EricLBuehler / candle-lora

Low rank adaptation (LoRA) for Candle.

EricLBuehler/candle-lora Issues

Could we have a written walkthrough of finetuning llama/mistral with this?
Updated 2 months ago1
In Llama model, only the embedding layer is converted to lora layer.
Updated 2 months ago2
Can lora also be implemented with stable diffusion?
Updated 3 months ago7
Updated candle-core, candle-nn [0.5.0] release breaks installation of candle-lora and candle-lora-macro dependencies
Closed 3 months ago
Is there any way to save lora-converted model?
Closed 3 months ago5
How to use canle_lora modle with rust auxm web server
Closed 4 months ago4
error[E0277]: expected a `Fn<(&candle_core::Tensor,)>` closure, found `BatchNorm`
Closed 5 months ago
Model Merging
Closed 5 months ago6
any example for llama_lora training
Closed 5 months ago2
replace_layer_fields and AutoLoraConvert not working as expected
Closed 7 months ago1
QA-LoRA Implementation and Review
Closed 9 months ago3
Add more LoRA transformers
Closed 8 months ago
Examples for Llama model architecture
Closed 10 months ago5
Question: Could we use the same mechanism for Quantization?
Closed 10 months ago4