Running On Windows

[grayerbeard]

Since development was on a Mac I found issues that I imagine most other would find routine to deal with. However it was a new experiance to me so with some help from Gemeni Pro this was my strategy.

My method based on a conversation with Gemeni Pro is as follwos. (This is for on Windows)

Enter The Following Commands which is generally as listed in README.md

python -m venv myLlmTestenv
 myLlmTestenv\Scripts\activate
 python.exe -m pip install --upgrade pip
pip install -r requirements.txt

Also I found because I am in Windows and Michael uses a Mac there is an issue ref Windows using CRLF and MAC using just Line feed.

So enter this command to set this repository to use LF like waht one does OnaMac.

git config core.autocrlf input 

Now adjust the screen size in config.yaml to significantly less than the size of your screem and the LLM parameters to what you prefer/Require. In my case I choose:

base_url: 'http://localhost:1234/v1' api_key: 'not-needed' questions_file: 'LLM Test.md' system_prompt: 'You are an intelligent assistant. You always provide well-reasoned answers that are both correct and helpful.' temperature: 0.2 font_size: 18 window_size: [1800, 900]

Now install and set up LM Studio and set up the local Inference server.

Then Make any changes required to the questions in the "LLM Test.md".

Then run tester with

python main.py

[SystemSculpt]

thanks so much for this update, I'll look into it. I'll try getting my hands on a virtual windows machine, maybe through parallels, to diagnose/fix up this problem; hopefully in the meantime this issue helps others trying to do the same

[grayerbeard]

Thanks. I added/refined a little to my notes a little see https://github.com/grayerbeard/myLlmTest/blob/main/MyNote.MD.

DT

On 16 Mar 2024, 18:17, at 18:17, SystemSculpt @.***> wrote:

thanks so much for this update, I'll look into it. I'll try getting my hands on a virtual windows machine, maybe through parallels, to diagnose/fix up this problem; hopefully in the meantime this issue helps others trying to do the same

-- Reply to this email directly or view it on GitHub: https://github.com/SystemSculpt/obsidian_llm_testing/issues/2#issuecomment-2002034985 You are receiving this because you authored the thread.

Message ID: @.***>