PyStormTracker 0.5.0.dev0 Released: TRACK Algorithm & Data I/O Upgrades

Day 95 of my #100DaysOfCode journey 🚀 Today I solved the Number of Enclaves problem using BFS on a grid. Problem intuition: We are given a grid of land (1) and water (0), and we need to count how many land cells are part of an enclave — meaning they cannot reach the boundary of the grid. Key idea: If a land cell is connected to the boundary, it is not an enclave. Approach: • Push all boundary land cells (1) into a queue • Use BFS to mark all land connected to the boundary as visited / water • After traversal, count the remaining land cells • Those remaining cells are the enclaves Concepts reinforced: • BFS on 2D grid • Boundary traversal pattern • Connected components • Grid-to-graph conversion Time Complexity: • O(n × m) → every cell is processed at most once This problem is another strong reminder that many “grid problems” are really just graph traversal problems in disguise. The more I practice these patterns, the more naturally the solutions start forming. 🌱 #100DaysOfCode #DSA #Graphs #BFS #GridProblems #Algorithms #Python #CodingJourney #ProblemSolving #SoftwareEngineering

I just published a new blog post on a verification flow I've been building: using Lean4 as a formally verified golden model for RTL simulation testbenches. The core idea: most simulation flows trust their reference model implicitly it's just Python or C that someone wrote and hoped was correct. What if the reference model had machine-checked proofs attached to it? The flow combines three things: → A Lean4 golden model with correctness properties proved interactively → A cocotb testbench that loads the compiled Lean4 model via FFI → Hypothesis stateful testing that auto-generates input sequences and checks RTL output against the proven model When a mismatch is found, you know the RTL is wrong not the model. Some non-obvious details in the post: why RTLD_GLOBAL matters, why you pass 1 (not 0) to the module initializer, and why *bv_omega* is your friend for BitVec arithmetic proofs. Link in the comments 👇 #FormalVerification #HardwareVerification #Lean4 #TheoremProving #VLSI #RTL

#PathSim goes #Embedded 🚀 The C-code generator for the pathsim ecosystem is live at code.pathsim.org, free for academic and personal use. Just copy your pathsim script into the editor or export directly from #PathView and generate single file #MISRA compliant flat C-code and compile it for your embedded target. The platform also provides a validator that runs trajectory tests of the generated code against the pathsim model directly in the UI and then provides a report for qualification and documentation. See the full pathsim to hardware workflow in our upcoming webinar: https://lnkd.in/dn3_emMZ The pathsim ecosystem at pathsim.org For professional inquiries visit milanrother.com #C #Hardware #HiL #Python

Milan has the productivity of a small corporation! I haven't dug into this yet, but it looks like a really important step forward for developing complex code for embedded systems. For academic users of existing proprietary block diagram modelling tools this can potentially reduce cost and make it easier to create labs where control algorithms are tested in simulation and practice. Check it out.

Friday observation: The most complex data pipelines aren't always the best ones. Too often, we get lost in the weeds of authentication, WebIDs, and boilerplate code, forgetting that the goal is simply to get data from A to B. If you are working with the PI System, I put together a deep-dive on how to streamline this. It covers everything from Kerberos auth to efficient streaming. It’s a great read for the weekend—hope it saves you a few hours of debugging next week: [https://lnkd.in/dwgPKh_m] Let me know your thoughts—what’s the one PI System challenge you’re currently working on? 👇 #PISharp #IndustrialData #DataEngineering #Python #Automation #FridayReads

Added a small but practical quant utility in Z-score.py: add_zscore_feature(). It adds robust z-score features to a pandas DataFrame with support for: - Full-series or rolling-window normalization. - Group-wise z-scores (e.g., by ticker/sector). - Safe handling of zero-variance edge cases. - Optional in-place updates for flexible workflows. Clean feature engineering like this helps keep signal pipelines reproducible and model-ready. #Python #QuantFinance #AlgorithmicTrading #SoftwareEngineer

🚀 GFG POTD – Day 4 Completed! 🧩 Problem: Consecutive 1’s Not Allowed Today’s challenge was a classic DP problem where we count binary strings of length n such that no two 1s are consecutive. 💡 Key Insight: This problem follows a Fibonacci pattern: If last bit is 0 → no restriction If last bit is 1 → previous must be 0 So, 👉 dp[n] = dp[n-1] + dp[n-2] ⚡ Optimized Approach: Used O(1) space by keeping track of just the last two values. 🐍 Python Code: class Solution: def countStrings(self, n): if n == 1: return 2 a, b = 2, 3 for _ in range(3, n + 1): a, b = b, a + b return b ✅ Result: All test cases passed 📈 Complexity: O(n) time | O(1) space Consistency is the key — Day 4 done, moving forward! 💪 #GFG #POTD #Day4 #DataStructures #DynamicProgramming #CodingJourney GeeksforGeeks R P Sarathy Institute of Technology (Autonomous)

Excited to share that my R package discoCVI is now submitted to CRAN! discoCVI implements the DISCO metric a Cluster Validity Index for evaluating density-based clustering results without ground truth labels. Key features: - Handles arbitrary-shaped clusters (rings, spirals, crescents) - Explicitly evaluates noise point quality - Returns interpretable scores in [-1, 1] The package is a faithful R translation of the Python reference implementation by Beer et al. (2025) arXiv:2503.00127, verified across 8 benchmark datasets with differences about < 10^-14. Install from GitHub: devtools::install_github("aminentezari/discoCVI") Special thanks to Prof. Davide Chicco for supervision and guidance. #RStats #DataScience #MachineLearning #Clustering #OpenSource #CRAN

Minimum Window Substring A sliding window algorithm that finds the smallest contiguous segment in a data stream containing a specific set of required labels. It uses dual pointers to expand for validity and contract for optimization. Why do we need this because in production for eg agri-tech , we need to trigger sensor calibrations the moment specific fruits pass through the camera like mango / banana. finding the min window minimizes the latency and memory overload on edge hardware. Code Pushed :- https://lnkd.in/gCdq87Xp #Python #SoftwareEngineering #AgriTech #Grit #BuildInPublic

Anthropic Accidentally Leaks 512K Lines of Claude Code via npm Error 📌 Anthropic accidentally leaked 512K lines of Claude Code via an npm packaging glitch, exposing proprietary AI architecture, feature flags, and even undercover modes. The breach, caused by a missing .npmignore rule, triggered supply chain risks as trojans spread across GitHub forks. Developers are now racing to reimplement the system in Python and Rust - all while Anthropic scrambles to contain the fallout. 🔗 Read more: https://lnkd.in/d4DBVafW #Anthropic #Claudecode #Npm #Cloudflare #Typescript