VIP20
本课程属于：005-Artificial Intelligence and Machine Learning Engineering English Technical Courses · 共 20 门相关课程，组合学习更完整。
查看所属VIP全部课程 →
← 返回所属目录 咨询 / 购买

Time Series Forecasting with Machine Learning Models

30 Chapters
View Course Samples Online
01
Time series forecasting fundamentals
components, patterns, and the role of machine learning models.
forecastingcomponents
02
Exploratory data analysis for time series
visualization, decomposition, stationarity, and autocorrelation.
Exploratoryanalysis
03
Feature engineering for time series
lag features, rolling statistics, calendar effects, and Fourier terms.
featuresstatistics
04
Data preprocessing and validation strategies
missing values, outliers, scaling, and time series cross-validation.
preprocessingvalidation
05
Baseline forecasting methods
naive, seasonal naive, mean, and drift as benchmarks for ML models.
Baselineforecasting
06
Linear regression for time series
trend, seasonality, and autoregressive distributed lag models.
regressionseasonality
07
Regularized regression
Ridge, Lasso, and Elastic Net for high-dimensional time series features.
Regularizedregression
08
Decision trees and random forests for time series forecasting
handling non-linearity and interactions.
Decisionforecasting
09
Gradient boosting machines
XGBoost, LightGBM, and CatBoost for tabular time series data.
Gradientboosting
10
Hyperparameter tuning and model selection for time series forecasting with scikit-learn and Optuna.
Hyperparameter tuning and model selection for time series forecasting with scikit-learn and Optuna.
Hyperparameterselection
11
Recurrent neural networks
RNN, LSTM, and GRU architectures for sequential forecasting.
Recurrentnetworks
12
Temporal convolutional networks and dilated causal convolutions for time series.
Temporal convolutional networks and dilated causal convolutions for time series.
Temporalconvolutional
13
Transformer-based models for time series
attention mechanisms and Informer, Autoformer, and PatchTST.
Transformer-basedattention
14
Hybrid and ensemble models
combining statistical, machine learning, and deep learning forecasts.
ensemblecombining
15
Probabilistic forecasting
quantile regression, conformal prediction, and prediction intervals.
Probabilisticforecasting
16
Multi-step ahead forecasting
recursive, direct, and multi-output strategies.
Multi-stepforecasting
17
Multivariate and hierarchical time series forecasting
vector autoregression, VARMA, and reconciliation.
Multivariatehierarchical
18
Handling non-stationarity and concept drift
online learning and adaptive models.
Handlingnon-stationarity
19
Anomaly detection and change point detection in time series using machine learning.
Anomaly detection and change point detection in time series using machine learning.
detection
20
Feature selection and dimensionality reduction for high-dimensional time series
PCA, autoencoders, and LASSO.
selectiondimensionality
21
Time series forecasting with exogenous variables
dynamic regression and causal inference.
forecastingexogenous
22
Evaluation metrics for forecasting
MAE, RMSE, MAPE, sMAPE, MASE, and pinball loss.
Evaluationforecasting
23
Backtesting and rolling-origin evaluation
best practices and common pitfalls.
Backtestingrolling-origin
24
Scalable forecasting pipelines
data ingestion, feature stores, and model deployment with MLflow and Airflow.
Scalableforecasting
25
Real-world case study
energy load forecasting using gradient boosting and LSTM.
Real-worldforecasting
26
Real-world case study
retail demand forecasting with hierarchical reconciliation and probabilistic outputs.
Real-worldforecasting
27
Real-world case study
financial time series forecasting with tree-based models and risk metrics.
Real-worldfinancial
28
Real-world case study
IoT sensor forecasting with edge deployment and online learning.
Real-worldIoT
29
Model interpretability and explainability for time series forecasting
SHAP, LIME, and attention weights.
interpretabilityexplainability
30
Capstone project
end-to-end machine learning forecasting system from data to deployment.
Capstoneend-to-end