gauravkr0071 / data-scientist-roadmap Goto Github PK

• Linear Algebra : eigen values, eigen vectors, PCA, ICA, Pseudo Inverse, Matrix\vector differntiation, Norm
• Central Limit theurom and its importance
• A/B testing, Hypothesis, P-value, F-value, T-test
• Chi-Squared test
• Sampling and its types
• Data cleansing
• Handling Missing data
• Outlier Detection
• Feature Selection Techniques
• Feature engineering
• Expectation, Variance and Mean
• Error analysis
• Bias and variance and its trade-off
• linear/non-lineer/multiple/logistic regression , assumptions, performance metrics
• R-value, Adjusted-R, P-value
• L1 and L2, ridge and lasso
• Ensemble techniques and why its work ?
• Decision tree(CART, ID3.4)
• How to compute Entropy for conntnious data ?
• Bagging, Boosting , Stacking, (why bagging works)
• Underfitting and Overfitting, Tradeoff, overcoming them , preventive methods
• Random Forest
• Adaboost, Gradient Boost, XGboost etc.
• Perceptron model, Activation function, Neural network/cnn, Backpropagation, Grad descent/asscent
• Effect of Batch size, learning rate.
• Loss function/ optimization/ how to derive them
• SVM and assumptions
• conditional probability, conditional idependence
• Naive bayes and assumption
• Confusion Matrix, AUC, ROC, false positive, false negative, etc.
• Performance metrics in CNN (mAP, confusion matrix etc.)
• K- nearest neighbour, Assumption, Performance metrics, Advantages and Disadvantages.
• K- mean, Assumption, Performance metrics, Advantages and Disadvantages.
• DSBCAN and Other clustering algorithm
• Gaussian mixture model
• Expectation maximization

If you are using this repo for prepration and in some work, and it helped you in anyway, please consider dropping a mail to me or cite this repo with link in any of your repo and follow me on github.

A big shoutout to Analytics vidhya , medium blog, statistics how to , and the whole internet for giving knowledge for free.