Here comes a new challenger... And his name is DeepSeek!
in Review on Llms, Ai, Deepseek, Deepseek-r1, Deepseek-r1-zero, Gpt, Llama, Cot, Distillation
review DeepSeek-R1 paper.
Simple Image Similarity
in Works on Simple, Python, Image similarity, Opencv, Imagehash
A simple Python app for retrieving the closest image by calculating image similarity
Development log
Simple Activity Trakcer
in Works on Simple, Python, Object detection, Opencv, Yolo
A simple Python webcam app for tracking position and evaluating the activity level of a subject based on YOLOv5
Development log
Image Generation and LoRA Service with Flux
in Works on Python, Flux, Api, Fine-tuning, Black forest labs, Replit
A web-based image generation and fine-tuning LoRA service with Black Forest Labs’ Flux model
Development log
Diffusion Drawing
in Works on Python, Diffusion model, Stable diffusion, Webui-forge, Docker, Aws, Pyqt
A Python drawing app with Stable Diffusion Webui Forge using Docker and a web server like an AWS instance
Development log
A Character Created Using ChatGPT And ElevenLabs
in Works on Python, Tts, Stt, Chatgpt, Elevenlabs, Fine-tuning, Frieren
A text- and voice-interactable character that embodies a fictional character’s personality and voice, created using ChatGPT assistants feature and ElevenLabs fine-tuned TTS
Development log
Try Fine-Tuning LLMs at Home
in Practice on Fine-tuning, Llms, Ai, Unsloth, Continued pre-training, Lora
practice fine-tuning with LLaMA.
figure1: A cute lama, generated by Flux 1.1 pro
High-Resolution Image Synthesis With Latent Diffusion Models
in Review
review Stable Diffusion.
In the image synthesis domain, diffusion models had shown very effective manipulation of images. However the models require abundant GPU calculations since they are operated in pixel space. To conserve computational resource while preserving flexibility and quality, Stable Diffusion (2022) [1] introduced latent space into the process.
Large Language Models before ChatGPT 3.5
in Summary
summarize the papers of GPT1, BERT, GPT3 and the scaling laws.
After the publication of “Attention Is All You Need”, a number of language models have chosen the transformer as their base structure. As a result, many powerful models have been derived from it, the most famous of which are the GPT models and BERT.
Denoising Diffusion Probabilistic Models
in Review
review DDPM.
Generative deep learning models like VAEs and GANs have shown brilliant performance. “Denoising Diffusion Probabilistic Models” (DDPM) [1] is novel generative model published in 2020. This model is based on “Deep Unsupervised Learning Using Nonequilibrium Thermodynamics (2015)”.