Machine Learning Cheat Sheet: Key Concepts and Definitions, Exams of Machine Learning

This machine learning cheat sheet provides a concise overview of essential concepts and definitions, covering topics from data preparation to model deployment. It includes key terms such as data augmentation, model evaluation, and various machine learning algorithms. The sheet also touches on important aspects like hyperparameter tuning, transfer learning, and performance metrics, making it a valuable resource for students and practitioners alike. It also includes real code examples and model picking guides. Useful for university and high school students. (410 characters)

Typology: Exams

2024/2025

Available from 11/26/2025

iamvincent
iamvincent 🇰🇪

1.3K documents

1 / 6

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Cheat Sheet for
Machine Learning
Machine Learning - answers A subset of AI that enables systems to learn and improve
from experience
Deep Learning - answers A subset of machine learning using neural networks with
multiple layers
Business Objectives - answers Define what problem you're solving
Data Preparation - answers Collect, prepare and label data
Data Augmentation - answers Increase dataset size and variety
Model Development - answers Choose model architecture
Model Training - answers Train the model with prepared data
Model Evaluation - answers Measure performance
Model Deployment - answers Implement in production
Model Monitoring - answers Continuous assessment
Business Objectives - answers Must relate to bottom line and business value
Specific, measurable goals - answers Define specific, measurable goals
100% accuracy necessity - answers Consider if 100% accuracy is necessary or if partial
improvement provides value
Current costs calculation - answers Calculate current costs of not solving the problem
Image Formation - answers Follow proper image acquisition fundamentals such as
camera position, field of view, resolution, lighting, and filters
Good image formation - answers Good image formation reduces required dataset size
Training set - answers Used to build the model
pf3
pf4
pf5

Partial preview of the text

Download Machine Learning Cheat Sheet: Key Concepts and Definitions and more Exams Machine Learning in PDF only on Docsity!

Cheat Sheet for

Machine Learning

Machine Learning - answers A subset of AI that enables systems to learn and improve from experience Deep Learning - answers A subset of machine learning using neural networks with multiple layers Business Objectives - answers Define what problem you're solving Data Preparation - answers Collect, prepare and label data Data Augmentation - answers Increase dataset size and variety Model Development - answers Choose model architecture Model Training - answers Train the model with prepared data Model Evaluation - answers Measure performance Model Deployment - answers Implement in production Model Monitoring - answers Continuous assessment Business Objectives - answers Must relate to bottom line and business value Specific, measurable goals - answers Define specific, measurable goals 100% accuracy necessity - answers Consider if 100% accuracy is necessary or if partial improvement provides value Current costs calculation - answers Calculate current costs of not solving the problem Image Formation - answers Follow proper image acquisition fundamentals such as camera position, field of view, resolution, lighting, and filters Good image formation - answers Good image formation reduces required dataset size Training set - answers Used to build the model

Validation set - answers Used to prevent overtraining Testing set - answers Used to evaluate performance Image Classification - answers Labeling whole images (e.g., 'good part' vs 'bad part') Object Detection - answers Finding items in images with bounding boxes Segmentation - answers Identifying regions at pixel level Instance Segmentation - answers Identifying individual objects at pixel level Data Augmentation - answers Techniques to increase dataset size when limited data is available Augmentation methods - answers Methods include rotation, flipping, cropping, color adjustments Platform Selection - answers High-control platforms (F1 steering wheel) provide access to all hyperparameters User-friendly platforms - answers User-friendly platforms (normal steering wheel) provide simplified interfaces Hyperparameters - answers Values given to the training model that aren't derived from data Transfer Learning - answers Using pre-trained models as starting points, repurposing models trained for one task to perform a related task Confusion Matrix - answers Table showing predicted vs actual results True Positive (TP) - answers Correctly identified positives True Negative (TN) - answers Correctly identified negatives False Positive (FP) - answers Incorrectly identified positives (Type I error) False Negative (FN) - answers Incorrectly identified negatives (Type II error) Precision - answers TP/(TP+FP) - How accurate positive predictions are Recall - answers TP/(TP+FN) - How complete the results are F-Score - answers 2(PrecisionRecall)/(Precision+Recall) - Combined metric

Predict - answers The use of the trained model to make guesses about future data. Feature - answers The input data used to make predictions. Target - answers The output that the model is trying to predict. LogisticRegression() - answers A model best for binary classification tasks. DecisionTreeClassifier() - answers A model that uses rule-based splitting for decision making. RandomForestClassifier() - answers A model that prioritizes accuracy over simplicity by using multiple decision trees. KNeighborsClassifier() - answers A model that finds local similarity by identifying the closest data points. SVM() - answers A model that creates complex boundaries between different classes. Sequential() (Keras) - answers A type of model that builds layers similar to a neural network. CNN - answers A model specialized for processing images. RNN/LSTM - answers A model designed for handling time series or text data. Loss Function - answers A metric that measures how wrong the model's predictions are, where lower values indicate better performance. accuracy_score - answers A function that grades the model's predictions against the actual outcomes. train_test_split - answers A function that divides data into training and testing sets. model.fit - answers The method used to teach the model using training data. model.predict - answers The method used to make predictions on new data. Optimizer - answers A mechanism that adjusts the model's parameters based on the errors made, similar to a boxing coach providing real-time feedback. adam - answers A popular optimizer known for being super efficient and adaptable in adjusting the learning rate.

sgd - answers An optimizer that follows a slow and steady approach to updating model parameters. rmsprop - answers An optimizer that performs well in noisy environments by adapting the learning rate. One-Hot Encoding - answers A method of transforming categorical labels into a binary format where each class is represented by a unique vector with a single 'on' position. Weights - answers Parameters in a model that represent the strength of connections, analogous to brain wires. Gradient - answers A measure of how steep the error surface is, indicating the direction and rate of change of the loss function. Backpropagation - answers The process of propagating errors back through the network to update weights based on how much each layer contributed to the error. Update - answers The action of adjusting the weights of a model to minimize future errors. Abstract Image of Learning - answers A metaphor where the brain is likened to rubber bands that are pulled in different directions based on errors, with the optimizer adjusting tension after each trial. Movie Metaphor (Inception) - answers A conceptual analogy where each layer of a model is compared to a dream within a dream, with backpropagation as the mechanism that wakes up each layer with new information. Import Tools - answers The initial step in coding that involves bringing in necessary libraries for machine learning tasks. Load and Split - answers The process of dividing the dataset into features and target variables, and then splitting them into training and testing sets. Create and Train - answers The phase where a model is instantiated and trained on the training data. Predict and Score - answers The final step where predictions are made on the test set and the accuracy of the model is evaluated. Syntax - answers The structure and rules governing the code used in machine learning implementations.