#datatalksclub 搜尋結果

XGBoost is an implementation of gradient boosting. My github repo: github.com/kunlemariam08/… #AI #DataTalksClub #MachineLearing #Lifelonglearning #MLZoomcamp #DataScience #AlexeyGrigorev

#AI #DataTalksClub #MachineLearing #Lifelonglearning #MLZoomcamp #DataScience #AlexeyGrigorev

Module 4 homework? Completed ✅ Accuracy is not all it seems 👀 So good to cement my understanding of binary classification evaluation metrics #mlzoomcamp #DataTalksClub

"Understanding Model Evaluation in Lead Scoring: A Practical Walkthrough" by @VinAnoliefo #DataTalksClub #alexeygrigorev #DEVCommunity dev.to/techkene/under…

Just wrapped up a lead scoring model project Learned how to: ✅ Evaluate models using ROC AUC, Precision, Recall & F1 ✅ Use 5-Fold CV for stability ✅ Tune hyperparameters (Best C = 1) ✅ Identify top feature — lead_score #MachineLearning #DataTalksClub #alexeygrigorev

Cross-validation is key to building reliable models. Instead of a single train/test split, it rotates data to create multiple validation sets. This provides a more accurate measure of model performance and helps prevent overfitting. #MachineLearning #DataTalksClub #mlzoomcamp

Accuracy can be deceiving. The confusion matrix reveals the truth behind your model's performance by breaking down its correct and error types. It's the foundation for crucial metrics like Precision and Recall. #MachineLearning #DataTalksClub

ROC Curve - a way to evaluate the performance at all thresholds; okay to use with imbalance K-Fold CV - more reliable estimate for performance (mean + std). My github repo: github.com/kunlemariam08/… #DataTalksClub

What's a single, powerful metric for model comparison? AUC (Area Under the Curve). It measures a model's ability to separate positive and negative classes, independent of any specific threshold. 🎯 1.0 = Perfect classifier 🤷 0.5 = Random guessing #MLZoomcamp #DataTalksClub

To really understand a binary classification model, you need a Confusion Matrix 📊. It splits every prediction into 4 buckets: ✅ True Positive (TP) ✅ True Negative (TN) ❌ False Positive (FP) (Type I Error) ❌ False Negative (FN) (Type II Error) #MLZoomcamp #DataTalksClub

Just wrapped up a Lead Scoring assignment 🧠📊 ✅ Cleaned and prepped real-world data ✅ Explored correlations & mutual info ✅ Trained a Logistic Regression model ✅ Achieved 68% validation accuracy #mlzoomcamp #DataTalksClub #AlexeyGrigorev.

"From Raw Leads to Predictive Insights: My Logistic Regression Assignment Journey" by @VinAnoliefo #DEVCommunity dev.to/vinanoliefo/fr… #mlzoomcamp #DataTalksClub #AlexeyGrigorev

#AI #DataTalksClub #MachineLearing #Lifelonglearning #Alex #MLZoomcamp #DataScience #AlexeyGrigorev

Ever wonder how an ML model makes a "yes/no" decision, like predicting customer churn? 🤔 It often comes down to Logistic Regression! It's a linear model with a secret weapon for classification: the sigmoid function. 🧠 #MLZoomcamp #DataTalksClub

Kicked off module 3 in #MLZoomcamp by @Al_Grigor! 🚀 We're building a classification model to predict customer churn. The goal: identify at-risk customers and send them targeted discounts to prevent revenue loss! 💰 #DataTalksClub #BusinessAnalytics #LearningInPublic

👉 github.com/kunlemariam08/… Excited to keep building and learning through the next module! 🚀 #MachineLearning #MLZoomcamp #DataTalksClub #AI #DataScience #AlexeyGrigorev #LifelongLearning

Just wrapped up Module 2 of #mlzoomcamp! 🚀 Built a car price regression model. Key lessons: 🔹 Tame "long-tail" data with a log transform 🔹 Feature engineering is crucial for accuracy 🔹 Regularization is essential to fix model instability #DataTalksClub #learninginpublic

This week on #MLZoomcamp a car price prediction model! 🚗Built a linear regression model 📈 from scratch to predict vehicle MSRP. From data cleaning and EDA to regularization, it was an incredible learning journey. #DataTalksClub #LearningInPublic #Regression #MachineLearning

Enjoying the machine learning Zoomcamp on DataTalks, quite a meaningful learning platform. #mlzoomcamp #DataTalksClub

Machine Learning Zoomcamp journey! 🤖 The core concept? Let the machine find the patterns in the data. We provide the examples, it learns the rules. Mind-blowing stuff! 🤯 #MachineLearning #MLZoomcamp #DataTalksClub

#DataTalksClub #LearningInPublic #MachineLearning #CrossValidation #DataScience

Ever wonder how an ML model makes a "yes/no" decision, like predicting customer churn? 🤔 It often comes down to Logistic Regression! It's a linear model with a secret weapon for classification: the sigmoid function. 🧠 #MLZoomcamp #DataTalksClub

Kicked off module 3 in #MLZoomcamp by @Al_Grigor! 🚀 We're building a classification model to predict customer churn. The goal: identify at-risk customers and send them targeted discounts to prevent revenue loss! 💰 #DataTalksClub #BusinessAnalytics #LearningInPublic

We used a Logistic Regression model on the Telco Customer Churn dataset to turn data into actionable business intelligence. #DataTalksClub #MachineLearning #Classification #Business Analytics

Starting my new #MLOps project with a price prediction model for milk in Mexico 🥛🇲🇽 Chose Pipenv over Conda: ✅ Pure Python ✅ Lightweight for Lambda ✅ Reproducible with Pipfile.lock Sometimes simple > powerful. #MLZoomcamp #DataTalksClub #MachineLearning #OpenData #AWS

Good documentation isn't just a "nice to have," it's a sign of professional engineering. It transforms a personal project into a reusable, understandable, and maintainable blueprint for others (and my future self!). #MLOps #MLOPSZoomcamp #DataTalksClub #Portfolio

Just wrapped up Module 2 of #mlzoomcamp! 🚀 Built a car price regression model. Key lessons: 🔹 Tame "long-tail" data with a log transform 🔹 Feature engineering is crucial for accuracy 🔹 Regularization is essential to fix model instability #DataTalksClub #learninginpublic

Submitting the assignment of week #1🥳 Grasped the basic of building ml models and get my hands dirty on Numpy and Pandas🧑🏼‍💻 #mlzoomcamp #DataTalksClub

In my first week with data engineering, #zoomcamp with #DataTalksClub, I learned the concept of Docker, I set up a #GoogleCloud account and started ingesting data into Postgres, also some SQL queries refresher. #dezoomcamp

To really understand a binary classification model, you need a Confusion Matrix 📊. It splits every prediction into 4 buckets: ✅ True Positive (TP) ✅ True Negative (TN) ❌ False Positive (FP) (Type I Error) ❌ False Negative (FN) (Type II Error) #MLZoomcamp #DataTalksClub

Enjoy this great podcast with the bit.ly/3OFInD0 'StoryTime for DataOps'. bit.ly/3OwKvgq #DataTalksClub #DataOps

I had al ot of fun on this podcast with the bit.ly/38oVFmS 'StoryTime for DataOps'. bit.ly/3kjEFkw #DataTalksClub #DataOps

The four levels of model deployment exemplified in a churn prediction web service can be shown succinctly in this sketch from the #mlzoomcamp course. #datatalksclub

🔍 Dealing with categorical variables in slight variations (e.g., 1 Series, 1_series, 1 series)? Ensure consistency with this code snippet! Clean features lead to better model performance. #mlzoomcamp #datatalksclub

🌟 CRoss Industry Standard Process for Data Mining (CRISP-DM) is a powerful methodology for structuring ML projects. It includes: - Business Understanding - Data Understanding - Data Preparation - Modeling - Evaluation - Deployment 🚀 #mlzoomcamp #datatalksclub

📈 The ROC curve plots the true positive rate (recall) against the false positive rate (FPR) across thresholds from 0 to 1. To compare the 'skill' of binary classifiers, we use the Area Under the ROC Curve (AUC). 🔍✨ #mlzoomcamp #DataTalksClub

🎯 AUC isn't just for evaluating binary classifiers! It can also assess the importance of numerical features in predicting a binary target. 📊 In Python, calculate AUC for each feature. Higher AUC implies higher importance!💡 #mlzoomcamp #DataTalksClub

📊 Preparing your data for machine learning? Normalize your columns with this code snippet! It converts all column names (features and targets) to lowercase and replaces spaces with underscores. Clean data leads to better models! #mlzoomcamp #datatalksclub

The risk ratio is a simple method for assessing feature importance in churn prediction. By dividing group churn rate by global churn rate, you get the risk rate. A rate > 1 indicates higher churn likelihood, while < 1 suggests the opposite. 📊 #mlzoomcamp #datatalksclub