Comparing different ML models for sentiment classification based on movie review datasets

Implementation of DALL-E 2, OpenAI's updated text-to-image synthesis neural network, in Pytorch

A very simple framework for state-of-the-art Natural Language Processing (NLP)

Code used for evaluation and baselines in the PAN shared tasks.

🧠💬 Articles I wrote about machine learning, archived from MachineCurve.com.

A corpus of comments tagged for multiple attributes of unhealthiness.

A corpus of comments tagged for multiple attributes of unhealthiness.

BertViz: Visualize Attention in Transformer Models

Baseline model for multimodal classification based on images and text. Text representation obtained from pretrained BERT base model and image representation obtained from VGG16 pretrained model.

Understanding Deep Networks via Extremal Perturbations and Smooth Masks

Multimodal Meme Classification: Identifying Offensive Content in Image and Text

🏆 MedHacks 2020 Winner - 2nd Place, Best use of Google Cloud — Using ML to create communities and support groups for COVID patients with similar conditions 🏥.

Identifying offensive content in image and text

[NeurIPS'20-Competition] Detecting Hate Speech in Memes Using Multimodal Deep Learning Approaches: Prize-winning solution to Hateful Memes Challenge. https://arxiv.org/abs/2012.12975

Factify is a Multi-Modal Fact Verification dataset released for a shared task as part of the De-Factify workshop in AAAI-21.

Propaganda Detection in News articles.

Multimodal model for text and tabular data with HuggingFace transformers as building block for text data

A speaker embedding network in Pytorch that is very quick to set up and use for whatever purposes.

Notebooks using the Hugging Face libraries 🤗

🎯 Task-oriented embedding tuning for BERT, CLIP, etc.

Deep Learning - one shot learning for speaker recognition using Filter Banks

Encoding position with the word embeddings.

🤗 The largest hub of ready-to-use datasets for AI models with fast, easy-to-use and efficient data manipulation tools

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Self-Supervised Speech Pre-training and Representation Learning Toolkit

Speech emotion recognition models using fully-convolutional and convolutional-recurrent models

Neural network visualization toolkit for keras