Skip to content

Python

Self-Speculative Decoding Implementation: LayerSkip Model, Bayesian Optimization, and Adaptive Draft-Exiting Mechanism (Here are gemma-2-9b-it Experiment Results)

Over the past week, I dedicated some time to reproducing the Self-Speculative Decoding mechanism based on the ideas from the paper Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding, implementing the following modules:

  • A Decoder-only Transformer model with layer skipping (based on Llama and Gemma-2 architectures)
  • Adaptive Draft Exit Mechanism
  • Bayesian Optimization to discover the best layer-skipping strategy (optimizing draft model configurations)
  • Self-Speculative Decoding — achieving acceleration purely through the model itself
Read More »Self-Speculative Decoding Implementation: LayerSkip Model, Bayesian Optimization, and Adaptive Draft-Exiting Mechanism (Here are gemma-2-9b-it Experiment Results)

[Python] FastAPI Using Server-Sent Events (SSE) for Streaming Responses

I have recently set up numerous backend API servers for Chatbots. Initially, I received user messages and returned the entire LLM-generated reply in one go to the frontend interface. However, this approach did not provide a good user experience. I then switched to HTTP streaming, sending each generated token to the frontend as it was produced. Later, I found that some users' devices experienced packet sticking, so I finally switched to using WebSocket.

Read More »[Python] FastAPI Using Server-Sent Events (SSE) for Streaming Responses

Using Finite State Machine (FSM) and Rollback Mechanism to Restrict LLM from Generating Banned Words

When implementing various services through LLMs, do you worry about uncontrolled language generation? Recently, at a critical juncture in wrapping up a project, I used tools like Outlines to constrain LLM decoding, which effectively controlled the model's output to follow the desired patterns. However, a colleague posed a deep question: What if I want it not to generate specific words?

Read More »Using Finite State Machine (FSM) and Rollback Mechanism to Restrict LLM from Generating Banned Words

[Python] Using Locust Open Source Load Testing Framework for Stress Testing

Locust is an open-source load testing tool that helps simulate heavy user traffic on web applications and APIs. Compared to traditional load testing tools, Locust offers more customization and scalability—it supports Python as the scripting language, allowing us to write tests specific to our API or web service use cases.

Read More »[Python] Using Locust Open Source Load Testing Framework for Stress Testing

Notes on Fine-Tuning a Multi-Modal Large Language Model Using SFTTrainer (Taking LLaVa-1.5 as an Example)

A multi-modal large language model (Multi-Modal Large Language Model) isn’t limited to text only. I know this might sound contradictory, but this is a term that has become widely accepted. What I want to document today is how to fine-tune a multi-modal model using a script.

Read More »Notes on Fine-Tuning a Multi-Modal Large Language Model Using SFTTrainer (Taking LLaVa-1.5 as an Example)

[Python] Extracting Text from PPT Using the python-pptx Library

Introduction

Recently, while handling some work-related matters, I noticed that the client might potentially need a way to extract text from PPT files. I discussed this with the PM and my supervisor, and they mentioned that the client could simply copy the text from the PPT slides manually. Unless the client explicitly requests us to extract it programmatically.

Read More »[Python] Extracting Text from PPT Using the python-pptx Library

[Machine Learning] Vector Quantization (VQ) Notes

The first time I heard about Vector Quantization (VQ) was from a friend who was working on audio processing, which gave me a vague understanding that VQ is a technique used for data feature compression and representation. At that time, I still wasn't clear on how it differed from dimensionality reduction techniques like PCA.

Read More »[Machine Learning] Vector Quantization (VQ) Notes
Exit mobile version