Andy Murtha’s Post

What is {messy} and what do I like about it? “The {messy} R package takes a clean dataset, and randomly adds mess to create data more similar to that which you'd find in the real world. This is an easy way for educators to create data sets that give students the opportunity to practice their data wrangling skills without having to change all of their examples.” In many statistics courses, we learn how to work with data that is pretty much set up for the analysis we want to perform, but that’s not reality. What I like about this package it that it makes it easy to add a layer of complexity to assignments for learners. I can use it to offer another element of the critical thinking needed in real-world data analysis in my classroom, helping them be more prepared to hit the ground running in the next data project they work on. I am looking forward to implementing it in a future course 💛

Day 34 of #100DaysOfMachineLearning Today, I implemented a Random Forest algorithm. This experience provided valuable insights into ensemble learning and the power of combining multiple decision trees to improve predictive accuracy. Excited to see how this model performs on various datasets! GitHub: https://lnkd.in/ec6PGE5E #MachineLearning #DataScience #RandomForest #EnsembleLearning #LearningJourney #ML

Partial fitting, sparse data structures, semi supervised learning, pipeline caching, sample weights. Scikit-learn can do THAT?! Yep, it really does do that and more. Related: my PyData Eindhoven talk on the topic is out. I hope folks enjoy it!

🔴 what is KBinsDiscretizer Is a preprocessing technique in scikit-learn used to discretize continuous features into discrete bins. It divides the range of the continuous variable into k intervals, or bins, and then assigns each data point to the appropriate bin 🔴 Parameters: 💠 n_bins: Specifies the number of bins. 💠 encode: Determines encoding of transformed data 💠 strategy: Determines binning strategy ('uniform', 'quantile', 'kmeans') #DataScience #DataPreprocessing#MachineLearning

🚀 Excited to continue the Machine Learning Series journey with you all!💡 📝 Today's focus: Pandas Learning Source: https://lnkd.in/g5ZcZR93 Mind Map Link: https://lnkd.in/gePvA7j9 Stay tuned for daily updates as I progress through my #MachineLearningSeries! 📈 Excited to share insights and learnings along the way. Let's grow together! 🌱 #Day2 #MachineLearning #DataScience #Pandas #LearningJourney

Day 13/21 : Today, I dove into the captivating realm of subarrays equaling a sum, harnessing the mighty power of prefix sums. Mastering this concept is pivotal in conquering data structures and algorithms. Mohammad F. Pulkit Chhabra #CodingJourney #PrefixPower #AlgorithmMastery #DataStructures"

Discover the most common and openly accessible datasets, ready to be used to practice and train your classification models. #DataSets #ClassificationModels #IrisDataset #MNIST

🤔 Facing a dilemma with a column comprising nearly 40% NaN values? Removing them risks losing crucial data. What's the solution?   💡 Enter pd.isna()! This pandas function is a lifesaver, pinpointing missing or NaN (Not a Number) values within a DataFrame or Series. It generates a boolean mask, highlighting where values are NA, enabling seamless data cleaning and preprocessing. 🐼💻   Check out the code snippet below!   Remember, every NaN value represents an opportunity for insight, not a setback. Let's harness the magic of pd.isna() to uncover hidden gems within our data. Keep exploring, keep learning! 🌟🐼 #DataAdventures #PandasMagic #Pandas #DataCleaning #DataScience

Just completed a course on XGBoost, enhancing my understanding of its applications in classification and regression tasks. Explored key hyperparameters for model optimization and integration into pipelines, alongside practical tips for working with Pandas DataFrames and scikit-learn. Excited to apply these insights in my daily work! #DataScience #XGBoost #ContinuousLearning"

TrackCode:DS📌 Task 1: Create a bar chart to visualize the distribution of a categorical or continuous variable ✅ Platform Used: Google Colab Dataset: "Summer-Olympic-Medals" This bar chart showcases the distribution of selected category or continuous variables within the Tasker platform. Utilizing Google Colab, the data was processed and visualized to illustrate the frequency or spread of these variables, enabling easy analysis and interpretation📈 #prodigy #datascience #prodigyinfotech #task1