The trec_webtrack from joaolages

trec_webtrack's Introduction

Hi there 👋

🙋‍♂️ My name is João Lages
👷‍ Deep Learning Engineer @ Revolut
🌱 I’m interested in everything about machine learning, with focus on deep learning applied to text, images, tabular data, video, speech, time-series, anything!
📫 How to reach me: [email protected]
✍️ Blog posts:
- Model Merging: MoE, Frankenmerging, SLERP, and Task Vector Algorithms 🧌 - Deep dive on how LLM merging methods work (co-authored with Deci AI)
- OpenAI JSON Mode vs Functions - Practical differences between these two ways of using OpenAI API
- Direct Preference Optimization (DPO) - A simplified explanation of the DPO algorithm applied to large language models, like Zephyr
- Reinforcement Learning from Human Feedback (RLHF) 🙋‍♂️ - A simplified explanation of the RLHF algorithm applied to large language models, like ChatGPT
- Transformers KV Caching Explained 💾 - A short writing on how Key and Value states are cached in transformers for faster inference
- Transformers Positional Encodings Explained 📝 - Positional encoding and how it limits the input size of language models
- Mahalanobis for outlier detection - A simple demo on how to use mahalanobis distance for outlier detection
⭐ Main open-source contributions:
- Diffusers-Interpret 🤗🧨🕵️‍♀️ - Own package, a model explainability tool built on top of 🤗 Diffusers
- Ecco - Major contributions to this package that is used to explain, analyze, and visualize NLP language models
- AI Reading Group - Co-author of an open AI reading group from 2019-2023
- RATransformers 🐭 - Own package, used to make transformer models relation-aware
- 🤗 datasets - implemented the mahalanobis distance metric

trec_webtrack's People

Contributors

Stargazers

Watchers

trec_webtrack's Issues

Running RE-PACRR on custom data

Hi, I was able to run the repacrr code provided by @khui and now I am trying to run repacrr on some custom data and see the results. It would be very helpful if you can provide some instructions (or any advice) for the same. For example, I understand how to generate the similarty matrices using word vectors but what I fail to understand is the matrix structure itself. i.e: should I keep the query's and document's word ordering or should I use idf sorting etc. Thank you.

Recommend Projects