mirror of
https://github.com/huggingface/candle.git
synced 2025-06-19 03:54:56 +00:00
Metal quantized modifications proposal.
- Add a device param, wherever needed. - Create new QMetal storage thing that implements QuantizedType. - Update everywhere needed. Fix Python. Fixing examples. Fix: fmt + clippy + stub. Moving everything around. Only missing the actual implems. Fixing everything + adding dequantized kernels. More work. Fixing matmul. Fmt + Clippy Some clippy fixes. Working state. Q2K Metal -> Bugged (also present in GGML). Q4K CPU -> Bugged (present previously, new test catch it). Q5K CPU -> Bugged (present previously). Q8_1 Both -> Never really implemented it seems Q8K metal -> Never implemented in metal Fixing Q2K bug (present in ggml).
This commit is contained in:

committed by
Nicolas Patry

parent
3a7304cb0d
commit
f97fcd4712
@ -132,7 +132,8 @@ impl T5ModelBuilder {
|
||||
}
|
||||
|
||||
pub fn build_model(&self) -> Result<t5::T5ForConditionalGeneration> {
|
||||
let vb = t5::VarBuilder::from_gguf(&self.weights_filename)?;
|
||||
let device = Device::Cpu;
|
||||
let vb = t5::VarBuilder::from_gguf(&self.weights_filename, &device)?;
|
||||
Ok(t5::T5ForConditionalGeneration::load(vb, &self.config)?)
|
||||
}
|
||||
|
||||
|
Reference in New Issue
Block a user