ruda-nn 0.21.46

Ruda neural network layers, activation modules and losses.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52

mod decoder;
mod encoder;
mod pwff;
mod dense;
mod heads;
mod adapted_head;
mod embeddings;
mod adapters;
mod adapter_record;
mod adapted_stack;
mod adapted_decoder;
mod decoder_adapter_record;
mod packed;
mod packed_decoder;
mod cached;
mod cached_decoder;
mod awq;
mod awq_model;
mod projection;
mod projected_adapters;
mod projected_decoder;
mod projected_paired_model;
mod native_attention;
mod moe;
mod moe_model;
mod moe_adapters;
#[cfg(feature="tensor-parallel")]
mod expert_parallel;

pub use decoder::*;
pub use encoder::*;
pub use pwff::*;
pub use dense::*;
pub use heads::*;
pub use adapted_head::*;
pub use embeddings::*;
pub use adapters::*;
pub use adapter_record::*;
pub use adapted_stack::*;
pub use adapted_decoder::*;
pub use decoder_adapter_record::*;
pub use awq::*;
pub use awq_model::*;
pub use projection::*;
pub use projected_adapters::*;
pub use projected_decoder::*;
pub use projected_paired_model::*;
pub use moe::*;
pub use moe_model::*;
pub use moe_adapters::*;
#[cfg(feature="tensor-parallel")]
pub use expert_parallel::*;