Training personalized model

[emjay73]

Although the instructions state that we can train our own personalized model using Lora as follows,

How to use it without any coding
Get lora models: train lora model with [A1111](https://github.com/continue-revolution/sd-webui-animatediff) based on a collection of your own favorite images (e.g., tutorials [English](https://www.youtube.com/watch?v=mfaqqL5yOO4), [Japanese](https://www.youtube.com/watch?v=N1tXVR9lplM), [Chinese](https://www.bilibili.com/video/BV1fs4y1x7p2/)) or download Lora models from [Civitai](https://civitai.com/).

Animate lora models: using gradio interface or A1111 (e.g., tutorials [English](https://github.com/continue-revolution/sd-webui-animatediff), [Japanese](https://www.youtube.com/watch?v=zss3xbtvOWw), [Chinese](https://941ai.com/sd-animatediff-webui-1203.html))

Be creative togther with other techniques, such as, super resolution, frame interpolation, music generation, etc.

Table 1 in the paper shows that ToonYou is trained using Dreambooth. I wonder how I can train a model like ToonYou on my own to utilize it.

Thank you in advance.

[rkfg]

I read the paper, it seems they refer to the base SD models and regular Loras, not the motion module. So this question isn't related to AnimateDiff at all, there are many guides about training models and loras, try kohya_ss for example.

From this issue title I thought it'd be about training your own motion module, but this procedure is described in the readme. I wonder if it'd be possible to make a motion module Lora, to only slightly finetune the base MM that already knows what "motion" in general is but doesn't understand certain words or situations. If it can be done with just a few videos it'd be great! I mean it works perfectly with a few images and base SD models so it probably should work with motions as well?