add readme
This commit is contained in:
Generated
+1
-1
@@ -19,7 +19,7 @@ checksum = "320119579fcad9c21884f5c4861d16174d0e06250625266f50fe6898340abefa"
|
|||||||
|
|
||||||
[[package]]
|
[[package]]
|
||||||
name = "aha"
|
name = "aha"
|
||||||
version = "0.0.0"
|
version = "0.1.0"
|
||||||
dependencies = [
|
dependencies = [
|
||||||
"anyhow",
|
"anyhow",
|
||||||
"base64 0.22.1",
|
"base64 0.22.1",
|
||||||
|
|||||||
+2
-2
@@ -1,10 +1,10 @@
|
|||||||
[package]
|
[package]
|
||||||
name = "aha"
|
name = "aha"
|
||||||
version = "0.0.0"
|
version = "0.1.0"
|
||||||
edition = "2024"
|
edition = "2024"
|
||||||
repository = "https://github.com/jhqxxx/aha"
|
repository = "https://github.com/jhqxxx/aha"
|
||||||
license = "Apache-2.0"
|
license = "Apache-2.0"
|
||||||
description = "aha"
|
description = "aha model inference library, now supports Qwen2.5VL, MiniCPM4, and VoxCPM"
|
||||||
|
|
||||||
[dependencies]
|
[dependencies]
|
||||||
candle-core = { git = "https://github.com/huggingface/candle.git", version = "0.9.1"}
|
candle-core = { git = "https://github.com/huggingface/candle.git", version = "0.9.1"}
|
||||||
|
|||||||
@@ -0,0 +1,99 @@
|
|||||||
|
# aha
|
||||||
|
一个基于 Candle 框架的 Rust 模型推理库,提供高效、易用的多模态模型推理能力。
|
||||||
|
|
||||||
|
## 特性
|
||||||
|
* 🚀 高性能推理 - 基于 Candle 框架,提供高效的张量计算和模型推理
|
||||||
|
* 🎯 多模型支持 - 集成视觉、语言和多模态模型
|
||||||
|
* 🔧 易于使用 - 简洁的 API 设计,快速上手
|
||||||
|
* 🛡️ 内存安全 - 得益于 Rust 的所有权系统,确保内存安全
|
||||||
|
* 📦 轻量级 - 最小化依赖,编译产物小巧
|
||||||
|
* ⚡ GPU 加速 - 可选 CUDA 支持
|
||||||
|
* 🧠 注意力优化 - 可选 Flash Attention 支持,优化长序列处理
|
||||||
|
|
||||||
|
## 支持的模型
|
||||||
|
### 当前已实现
|
||||||
|
* Qwen2.5VL - 阿里通义千问 2.5 多模态大语言模型
|
||||||
|
* MiniCPM4 - 面壁智能 MiniCPM 系列语言模型
|
||||||
|
* VoxCPM - 面壁智能语音生成模型
|
||||||
|
|
||||||
|
## 计划支持
|
||||||
|
我们持续扩展支持的模型列表,欢迎贡献!
|
||||||
|
|
||||||
|
## 安装
|
||||||
|
### 作为库使用
|
||||||
|
* cargo add aha
|
||||||
|
* 或者在Cargo.toml中添加
|
||||||
|
```toml
|
||||||
|
[dependencies]
|
||||||
|
aha = { git = "https://github.com/jhqxxx/aha.git" }
|
||||||
|
|
||||||
|
# 启用 CUDA 支持(可选)
|
||||||
|
candle-inference = { git = "https://github.com/jhqxxx/aha.git", features = ["cuda"] }
|
||||||
|
|
||||||
|
# 启用Flash Attention 支持(可选)
|
||||||
|
candle-inference = { git = "https://github.com/your-username/your-repo.git", features = ["cuda", "flash-attn"] }
|
||||||
|
```
|
||||||
|
|
||||||
|
### 从源码构建运行测试
|
||||||
|
```bash
|
||||||
|
git clone https://github.com/jhqxxx/aha.git
|
||||||
|
cd aha
|
||||||
|
# 修改测试用例中模型路径
|
||||||
|
# 运行 Qwen2.5VL 示例
|
||||||
|
cargo test -F cuda,flash-attn qwen2_5vl_generate -- --nocapture
|
||||||
|
|
||||||
|
# 运行 MiniCPM4 示例
|
||||||
|
cargo test -F cuda,flash-attn minicpm_generate -- --nocapture
|
||||||
|
|
||||||
|
# 运行 VoxCPM 示例
|
||||||
|
cargo test -F cuda,flash-attn voxcpm_generate -- --nocapture
|
||||||
|
```
|
||||||
|
|
||||||
|
## 开发
|
||||||
|
### 项目结构
|
||||||
|
```text
|
||||||
|
.
|
||||||
|
├── Cargo.toml
|
||||||
|
├── README.md
|
||||||
|
├── src
|
||||||
|
│ ├── chat_template
|
||||||
|
│ ├── models
|
||||||
|
│ │ ├── common
|
||||||
|
│ │ ├── minicpm4
|
||||||
|
│ │ ├── qwen2_5vl
|
||||||
|
│ │ ├── voxcpm
|
||||||
|
│ │ └── mod.rs
|
||||||
|
│ ├── position_embed
|
||||||
|
│ ├── tokenizer
|
||||||
|
│ ├── utils
|
||||||
|
│ └── lib.rs
|
||||||
|
└── tests
|
||||||
|
├── test_minicpm4.rs
|
||||||
|
├── test_qwen2_5vl.rs
|
||||||
|
└── test_voxcpm.rs
|
||||||
|
```
|
||||||
|
|
||||||
|
### 添加新模型
|
||||||
|
* 在src/models/创建新模型文件
|
||||||
|
* 在src/models/mod.rs中导出
|
||||||
|
* 在tests/中添加测试和示例
|
||||||
|
|
||||||
|
## 许可证
|
||||||
|
本项目采用 Apache License, Version 2.0 许可证 - 查看 [LICENSE](./LICENSE) 文件了解详情。
|
||||||
|
|
||||||
|
## 致谢
|
||||||
|
* [Candle](https://github.com/huggingface/candle) - 优秀的 Rust 机器学习框架
|
||||||
|
* 所有模型的原作者和贡献者
|
||||||
|
|
||||||
|
## 支持
|
||||||
|
#### 如果你遇到问题:
|
||||||
|
1. 查看 Issues 是否已有解决方案
|
||||||
|
2. 提交新的 Issue,包含详细描述和复现步骤
|
||||||
|
|
||||||
|
## 更新日志
|
||||||
|
### v0.1.0
|
||||||
|
* 初始版本发布
|
||||||
|
* 支持 Qwen2.5VL, MiniCPM4, VoxCPM 模型
|
||||||
|
|
||||||
|
|
||||||
|
⭐ 如果这个项目对你有帮助,请给我们一个 Star!
|
||||||
Binary file not shown.
|
Before Width: | Height: | Size: 62 KiB |
@@ -1,6 +1,6 @@
|
|||||||
use crate::{
|
use crate::{
|
||||||
models::{
|
models::{
|
||||||
base_modules::{AttentionNobias, MLPNoBias},
|
common::{AttentionNobias, MLPNoBias},
|
||||||
minicpm4::config::MiniCPM4Config,
|
minicpm4::config::MiniCPM4Config,
|
||||||
},
|
},
|
||||||
position_embed::rope::compute_default_rope_parameters,
|
position_embed::rope::compute_default_rope_parameters,
|
||||||
@@ -155,7 +155,7 @@ impl MiniCPMDecoderLayer {
|
|||||||
+ xs.affine(
|
+ xs.affine(
|
||||||
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
||||||
0.0,
|
0.0,
|
||||||
))?;
|
)?)?;
|
||||||
Ok(xs)
|
Ok(xs)
|
||||||
}
|
}
|
||||||
|
|
||||||
@@ -180,7 +180,7 @@ impl MiniCPMDecoderLayer {
|
|||||||
+ xs.affine(
|
+ xs.affine(
|
||||||
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
||||||
0.0,
|
0.0,
|
||||||
))?;
|
)?)?;
|
||||||
Ok(xs)
|
Ok(xs)
|
||||||
}
|
}
|
||||||
pub fn clear_kv_cache(&mut self) {
|
pub fn clear_kv_cache(&mut self) {
|
||||||
|
|||||||
+1
-1
@@ -1,4 +1,4 @@
|
|||||||
pub mod base_modules;
|
pub mod common;
|
||||||
pub mod minicpm4;
|
pub mod minicpm4;
|
||||||
pub mod qwen2_5vl;
|
pub mod qwen2_5vl;
|
||||||
pub mod voxcpm;
|
pub mod voxcpm;
|
||||||
|
|||||||
@@ -1,7 +1,7 @@
|
|||||||
|
|
||||||
use crate::{
|
use crate::{
|
||||||
models::{
|
models::{
|
||||||
base_modules::{AttentionNobias, MLPNoBias},
|
common::{AttentionNobias, MLPNoBias},
|
||||||
voxcpm::config::VoxMiniCPM4Config,
|
voxcpm::config::VoxMiniCPM4Config,
|
||||||
},
|
},
|
||||||
position_embed::rope::compute_default_rope_parameters,
|
position_embed::rope::compute_default_rope_parameters,
|
||||||
@@ -193,7 +193,7 @@ impl MiniCPMDecoderLayer {
|
|||||||
+ xs.affine(
|
+ xs.affine(
|
||||||
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
||||||
0.0,
|
0.0,
|
||||||
))?;
|
)?)?;
|
||||||
res_add
|
res_add
|
||||||
} else {
|
} else {
|
||||||
let res_add = (residual + xs)?;
|
let res_add = (residual + xs)?;
|
||||||
@@ -206,7 +206,7 @@ impl MiniCPMDecoderLayer {
|
|||||||
+ xs.affine(
|
+ xs.affine(
|
||||||
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
self.scale_depth as f64 / (self.num_hidden_layers as f64).sqrt(),
|
||||||
0.0,
|
0.0,
|
||||||
))?;
|
)?)?;
|
||||||
res_add
|
res_add
|
||||||
} else {
|
} else {
|
||||||
let res_add = (residual + xs)?;
|
let res_add = (residual + xs)?;
|
||||||
|
|||||||
@@ -1,18 +1,14 @@
|
|||||||
use anyhow::{Ok, Result};
|
use anyhow::{Ok, Result};
|
||||||
use std::{collections::HashMap, time::Instant};
|
use std::{time::Instant};
|
||||||
|
|
||||||
use aha::{
|
use aha::{
|
||||||
models::voxcpm::{
|
models::voxcpm::{ generate::VoxCPMGenerate,
|
||||||
audio_vae::AudioVAE, config::VoxCPMConfig, generate::VoxCPMGenerate, model::VoxCPMModel,
|
|
||||||
tokenizer::SingleChineseTokenizer,
|
tokenizer::SingleChineseTokenizer,
|
||||||
},
|
},
|
||||||
utils::{
|
utils::{
|
||||||
audio_utils::save_wav,
|
audio_utils::save_wav,
|
||||||
utils::{find_type_files, get_device},
|
|
||||||
},
|
},
|
||||||
};
|
};
|
||||||
use candle_core::pickle::read_all_with_key;
|
|
||||||
use candle_nn::VarBuilder;
|
|
||||||
|
|
||||||
#[test]
|
#[test]
|
||||||
fn voxcpm_generate() -> Result<()> {
|
fn voxcpm_generate() -> Result<()> {
|
||||||
|
|||||||
Reference in New Issue
Block a user