TRBoom

Yes, a Convolutional Neural Net could do it or a plain Neural Net even.

You’ll want to create a sample of your handwriting with the letters isolated into their own picture and labelled. Maybe 10 of each letter to start with? If you get bad results with that make more samples. Each picture should be the same size as all the others. The starter course of Machine Learning I took way back when had us using a database of labeled numbers, each picture was 10 pixels by 10 pixels.

Then pick a CNN model (or better yet several) and train them on your handwriting. You can find some here: https://huggingface.co/models?other=CNN

Pick the one that does best. As part of that course I mentioned, I created an evolutionary algorithm to mutate, combine, and propagate CNNs to find out the best configurations for identifying images. The ones that performed the best got to combine with other top performers.

You might also be able to find a CNN specific to handwriting and then fine tune it to yours with your samples.

This is doing it raw and will have a lot of education for you along the way. There may be some prebuilt handwriting model you can just fine tune with easy instructions from the person who made it all wrapped up into a nice bit of python for you. Maybe.

Yup! If you installed the os on it.

So you have one usb with the iso flashed to it and a second to install the os on. Use the first to install to the second.

So a trick for the double drives is to pop in a low profile usb drive and install the os on that. Then you can use the ssd/hdd for other things.

Here’s mine, no judging for how I set the ip

name: jellyfin
services:
    jellyfin:
        container_name: jellyfin.live
        networks:
            mynet:
                ipv4_address: 192.168.5.3
        ports:
            - 8096:8096
        volumes:
            - jellyfin-config:/config
            - jellyfin-cache:/cache
            - type: bind
              source: /mnt
              target: /Media Disk 1
            - type: bind
              source: /mnt
              target: /Media Disk 2
            - type: bind
              source: /mnt
              target: /Media Disk 3
        environment:
            - HEALTHCHECK_URL=192.168.5.3
        restart: unless-stopped
        deploy:
            resources:
                reservations:
                    devices:
                        - driver: nvidia
                          count: all
                          capabilities:
                              - gpu
        image: jellyfin/jellyfin:latest
networks:
    mynet:
        external: true
        name: mynet
volumes:
    jellyfin-config:
        external: true
        name: jellyfin-config
    jellyfin-cache:
        external: true
        name: jellyfin-cache

The section on the GPU looks like yours. I had to do something to get it working, but it’s been forever.