Skip to content

📚 Tutorials & Guides

Welcome to Rankify tutorials! These guides cover everything from basic usage to advanced customization.

🚀 Retrieval & Search Techniques

Tutorial Description
📌 Introduction to Information Retrieval Overview of retrieval in Rankify
🔍 Using Sparse Retrievers (BM25) BM25 sparse retrieval
🧠 Using Dense Retrievers DPR, ANCE, ColBERT, BGE, Contriever
🤖 Hybrid Retrieval Combining sparse & dense methods
📂 Prebuilt Corpora & Indexes Using Wikipedia and MS MARCO indices
🔎 Custom Datasets & Indexing Building your own indices

📊 Re-Ranking Strategies

Tutorial Description
📌 Introduction to Re-Ranking Overview of 23 reranking methods
🎯 Pointwise Re-Ranking MonoBERT, MonoT5, UPR, FlashRank
🔄 Pairwise Re-Ranking RankGPT, InRanker, EchoRank
📃 Listwise Re-Ranking RankT5, LiT5, Transformer Rankers
🦾 API-Based Rerankers Voyage, Jina, MixedBread.ai
📈 Comparing Performance Benchmarking rerankers

🧠 Retrieval-Augmented Generation (RAG)

Tutorial Description
📌 Introduction to RAG Overview of 7 RAG methods
📥 Zero-Shot RAG GPT, LLaMA, vLLM backends
🔁 Fusion-in-Decoder (FiD) FiD architecture
📄 In-Context Learning Chain-of-Thought, Self-Consistency, ReAct
⚙️ Building RAG Pipelines End-to-end systems
📊 Evaluating RAG Models EM, F1, BLEU metrics

📂 Working with Datasets

Tutorial Description
📌 Prebuilt Benchmark Datasets NQ, TriviaQA, SQuAD, etc.
🛠 Creating Custom Datasets Build from your data
📥 Loading & Saving Dataset I/O
📊 Dataset Evaluation Evaluate retrieval quality

🛠 Evaluation & Benchmarking

Tutorial Description
📏 Retrieval Metrics Recall@k, MRR, P@k
📈 Reranking Metrics NDCG, MAP
🧠 RAG Metrics Exact Match, F1, Contains
📊 Method Comparisons Systematic benchmarking

⚡ Advanced Usage & Customization

Tutorial Description
🛠 Custom Retrievers Extend BaseRetriever
🔧 Custom Rerankers Extend BaseRanking
⚙️ Custom RAG Models Create new RAG methods
💾 Saving & Loading Model persistence

🚀 Deployment & Integration

Tutorial Description
🔌 Large-Scale Applications Batch processing, multi-GPU
🌍 External APIs OpenAI, Cohere, LiteLLM
🖥️ Cloud & GPUs vLLM, Docker, cloud deployment
🐞 Debugging Logging, profiling