Tutorial Overview

This section provides practical tutorials for Torch-RecHub in different recommendation scenarios to help developers get started quickly.

Code Resources: The project provides interactive Jupyter Notebook tutorials (in the tutorials/ directory) and complete Python example scripts (in the examples/ directory) that can be used alongside this documentation.

Tutorial List

Tutorial	Description	Link
CTR Prediction	Click-through rate prediction model training	CTR Prediction Tutorial
Matching Models	Two-tower matching model training	Matching Models Tutorial
Complete Pipeline	End-to-end recommendation system	Complete Pipeline Tutorial

Quick Navigation

CTR Prediction (Ranking)

Learn how to use DeepFM, DCN, and other models for click-through rate prediction.

from torch_rechub.models.ranking import DeepFM
from torch_rechub.trainers import CTRTrainer

model = DeepFM(deep_features, fm_features, mlp_params)
trainer = CTRTrainer(model)
trainer.fit(train_dl, val_dl)

View Full Tutorial →

Matching Models

Learn how to use DSSM, MIND, and other models for vector retrieval.

from torch_rechub.models.matching import DSSM
from torch_rechub.trainers import MatchTrainer

model = DSSM(user_features, item_features)
trainer = MatchTrainer(model)
trainer.fit(train_dl)

View Full Tutorial →