Upload README.md

README.md

@@ -1,76 +1,80 @@
 <div align="center">
-  <img src="https://raw.githubusercontent.com/limix-ldm/LimiX/refs/heads/main/doc/LimiX-Logo.png" alt="LimiX summary" width="89%">
+  <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/LimiX-Logo.png" alt="LimiX summary" width="89%">
 </div>
 
-# News :boom: 
- - 2025-08-29: LimiX V1.0 Released.
- - 2025-11-10: LimiX-2M is officially released! Compared to LimiX-16M, this smaller variant offers significantly lower GPU memory usage and faster inference speed. The retrieval mechanism has also been enhanced, further improving model performance while reducing both inference time and memory consumption.
+#  :boom:  最新进展
+ - 2025-11-10: LimiX-2M 正式发布！与 LimiX-16M 相比，此版本在显著降低 GPU 内存占用的同时，实现了更快的推理速度。在此基础上，LimiX 的样本检索机制也得到了优化，模型的性能进一步提升，并有效减少了该模式下的推理时间与显存消耗
+ - 2025-08-29: LimiX V1.0 发布
+# ⚡ 最新评测
+<div align="center">
+  <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/BCCO-CLS.png"  width="30%">
+  <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/TabArena-CLS.png"  width="30%">
+  <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/TabZilla-CLS.png" width="30%">  
+</div>
+<div align="center">
+  <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/BCCO-REG.png"  width="30%">
+  <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/TabArena-REG.png" width="30%">
+  <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/CTR23-REG.png" width="30%">
+</div>
 
-# ➤ Overview
+# ➤ 简介
 <div align="center">
   <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/LimiX_Summary.png" alt="LimiX summary" width="89%">
 </div>
-We introduce LimiX, the first installment of our LDM series. LimiX aims to push generality further: a single model that handles classification, regression, missing-value imputation, feature selection, sample selection, and causal inference under one training and inference recipe, advancing the shift from bespoke pipelines to unified, foundation-style tabular learning.
+我们推出 LDM 系列的首个模型—极数。极数的目标在于进一步提升通用性：在统一的训练与推理框架下，同时处理分类、回归、缺失值插补、特征选择、样本选择和因果推理等任务，从而推动表格学习从定制化流程迈向基础模型范式的转变。
 
-LimiX adopts a transformer architecture optimized for structured data modeling and task generalization. The model first embeds features X and targets Y from the prior knowledge base into token representations. Within the core modules, attention mechanisms are applied across both sample and feature dimensions to identify salient patterns in key samples and features. The resulting high-dimensional representations are then passed to regression and classification heads, enabling the model to support diverse predictive tasks. 
+极数基于专为结构化数据建模和任务泛化优化的 Transformer 架构。 模型首先将先验知识库中的特征 𝑋  与目标 𝑌 映射为 token 表示。在核心模块中，注意力机制同时作用于样本维度和特征维度，以捕捉关键样本与特征的显著模式。随后，高维表示被送入回归与分类模块，从而支持多种任务的输出。
 
-For details, please refer to the technical report at the link: [LimiX:Unleashing Structured-Data Modeling Capability for Generalist Intelligence](https://arxiv.org/abs/2509.03505) or [LimiX_Technical_Report.pdf](https://github.com/limix-ldm/LimiX/blob/main/LimiX_Technical_Report.pdf).
+技术报告详见：[LimiX:Unleashing Structured-Data Modeling Capability for Generalist Intelligence](https://arxiv.org/abs/2509.03505) / [LimiX_Technical_Report.pdf](https://github.com/limix-ldm/LimiX/blob/main/LimiX_Technical_Report.pdf)
 
-# ➤ Superior Performance 
-The LimiX model achieved SOTA performance across multiple tasks.
-
-## ➩ Classification
+# ➤ 对比测试
+LimiX模型在多个任务的测试中达到了当前最优性能.
+## ➩ 分类
 <div align="center">
   <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/Classifier.png" alt="Classification" width="80%">
 </div>
 
-## ➩ Regression
+## ➩ 回归
 <div align="center">
   <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/Regression.png" alt="Regression" width="60%">
 </div>
 
-## ➩ Missing Values Imputation
+## ➩ 缺失值插补
 <div align="center">
   <img src="https://github.com/limix-ldm/LimiX/raw/main/doc/MissingValueImputation.png" alt="Missing value imputation" width="80%">
 </div>
 
-# ➤ Tutorials 
-## ➩ Installation
-### Option 1 (recommended): Use the Dockerfile
-Download [Dockerfile](https://github.com/limix-ldm/LimiX/blob/main/Dockerfile)
+
+# ➤ 教程
+## ➩ 安装
+### 方式1，使用Dockerfile (建议)
+下载[Dockerfile](https://github.com/limix-ldm/LimiX/blob/main/Dockerfile)
 ```bash
 docker build --network=host -t limix/infe:v1 --build-arg FROM_IMAGES=nvidia/cuda:12.2.0-base-ubuntu22.04 -f Dockerfile .
 ```
-
-### Option 2: Build manually
-Download the prebuilt flash_attn files
+### 方式2，自行构建环境
+下载 flash_attn 预编译文件
 ```bash
 wget -O flash_attn-2.8.0.post2+cu12torch2.7cxx11abiTRUE-cp312-cp312-linux_x86_64.whl https://github.com/Dao-AILab/flash-attention/releases/download/v2.8.0.post2/flash_attn-2.8.0.post2+cu12torch2.7cxx11abiTRUE-cp312-cp312-linux_x86_64.whl
 ```
-Install Python dependencies
+环境安装
 ```bash
 pip install python==3.12.7 torch==2.7.1 torchvision==0.22.1 torchaudio==2.7.1
 pip install flash_attn-2.8.0.post2+cu12torch2.7cxx11abiTRUE-cp312-cp312-linux_x86_64.whl
 pip install scikit-learn  einops  huggingface-hub matplotlib networkx numpy pandas  scipy tqdm typing_extensions xgboost kditransform hyperopt
 ```
 
-### Download source code
-```bash
-git clone https://github.com/limix-ldm/LimiX.git
-cd LimiX
-```
-
-# ➤ Inference
-LimiX supports tasks such as classification, regression, and missing value imputation
-## ➩ Model download
-| Model size | Download link | Tasks supported |
+# ➤ 推理任务
+LimiX支持分类、回归、缺失值插补等任务。
+## ➩ 模型下载
+| 模型尺寸 | 下载链接 | 支持的任务 |
 | --- | --- | --- |
-| LimiX-16M | [LimiX-16M.ckpt](https://huggingface.co/stableai-org/LimiX-16M/tree/main) |  ✅ classification  ✅regression   ✅missing value imputation |
-| LimiX-2M | [LimiX-2M.ckpt](https://huggingface.co/stableai-org/LimiX-2M/tree/main) |  ✅ classification  ✅regression   ✅missing value imputation |
+|  LimiX-16M | [LimiX-16M.ckpt](https://huggingface.co/stableai-org/LimiX-16M/tree/main) |  ✅ 分类  ✅回归   ✅缺失值插补 |
+|  LimiX-2M | [LimiX-2M.ckpt](https://huggingface.co/stableai-org/LimiX-2M/tree/main) |  ✅ 分类  ✅回归   ✅缺失值插补 |
 
-## ➩ Interface description
 
-### Model Creation
+## ➩ 接口说明
+### 创建模型
 ```python
 class LimiXPredictor:
     def __init__(self,
@@ -86,108 +90,97 @@ class LimiXPredictor:
                  inference_with_DDP: bool = False,
                  seed:int=0)
 ```
-| Parameter | Data Type | Description |
+| 参数名 | 数据类型 | 参数说明 |
 |--------|----------|----------|
-| device | torch.device | The hardware that loads the model |
-| model_path | str | The path to the model that needs to be loaded |
-| mix_precision | bool | Whether to enable the mixed precision inference |
-| inference_config | list/str | Configuration file used for inference |
-| categorical_features_indices | list | The indices of categorical columns in the tabular data |
-| outlier_remove_std | float | The threshold is employed to remove outliers, defined as values that are multiples of the standard deviation |
-| softmax_temperature | float | The temperature used to control the behavior of softmax operator |
-| task_type | str | The task type which can be either "Classification" or "Regression" |
-| mask_prediction | bool | Whether to enable missing value imputation |
-| inference_with_DDP | bool | Whether to enable DDP during inference |
-| seed | int | The seed to control random states |
-### Predict
+| device | torch.device | 运行模型的硬件 |
+| model_path | str | 需要加载的模型的路径 |
+| mix_precision | bool | 是否启动混合进度计算 |
+| inference_config | list/str | 推理使用的配置文件 |
+| categorical_features_indices | list | 数据表中，分类列的序号 |
+| outlier_remove_std | float | 移除异常值时采用的标准差倍数阈值 |
+| softmax_temperature | float | Softmax 温度 或 温度系数 |
+| task_type | str | 任务类型，取值范围为：Classification, Regression |
+| mask_prediction | bool | 是否启用缺失值插补功能 |
+| inference_with_DDP | bool | 在推理时是否开启DDP |
+| seed | int | 随机种子 |
+### 数据推理
 ```python
 def predict(self, x_train:np.ndarray, y_train:np.ndarray, x_test:np.ndarray) -> np.ndarray:
 ```
-| Parameter   | Data Type    | Description           |
+| 参数名   | 数据类型    | 参数说明           |
 | ------- | ---------- | ----------------- |
-| x_train  | np.ndarray  | The input features of the training set   |
-| y_train  | np.ndarray  | The target variable of the training set   |
-| x_test   | np.ndarray  | The input features of the test set   |
+| x_train  | np.ndarray  | 训练集的 feature   |
+| y_train  | np.ndarray  | 训练集的预测目标   |
+| x_test   | np.ndarray  | 测试集的 feature   |
 
-## Inference Configuration File Description
-| Configuration File Name | Description | Difference |
+## ➩ 推理配置文件说明
+| 配置文件名称 | 配置文件说明 | 差异 |
 | ------- | ---------- | ----- |
-| cls_default_retrieval.json | Default **classification task** inference configuration file **with retrieval** | Better classification performance |
-| cls_default_noretrieval.json | Default **classification task** inference configuration file **without retrieval** | Faster speed, lower memory requirements |
-| reg_default_retrieval.json | Default **regression task** inference configuration file **with retrieval** | Better regression performance |
-| reg_default_noretrieval.json | Default **regression task** inference configuration file **without retrieval** | Faster speed, lower memory requirements |
-| reg_default_noretrieval_MVI.json | Default inference configuration file for **missing value imputation task** |  |
-
-## ➩ Ensemble Inference Based on Sample Retrieval
-
-For a detailed technical introduction to Ensemble Inference Based on Sample Retrieval, please refer to the [technical report](https://github.com/limix-ldm/LimiX/blob/main/LimiX_Technical_Report.pdf).
-
-Considering inference speed and memory requirements, ensemble inference based on sample retrieval currently only supports hardware with specifications higher than the NVIDIA RTX 4090 GPU.
-
-### Classification Task
-
+| cls_default_retrieval.json | 默认的**含有retrieval**功能的**分类任务**推理配置文件 | 分类性能更好 |
+| cls_default_noretrieval.json | 默认的**不含有retrieval**功能的**分类任务**推理配置文件 | 速度更快、显存需求更低 |
+| reg_default_retrieval.json | 默认的**含有retrieval**功能的**回归任务**推理配置文件 | 回归性能更好 |
+| reg_default_noretrieval.json | 默认的**不含有retrieval**功能的**回归任务**推理配置文件 | 速度更快、显存需求更低 |
+| reg_default_noretrieval_MVI.json | 默认的**缺失值插补**任务推理配置文件 |  |
+
+## ➩ 基于样本检索的ensemble推理
+基于样本检索的ensemble推理的详细技术描述详见[LimiX技术报告](https://github.com/limix-ldm/LimiX/blob/main/LimiX_Technical_Report.pdf)
+考虑到推理速度、显存占用，基于样本检索的ensemble推理目前只支持基于版本高于NVIDIA-RTX 4090显卡的硬件条件。
+### 分类任务
 ```
 python inference_classifier.py --save_name your_save_name --inference_config_path path_to_retrieval_config --data_dir path_to_data
 ```
 
-### Regression Task
-
+### 回归任务
 ```
 python inference_regression.py --save_name your_save_name --inference_config_path path_to_retrieval_config --data_dir path_to_data
 ```
 
-### Customizing Data Preprocessing for Inference Tasks
-#### First, Generate the Inference Configuration File
-
+### 个性化设置推理任务的数据预处理方式
+#### 首先生成inference_config文件
 ```python
-generate_inference_config()
+generate_infenerce_config()
 ```
 
-### Classification Task
-#### Single GPU or CPU
-
+### 分类任务
+#### 单卡或者CPU
 ```
 python  inference_classifier.py --save_name your_save_name --inference_config_path path_to_retrieval_config --data_dir path_to_data
 ```
-
-#### Multi-GPU Distributed Inference
-
+#### 多卡分布式推理
 ```
 torchrun --nproc_per_node=8  inference_classifier.py --save_name your_save_name --inference_config_path path_to_retrieval_config --data_dir path_to_data --inference_with_DDP
 ```
 
-### Regression Task
-#### Single GPU or CPU
-
+### 回归任务
+#### 单卡或者CPU
 ```
 python  inference_regression.py --save_name your_save_name --inference_config_path path_to_retrieval_config --data_dir path_to_data
 ```
-
-#### Multi-GPU Distributed Inference
-
+#### 多卡分布式推理
 ```
 torchrun --nproc_per_node=8  inference_regression.py --save_name your_save_name --inference_config_path path_to_retrieval_config --data_dir path_to_data --inference_with_DDP
 ```
 
-### Retrieval Optimization Project
-This project implements an optimized retrieval system. To achieve the best performance, we utilize Optuna for hyperparameter tuning of retrieval parameters.
-#### Installation
-Ensure you have the required dependencies installed:
+### 样本检索的超参检索
+本项目提供了一个样本检测的超参搜索套件。为了获得最佳性能，我们使用 Optuna 对检索参数进行超参数优化
+#### 安装
 ```
 pip install optuna
 ```
-#### Usage
-For standard inference using pre-optimized parameters, refer to the code below:
+#### 使用方法
+若要使用超参数优化进行标准推理，请参考如下代码，这段代码将启动一个 Optuna 搜索，用于为特定数据集和应用场景寻找最佳的检索参数组合：
 ```
 searchInference = RetrievalSearchHyperparameters(
-           dict(device_id=0,model_path=model_path), X_train, y_train, X_test, y_test,
+           dict(device_id=0, model_path=model_path), X_train, y_train, X_test, y_test,
 )
-config, result = searchInference.search(n_trials=10, metric="AUC",
-              inference_config='config/cls_default_retrieval.json',task_type="cls")
+config, result = searchInference.search(
+              n_trials=10, metric="AUC",
+              inference_config='config/cls_default_retrieval.json',
+              task_type="cls")
 ```
-This will launch an Optuna study to find the best combination of retrieval parameters for your specific dataset and use case.
 
-## ➩ Classification
+## ➩ 分类
+
 ```python
 from sklearn.datasets import load_breast_cancer
 from sklearn.metrics import accuracy_score, roc_auc_score
@@ -217,9 +210,9 @@ prediction = clf.predict(X_train, y_train, X_test)
 print("roc_auc_score:", roc_auc_score(y_test, prediction[:, 1]))
 print("accuracy_score:", accuracy_score(y_test, np.argmax(prediction, axis=1)))
 ```
-For additional examples, refer to [inference_classifier.py](./inference_classifier.py)
+更加详细的样例详见： [inference_classifier.py](./inference_classifier.py)
 
-## ➩ Regression
+## ➩ 回归
 ```python
 from functools import partial
 
@@ -253,6 +246,7 @@ y_std = y_train.std()
 y_train_normalized = (y_train - y_mean) / y_std
 y_test_normalized = (y_test - y_mean) / y_std
 
+data_device = f'cuda:0'
 model_path = hf_hub_download(repo_id="stableai-org/LimiX-16M", filename="LimiX-16M.ckpt", local_dir="./cache")
 
 model = LimiXPredictor(device='cuda', model_path=model_path, inference_config='config/reg_default_retrieval.json')
@@ -266,22 +260,21 @@ r2 = r2_score(y_test_normalized, y_pred)
 print(f'RMSE: {rmse}')
 print(f'R2: {r2}')
 ```
-For additional examples, refer to [inference_regression.py](./inference_regression.py)
+更加详细的样例详见： [inference_regression.py](./inference_regression.py)
 
-## ➩ Missing value imputation
-For the demo file, see [examples/demo_missing_value_imputation.py](examples/inference_regression.py)
+## ➩ 缺失值插补
+样例详见： [examples/demo_missing_value_imputation.py](examples/inference_regression.py)
 
-# ➤ Link
- - LimiX:Unleashing Structured-Data Modeling Capability for Generalist Intelligence: [LimiX:Unleashing Structured-Data Modeling Capability for Generalist Intelligence](https://arxiv.org/abs/2509.03505)
- - LimiX Technical Report: [LimiX_Technical_Report.pdf](https://github.com/limix-ldm/LimiX/blob/main/LimiX_Technical_Report.pdf)
- - Detailed instructions for using Limix: [Visit the official Limix documentation](https://www.limix.ai/doc/)
- - Balance Comprehensive Challenging Omni-domain Classification Benchmark: [bcco_cls](https://huggingface.co/datasets/stableai-org/bcco_cls)
- - Balance Comprehensive Challenging Omni-domain Regression Benchmark: [bcco_reg](https://huggingface.co/datasets/stableai-org/bcco_reg)
+# ➤ 链接
+ - LimiX(极数)：结构化数据通用大模型：[LimiX:Unleashing Structured-Data Modeling Capability for Generalist Intelligence](https://arxiv.org/abs/2509.03505)
+ - LimiX技术报告：[LimiX_Technical_Report.pdf](https://github.com/limix-ldm/LimiX/blob/main/LimiX_Technical_Report.pdf)
+ - 平衡、全面、有挑战、跨领域的分类数据集：[bcco_cls](https://huggingface.co/datasets/stableai-org/bcco_cls)
+ - 平衡、全面、有挑战、跨领域的回归数据集：[bcco_reg](https://huggingface.co/datasets/stableai-org/bcco_reg)
 
-# ➤ License
-The code in this repository is open-sourced under the [Apache-2.0](LICENSE.txt) license, while the usage of the LimiX model weights is subject to the Model License. The LimiX weights are fully available for academic research and may be used commercially upon obtaining proper authorization.
+# ➤ 协议
+本仓库的代码依照 [Apache-2.0](LICENSE.txt) 协议开源，LimiX 模型的权重的使用则需要遵循 Model License。LimiX 权重对学术研究完全开放，在进行授权后允许商业使用。
 
-# ➤ Citation
+# ➤ 引用
 ```
 @article{LimiX,
   title={LimiX:Unleashing Structured-Data Modeling Capability for Generalist Intelligence},
@@ -289,4 +282,4 @@ The code in this repository is open-sourced under the [Apache-2.0](LICENSE.txt)
   journal={arXiv preprint arXiv:2509.03505},
   year={2025}
 }
-```
+```