Suman Saha’s Post

𝗣𝘆𝘁𝗵𝗼𝗻 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗣𝗮𝘁𝘁𝗲𝗿𝗻𝘀 🐍 | 𝗡𝘂𝗺𝗣𝘆 – 𝗦𝘂𝗺 & 𝗣𝗿𝗼𝗱 ➕✖️ | 📅 𝗗𝗮𝘆 𝟲𝟯 🚀 Today’s task: ✅ 𝗧𝗮𝗸𝗲 a 2D array (matrix). ✅ 𝗖𝗮𝗹𝗰𝘂𝗹𝗮𝘁𝗲 sum across rows. ✅ 𝗧𝗵𝗲𝗻 take product of the result. Core idea from the code: 𝙣𝙪𝙢𝙥𝙮.𝙨𝙪𝙢(𝙖𝙧𝙧, 𝙖𝙭𝙞𝙨=0) ➡️ Adds elements column-wise Then: 𝙣𝙪𝙢𝙥𝙮.𝙥𝙧𝙤𝙙(...) ➡️ Multiplies all resulting values Example concept: Matrix: [[1 2] [3 4]] Step 1 → Sum (axis=0) [1+3, 2+4] → [4, 6] Step 2 → Product 4 * 6 = 24 💡 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗧𝗮𝗸𝗲𝗮𝘄𝗮𝘆: Understanding axis is key: • axis=0 → column-wise • axis=1 → row-wise Strong candidates understand: • Reduction operations • Combining multiple NumPy functions • Data aggregation patterns Because real-world data tasks are all about: Transform → Aggregate → Compute Master these patterns — and NumPy becomes your superpower. #Python #NumPy #InterviewPrep #HackerRank #DataScience #ProblemSolving #DailyCoding #Consistency

𝗣𝘆𝘁𝗵𝗼𝗻 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗣𝗮𝘁𝘁𝗲𝗿𝗻𝘀 🐍 | 𝗡𝘂𝗺𝗣𝘆 – 𝗠𝗶𝗻 & 𝗠𝗮𝘅 🔍 | 📅 𝗗𝗮𝘆 𝟲𝟰 🚀 Today’s task: ✅ 𝗧𝗮𝗸𝗲 a 2D array (matrix). ✅ 𝗙𝗶𝗻𝗱 minimum of each row. ✅ 𝗧𝗵𝗲𝗻 find the maximum among those values. Core idea from the code: 𝙣𝙪𝙢𝙥𝙮.𝙢𝙞𝙣(𝙖𝙧𝙧, 𝙖𝙭𝙞𝙨=1) ➡️ Finds minimum in each row Then: 𝙣𝙪𝙢𝙥𝙮.𝙢𝙖𝙭(...) ➡️ Picks the maximum from those minimum values Example concept: Matrix: [[2 5] [3 7] [1 3]] Step 1 → Row-wise min [2, 3, 1] Step 2 → Max of result max(2, 3, 1) = 3 💡 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗧𝗮𝗸𝗲𝗮𝘄𝗮𝘆: This is a classic pattern: 👉 Min → then Max Strong candidates understand: • axis=1 → row-wise operations • Chaining NumPy functions • Data reduction strategies Because many real problems are about: Finding optimal values from constraints Learn to combine operations — that’s where real power lies. #Python #NumPy #InterviewPrep #HackerRank #DataScience #ProblemSolving #DailyCoding #Consistency

𝗣𝘆𝘁𝗵𝗼𝗻 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗣𝗮𝘁𝘁𝗲𝗿𝗻𝘀 🐍 | 𝗡𝘂𝗺𝗣𝘆 – 𝗙𝗹𝗼𝗼𝗿, 𝗖𝗲𝗶𝗹 & 𝗥𝗶𝗻𝘁 🔢 | 📅 𝗗𝗮𝘆 𝟲𝟮 🚀 Today’s task: ✅ 𝗧𝗮𝗸𝗲 a floating-point array. ✅ Apply: • 𝗙𝗹𝗼𝗼𝗿 • 𝗖𝗲𝗶𝗹 • 𝗥𝗶𝗻𝘁 Core idea from the code: numpy.floor(a) → Round down numpy.ceil(a) → Round up numpy.rint(a) → Round to nearest integer Example concept: Input → [1.1, 2.5, 3.9] Floor → [1, 2, 3] Ceil → [2, 3, 4] Rint → [1, 2, 4] 💡 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗧𝗮𝗸𝗲𝗮𝘄𝗮𝘆: Different rounding methods = different results. Strong candidates understand: • Floor vs Ceil difference • Rounding edge cases (like 0.5) • Element-wise operations in arrays Because in real-world data processing, small rounding choices can change outcomes. Precision matters. Details matter. #Python #NumPy #InterviewPrep #HackerRank #DataScience #ProblemSolving #DailyCoding #Consistency

I share open-source projects in my newsletter every week. Last week, the focus was on the freestiler project - a new geospatial library for R and Python by Kyle Walker. Key Features ⚙️ ✅ Generates PMTiles vector tilesets from spatial data objects, files, or database queries. ✅ Works with R and Python, enabling flexible integration into data science and geospatial workflows. ✅ Accepts multiple input sources, including sf objects, GeoParquet files, Shapefiles, GeoPackages, and DuckDB queries. ✅ Uses a Rust-based tiling engine that runs in-process, avoiding the need for external tile-building tools. ✅ Supports large-scale datasets through streaming pipelines that process data without loading everything into memory. More details are available in the project documentation. 🔗: https://lnkd.in/gS-kR7F8 License: MIT 🦄 📌 Subscribe to receive weekly updates: https://lnkd.in/gb3P8YdE #rstats #python #datascience

𝐓𝐨𝐩 𝐒𝐞𝐚𝐛𝐨𝐫𝐧 𝐏𝐥𝐨𝐭𝐬 𝐄𝐯𝐞𝐫𝐲 𝐃𝐚𝐭𝐚 𝐀𝐧𝐚𝐥𝐲𝐬𝐭 𝐌𝐮𝐬𝐭 𝐊𝐧𝐨𝐰 𝐢𝐧 𝟐𝟎𝟐𝟔 Data analysts rely heavily on visualizations to understand patterns hidden inside datasets. Python’s Seaborn library simplifies statistical visualization and helps analysts create clear, attractive charts with minimal code. This guide explains the most important Seaborn plots every data analyst should know in 2026. From scatter plots to heatmaps, these visualizations help uncover trends, correlations, and patterns quickly. #DataAnalytics #PythonVisualization #SeabornPlots #DataScience #PythonProgramming #analyticsinsight #analyticsinsightmagazine Read More 👇 https://zurl.co/mvmNa

Don't flatten what naturally has structure. It's tempting to model everything in a single class. Easy to write, easy to read, at least until your data grows. This is where most codebases start, with just one model. But with model composition, each model has a single responsibility. And Pydantic handles nested validation automatically. Structure your models the way your domain is actually structured. The code gets cleaner, the errors get clearer, and reuse becomes obvious. This and other real-world modelling patterns are covered in Practical Pydantic: 👉 https://lnkd.in/eGiB7ZxU Model your domain. Not just your data. #Python #Pydantic #Data #Models #Patterns

𝗣𝘆𝘁𝗵𝗼𝗻 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗣𝗮𝘁𝘁𝗲𝗿𝗻𝘀 🐍 | 𝗡𝘂𝗺𝗣𝘆 – 𝗭𝗲𝗿𝗼𝘀 & 𝗢𝗻𝗲𝘀 🔢 | 📅 𝗗𝗮𝘆 𝟱𝟵 🚀 Today’s task: ✅ 𝗧𝗮𝗸𝗲 𝗮𝗻 𝗮𝗿𝗿𝗮𝘆 𝘀𝗵𝗮𝗽𝗲. ✅ 𝗖𝗿𝗲𝗮𝘁𝗲 𝗮 𝗺𝗮𝘁𝗿𝗶𝘅 𝗳𝗶𝗹𝗹𝗲𝗱 𝘄𝗶𝘁𝗵 0s. ✅ 𝗖𝗿𝗲𝗮𝘁𝗲 𝗮𝗻𝗼𝘁𝗵𝗲𝗿 𝗺𝗮𝘁𝗿𝗶𝘅 𝗳𝗶𝗹𝗹𝗲𝗱 𝘄𝗶𝘁𝗵 1s. Only if you understand how NumPy initializes arrays. Core idea from the code: 𝙣𝙪𝙢𝙥𝙮.𝙯𝙚𝙧𝙤𝙨(𝙨𝙞𝙯𝙚, 𝙙𝙩𝙮𝙥𝙚=𝙞𝙣𝙩) Creates an array of zeros with the given shape. 𝙣𝙪𝙢𝙥𝙮.𝙤𝙣𝙚𝙨(𝙨𝙞𝙯𝙚, 𝙙𝙩𝙮𝙥𝙚=𝙞𝙣𝙩) Creates an array of ones with the same dimensions. Example concept: Shape → (2,3) Zeros: [[0 0 0] [0 0 0]] Ones: [[1 1 1] [1 1 1]] 💡 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗧𝗮𝗸𝗲𝗮𝘄𝗮𝘆: NumPy provides fast array initialization for many tasks. Strong candidates understand: • Array shape vs dimensions • Data type control using dtype • Efficient matrix initialization Because in data science and analytics, arrays are the foundation of computation. Master the basics — and complex operations become easier. #Python #NumPy #InterviewPrep #HackerRank #DataStructures #DailyCoding #Consistency

📊 Data Science Foundations Series – Part 1: NumPy Basics I’ve started strengthening my fundamentals in data science, beginning with NumPy. Here are some key takeaways: ✅ NumPy is faster than Python lists due to contiguous memory storage ✅ Supports vectorized operations (no need for loops) ✅ Efficient for handling large numerical datasets Some concepts I explored: 🔹 Array creation using np.array() and np.arange() 🔹 Reshaping data with .reshape() 🔹 Indexing and slicing (including negative indexing) 🤯 One interesting learning: m1[-5:-1:-1] returns an empty array. Reason: When stepping backwards, the start index must be greater than the stop index. ✔️ Correct approaches: m1[-1:-5:-1] m1[-5::-1] This small detail helped me better understand how slicing actually works under the hood. 📌 Next: Vectorization & Broadcasting #DataScience #Python #NumPy #LearningInPublic #CareerGrowth

🚀 Simplifying Trees in DSA! 🌳💻 While Arrays and Linked Lists are great linear structures, hierarchical data requires a Non-Linear approach—like Trees! To make revising easier, I created this visual cheat sheet. Just like a real-world tree has a Root and Leaves, a Tree data structure starts at the Root Node and branches out to Intermediate and Leaf Nodes. Here is what I have visually summarized in these notes: ✅ The core difference between Linear and Non-Linear structures ✅ 7 Types of Trees (including BST, Strict, Complete, and Skew Trees) ✅ Array Representation vs. Logical View ✅ Tree Traversal logic (Pre-order, In-order, Post-order) complete with Python code! 🐍 Visualizing the flow from the root down to the leaf nodes is a game-changer for understanding algorithms. Take a look and let me know in the comments—what is your favorite data structure to work with? 👇 #DSA #DataStructures #Algorithms #Python #CodingJourney #TechNotes #SoftwareEngineering #LearnInPublic

Just learned something interesting today 👇 In Data Analytics, cleaning data takes up almost 70–80% of the total work—not analysis. That means the real skill isn’t just knowing tools like Excel or Python… It’s knowing how to handle messy, real-world data. Small lesson, big perspective shift. What’s something surprising you’ve learned recently? #DataAnalytics #LearningInPublic #DataScience #GrowthMindset