Open quocanh34 opened 7 months ago
Are you using a single card for training? , when I use a single card for training, I get an error about insufficient storage space. Even though I still have a lot of storage space.
@wen020 yes, I use 4090 to train, may the training requires at least more than 24Gb I think, otherwise it causes oom.
The config of mine works on 4090, just switch to small model like 1B. The only problem I have is the lora result, it is bad.
I used this data(https://huggingface.co/dome272/stable-cascade/blob/main/fernando.tar) to finetuning the lora,i find the result is ok(the results are below). I immediately started training on my own data. prompt: cinematic photo of a dog [fernando] wearing a space suit
The results of LoRA are very bad with my custom dataset
@wen020 yes, especially on faces :(((
Hey, we never tried out the 1B model on LoRAs. We just used the 3.6B and I could only give feedback on this. The 1B model is very undertrained.
@dome272 Actually the 3B model trained on LoRAs is still bad, especially on faces. Have you ever tried to train faces with LoRA?
@quocanh34 I also has trained the lora by 4090. how long do your train 40000 steps? I will cost 3 hour.
The result training style lora is also bad.
@quocanh34 how to train a lora in 3.6B model with 4090(24Gb)?
@wen020 I can only train 1B model on 4090, otherwise it will cause oom.
I have trained LoRA several times with faces but it seems the model can't learn anything.
My dataset is as follow:
Here is my config: