Skip to main content

load_vec

Function load_vec 

Source
pub fn load_vec(
    content: &Content,
    reader: &mut (impl Read + Seek),
    name: &str,
    device: &Device,
) -> Result<Vec<f32>>
Expand description

Dequantize a named GGUF tensor to F32 and flatten it to a plain Vec<f32>.