
- Some of these job postings out there are absolutely hilarious. WTF is this jumbled mess of shit? (145 upvotes; 88 replies)
- Is this group moderated actively? Can it be improved? (80 upvotes; 45 replies)
- I'm a Data Engineer. How do I become a better Software Engineer? (76 upvotes; 37 replies)
- What are some of the best data engineering jokes you have seen? (53 upvotes; 42 replies)
- [D] How OpenAI Sold its Soul for $1 Billion: The company behind GPT-3 and Codex isn’t as open as it claims. (607 upvotes; 110 replies)
- [P] CLIP Guided Diffusion: Generates images from text prompts Web Demo (602 upvotes; 22 replies)
- [P] OpenSource d*ck pic detection model to improve womens online life (456 upvotes; 23 replies)
- [D] Resources for Understanding The Original Transformer Paper (281 upvotes; 38 replies)
- [D] Schmidhuber: The most cited neural networks all build on work done in my labs (242 upvotes; 151 replies)
- Real Time Recognition of Handwritten Math Functions and Predicting their Graphs using Machine Learning (1176 upvotes; 29 replies)
- ML Meme War- This is one of my favourite Machine Learning Memes, what´s yours? (732 upvotes; 13 replies)
- Andrew Ng Courses - All Machine Learning And Deep Learning Courses (264 upvotes; 20 replies)
- A Lexicon of Forbidden Letters Generated by Machine Learning (182 upvotes; 16 replies)
- Bomberland competition: build ML agents for Bomberman (details in comments) (145 upvotes; 5 replies)
- If assumptions can be tested, why are they 'assumed'? (16 upvotes; 15 replies)
- What exactly is a probability distribution? (13 upvotes; 7 replies)
- What are good books to learn about stochastic and mechanistic modeling? (9 upvotes; 3 replies)
- Forecast Pooling/Combinations: Robust weighting methods that account for correlated forecasts? (9 upvotes; 0 replies)
- Statistical Analysis to Compare the Performance of Different Machine Learning Approaches (8 upvotes; 8 replies)
- Make Slow Motion Videos With AI! TimeLens explained: a new model for video frame interpolation published at CVPR2021 (13 upvotes; 1 replies)
- Introducing Spice.ai - open source, time series AI for developers (10 upvotes; 0 replies)
- You Can Now Deepfake In Real-Time (10 upvotes; 0 replies)
- [2108.13976] WarpDrive: Extremely Fast End-to-End Deep Multi-Agent Reinforcement Learning on a GPU (8 upvotes; 2 replies)
- [2106.01487] LLC: Accurate, Multi-purpose Learnt Low-dimensional Binary Codes (2 upvotes; 1 replies)
- How to get more effective at reading ML/AI/NLP papers? (13 upvotes; 4 replies)
- Topics Recommendation for ML Project (11 upvotes; 2 replies)
- Getting started with GraphML (10 upvotes; 1 replies)
- How do I implement predictive maintenance for a software project system? (8 upvotes; 3 replies)
- (PyTorch Lightning) Model Checkpoint seems to save the last epoch, even though I thought I'd set it to save the epoch with the minimum validation loss. Any ideas where my code is in error? (7 upvotes; 3 replies)
Let's keep in touch,
Pedro.
website | twitter | medium | github | stackoverflow | linkedin