# rpdnn | jerrygao | Knowledge Base | Instagit

This repository contains code for the paper "RP-DNN: A Tweet level propagation context based deep neural networks for early rumor detection in Social Media" By J. Gao, S. Han, X. Song, et al. - LREC 2020

GitHub Stars: 34

Repository: https://github.com/jerrygaolondon/rpdnn

---

## Articles

### [Common Failure Modes and Debugging Strategies for RP-DNN Training](/jerrygaolondon/rpdnn/what-are-common-failure-modes-and-debugging-strategies-for-rp-dnn-training)

Debug RP-DNN training failures like missing files, OOM errors, and NaN gradients. Learn common failure modes and effective debugging strategies for your RP-DNN models with this guide.

- Tags: debugging-strategies
- Published: 2026-03-04

### [How to Add Custom Context Features to the context_features_extractor Module in RPDNN](/jerrygaolondon/rpdnn/how-to-add-custom-context-features-to-the-context-features-extractor-module)

Learn to add custom context features to the context features extractor module. Extend functions, increment dimensions, and verify output shape for your RPDNN models.

- Tags: how-to-guide
- Published: 2026-03-04

### [RPDNN Source Tweet Encoding vs Social Context Encoding: Architecture and Implementation](/jerrygaolondon/rpdnn/what-is-the-role-of-source-tweet-encoding-versus-social-context-encoding-in-the-model)

Explore RPDNN source tweet encoding and social context encoding. Understand how semantic meaning and propagation patterns are captured and combined in this architecture.

- Tags: deep-dive
- Published: 2026-03-04

### [How the Hierarchical Encoder Processes Multi-Level Propagation Trees in RPDNN](/jerrygaolondon/rpdnn/how-does-the-hierarchical-encoder-process-multi-level-propagation-trees)

Learn how the hierarchical encoder processes multi-level propagation trees by converting JSON to tensor sequences, encoding tweets, and using attention layers for classification.

- Tags: deep-dive
- Published: 2026-03-04

### [How to Reproduce the Leave-One-Out Cross-Validation (LOO-CV) Experimental Setup in RPDNN](/jerrygaolondon/rpdnn/how-to-reproduce-the-leave-one-out-cross-validation-loo-cv-experimental-setup)

Easily reproduce the Leave-One-Out Cross-Validation LOO-CV experimental setup using the RPDNN repository. Train and evaluate models by iterating over pre-computed event data splits for accurate results.

- Tags: how-to-guide
- Published: 2026-03-04

### [Computational Bottlenecks in the RPDNN Training Pipeline: 3 Critical Stages Explained](/jerrygaolondon/rpdnn/what-are-the-computational-bottlenecks-in-the-training-pipeline)

Discover the 3 computational bottlenecks in the RPDNN training pipeline including I/O, ELMo embedding calls, and vocabulary construction. Optimize your training efficiency today.

- Tags: performance
- Published: 2026-03-04

### [How to Integrate RP-DNN with External NLP Pipelines for Production Deployment](/jerrygaolondon/rpdnn/how-to-integrate-rp-dnn-with-external-nlp-pipelines-for-production-deployment)

Deploy RP-DNN in production NLP pipelines. Package your AllenNLP model as a Predictor, expose it via API, and inject external features using the context tensor interface.

- Tags: how-to-guide
- Published: 2026-03-04

### [RPDNN Evaluation Metrics: F1, Accuracy, Precision, and Recall Explained](/jerrygaolondon/rpdnn/what-evaluation-metrics-f1-accuracy-does-the-model-use-for-performance-assessment)

Understand RPDNN evaluation metrics like F1, accuracy, precision, and recall. Learn how the Rumour DNN model assesses its performance for better results.

- Tags: performance
- Published: 2026-03-04

### [How to Fine-Tune ELMo Embeddings for Domain-Specific Rumor Detection with RP-DNN](/jerrygaolondon/rpdnn/how-to-fine-tune-elmo-embeddings-for-domain-specific-rumor-detection)

Fine-tune ELMo embeddings for domain-specific rumor detection using RP-DNN. Learn to load weights and encode tweets for enhanced accuracy in your NLP projects.

- Tags: how-to-guide
- Published: 2026-03-04

### [GPU Memory Requirements for Batch Sizes and Context Sizes in RPD‑DNN](/jerrygaolondon/rpdnn/what-are-the-gpu-memory-requirements-for-different-batch-sizes-and-context-sizes)

Discover the GPU memory needs for RPD-DNN batch and context sizes. Understand activation tensor requirements from 300MB to 4.8GB, plus static ELMo and LSTM encoder costs.

- Tags: performance
- Published: 2026-03-04

### [How to Use rumour_dnn_evaluator.py with Custom Trained Models](/jerrygaolondon/rpdnn/how-to-use-the-rumour-dnn-evaluator-py-script-with-custom-trained-models)

Evaluate your custom-trained rumour_dnn models using rumour_dnn_evaluator.py. Load your AllenNLP archive and apply mirrored training configurations for consistent inference.

- Tags: how-to-guide
- Published: 2026-03-04

### [How RPDNN Handles Class Imbalance in Rumor Detection Datasets](/jerrygaolondon/rpdnn/how-does-the-model-handle-class-imbalance-in-rumor-detection-datasets)

Discover how RPDNN effectively handles class imbalance in rumor detection datasets using hybrid oversampling and undersampling techniques for balanced training data.

- Tags: deep-dive
- Published: 2026-03-04

### [Required Preprocessing Steps for Credbank and PHEME Datasets in RPDNN](/jerrygaolondon/rpdnn/what-are-the-required-preprocessing-steps-for-credbank-and-pheme-datasets)

Learn the essential preprocessing steps for Credbank and PHEME datasets in RPDNN. Discover normalization, deduplication, and balancing techniques for cleaner data.

- Tags: how-to-guide
- Published: 2026-03-04

### [How to Implement Custom Attention Mechanisms in the Hierarchical Attention Network](/jerrygaolondon/rpdnn/how-to-implement-custom-attention-mechanisms-in-the-hierarchical-attention-network)

Learn to implement custom attention mechanisms in your Hierarchical Attention Network by subclassing and overriding the forward method. Explore dot-product, multi-head, and normalized variants easily.

- Tags: how-to-guide
- Published: 2026-03-04

### [What Is the Difference Between Context Metadata and Context Content in RP-DNN?](/jerrygaolondon/rpdnn/what-is-the-difference-between-context-metadata-cm-and-context-content-cc-in-the-rp-dnn-architecture)

Understand the distinction between context metadata and context content in RP-DNN. Learn how CM captures user reaction data and CC represents the text's semantic meaning.

- Tags: deep-dive
- Published: 2026-03-04

### [How the RPDNN Model Processes the PHEME Dataset Structure for Rumor Detection](/jerrygaolondon/rpdnn/how-does-the-model-process-pheme-dataset-structure-for-rumor-detection-tasks)

Learn how the RPDNN model processes the PHEME dataset structure for rumor detection. Discover steps like directory scanning, metadata extraction, class balancing, and feature enrichment.

- Tags: internals
- Published: 2026-03-04

### [How to Configure Context Type Filtering (Disable Reply vs Retweet) in rpdnn Training and Evaluation](/jerrygaolondon/rpdnn/how-to-configure-context-type-filtering-disable-reply-vs-retweet-in-training-and-evaluation)

Control context type filtering in rpdnn training and evaluation using the disable context type flag. Exclude replies or retweets to enhance rumor detection models.

- Tags: how-to-guide
- Published: 2026-03-04

### [ELMo Embedding Integration in the AllenNLP-Based RP-DNN Model: A Deep Dive](/jerrygaolondon/rpdnn/how-does-elmo-embedding-integration-work-in-the-allenlp-based-rp-dnn-model)

Discover how ELMo embedding integration enhances the AllenNLP-based RP-DNN model for rumor classification. Explore token and sentence level encoding for improved accuracy.

- Tags: deep-dive
- Published: 2026-03-04

### [Maximum Social Context Size in RPDNN: Balancing Memory and Accuracy](/jerrygaolondon/rpdnn/what-is-the-maximum-social-context-size-and-how-does-it-impact-memory-usage-and-accuracy)

Discover the maximum social context size in RPDNN and learn how it balances GPU memory and rumor detection accuracy. Optimize your settings for better results.

- Tags: deep-dive
- Published: 2026-03-04

### [How RP-DNN Implements Hierarchical Attention for Early Rumor Detection in Social Media](/jerrygaolondon/rpdnn/how-does-rp-dnn-implement-hierarchical-attention-for-early-rumor-detection-in-social-media)

Discover how RP-DNN leverages hierarchical attention in PyTorch for early rumor detection on social media. Understand its sequence and multimodal fusion levels.

- Tags: deep-dive
- Published: 2026-03-04