So, I decided I wanted to have a local installation of TriplePlaneGaussian from github for research and experimentation. Ran into version conflicts, as usual; so I created a venv (python virtual environment). No big deal so far.Well, when going to install the appropriate versions of pyTorch and associated packages, (1.13.1+cu113); I couldn’t find them on
The following post is a collection of notes and links on SDUI usage and modification.Adding stabilityai/stable-video-diffusion-img2vid-xt to Automatic1111’s SDUI. See:After installing the appropriate virtual environment:and the required modules to allow forge to function, you need to git pull the svd model into the ./models/svd directory:Adding zero123 to forgeSo the SVD and z123 implementations works, now
This post is a place where I can store the output of queries from an LLM for further review and research; specifically validation of claims. This particular series centers around interest in using gene therapy techniques to achieve lasting effects similar to transient ones achieved from pharmaceutical agents. For example, ADD treatment often involves various
The principles of electrochemistry have long employed an electrical framework of understanding to characterize various chemical reactions (mostly oxidation and reduction reactions, tabulated in a so-called redox reaction list). However, it came so my attention from scroll addiction that MIT has recently discovered a way to employ applied voltage to reactions which do not fall
Regarding the underlying mechanisms of machine learning, matrix operations are the backbone of the mathematical framework in which all emergent phenomena arise. Something like a ‘mathematical aether’, this conceptual framework provides the necessary richness of context which patterns of a useful nature require to be defined. In contemporary technologies, dimension of the involved matrices is
In order to construct more interesting tools for monkeys to use, it is helpful to be able to integrate the modular tools you are already experimenting with. To that end…”Using the gradio_client library, we can easily use the Gradio as an API to transcribe audio files programmatically.Here’s the entire code to do it:The Gradio client works with
Using ollama, and the dolphin-2.5-mixtral-8x7b model a simple script to test TTS and requests from Ollama server is being constructed. These are some reference notes.Okay this was successful and resulted in a script called askbot.py which is run from a shell script called askbot.sh.Recording Audio for an undetermined timeThis is the next obvious step, recording
So, in experimenting with these neural network trained models, it has become apparent that either one must buy a lot of expensive individual GPU cards to achieve the core density and VRAM necessary for interesting experiments like training or local model services; or, one can set up a server with several slightly older model GPUs
This post proceeds as prompt, then output. Model used dolphin-mixtral:latest in ollama, no-gpu.Vimanas>>> describe the mercury vortex engine of the vimana and how to construct one with modern methodsThe Mercury Vortex Engine (MVE) is a hypothetical propulsion system mentioned in the ancient Indian texts called “Vimana Shastras.” These texts describe advanced aerial vehicles with seemingly
This post is to keep track of the Loras obtained from civit.ai facilitating the generation of various types of spaceships. You can follow prompts as well, WebUI used was Fooocus; model checkpoint is SDXL 2.The Black Squid ship was generated using prompts related to black metal squid star fighter, black metal squid spaceship, with the