aprender-train-0.41.0
- aprender-train 0.41.0
- Permalink
- Docs.rs crate page
- MIT
- Links
- Repository
- crates.io
- Source
- Owners
- noahgift
- Dependencies
- aprender-core ^0.41.0 normal
  arrow ^57 normal optional
  axum ^0.8 normal optional
  bytemuck ^1.19 normal
  chrono ^0.4 normal
  clap ^4.5 normal
  clap_complete ^4.5 normal
  ctrlc ^3.4 normal
  dhat ^0.3 normal optional
  dirs ^5.0 normal
  ed25519-dalek ^2.1 normal
  fs4 ^0.13 normal
  getrandom ^0.3 normal optional
  half ^2.4 normal
  hex ^0.4 normal
  hostname ^0.4.2 normal
  js-sys ^0.3 normal optional
  jsonschema ^0.28 normal
  ndarray ^0.16 normal
  nvml-wrapper ^0.10 normal optional
  presentar-core ^0.3.4 normal
  presentar-terminal ^0.3.5 normal optional
  provable-contracts-macros ^0.2 normal
  rand ^0.9 normal
  realizar ^0.8 normal optional
  regex ^1.10 normal
  aprender-profile ^0.41.0 normal optional
  rusqlite ^0.32 normal
  safetensors ^0.7 normal
  schemars ^0.8 normal
  serde ^1.0 normal
  serde_json ^1.0 normal
  serde_yaml ^0.9 normal
  sha2 ^0.10 normal
  thiserror ^2.0 normal
  tokio ^1.41 normal optional
  toml ^0.8 normal
  tower ^0.5 normal optional
  tower-http ^0.6 normal optional
  trueno ^0.17 normal
  trueno-db ^0.3.16 normal optional
  aprender-gpu ^0.41.0 normal optional
  trueno-rag ^0.2.0 normal optional
  trueno-viz ^0.2 normal
  unicode-normalization ^0.1 normal
  uuid ^1.11 normal optional
  validator ^0.20 normal
  wasm-bindgen ^0.2 normal optional
  approx ^0.5 dev
  arrow ^57 dev
  criterion ^0.7 dev
  dirs ^5.0 dev
  insta ^1.42 dev
  jugar-probar ^1.0 dev
  parquet ^57 dev
  proptest ^1.4 dev
  provable-contracts ^0.2 dev
  tempfile ^3.8 dev
  serde ^1 build
  serde_yaml_ng ^0.10 build
  alimentar ^0.2.4 normal optional
  hf-hub ^0.4 normal optional
  reqwest ^0.12 normal optional
  zip ^2.2 normal
- Versions
- 100% of the crate is documented
Platform
- x86_64-unknown-linux-gnu
Feature flags

docs.rs

Rust

Skip to main content

ste_backward

entrenar0.41.0

In entrenar::quant

entrenar::quant

Function ste_backward

pub fn ste_backward(grad_output: &Tensor) -> Tensor

Expand description

Convenience function for STE backward pass