feat: 'README'

README.md

@@ -0,0 +1,50 @@
+<div align="center">
+<img alt="Logo" src="docs/figures/FLAME-LOGO.png" width="80%"/>
+</div>
+
+
+
+ [![Python](https://img.shields.io/badge/Python-3.10%2B-blue)](https://www.python.org/)  [![PyTorch](https://img.shields.io/badge/PyTorch-2.4.1-blue)](https://pytorch.org/) 
+
+# FLAME: Flow Enhanced Legendre Memory Models for General Time Series Forecasting
+
+This is the official repository of **FLAME**: Flow Enhanced Legendre Memory Models for General Time Series Forecasting.
+
+
+
+## Introduction
+FLAME is a family of extremely **lightweight** and highly capable time series foundation models. Based on the normalization-based forecasting head, it can support both the **deterministic** and **probabilistic** forecasting.
+
+To our best knowldege, FLAME is the first time series foundation model possessing both lightweight backbones and generative prediction capabilities!
+
+
+
+<div align="center">
+<img alt="pic" src="docs/figures/intro.png" width="80%"/>
+</div>
+
+
+## Architecture
+
+FLAME adopts the Channel-Independent pretraining paradigm, and each variable is preprocessed through Instance Normalization to mitigate the value discrepancy. FLAME utilizes the Re-Norm to further mitigate the statistical differences between inputs and forecasts, and its backbone mainly consists of three modules: 1) Encoding, including Time Series Tokenization, **Local-Perception**, and MSA-Encoder, which tokenize the time series and enhance them through fusing the local environmental information with LegT; 2) Decoding, including **LegS based SSD-Decoder** and MCA-Enhancer, which utilize the SSD layers and MCA layers to make long-term inference ; 3) **Flow-based Head**, which leverages the Normalization Flow to support generative probabilistic forecasting, with both efficiency and accuracy. 
+
+<div align="center">
+<img alt="pic" src="docs/figures/overview.png" width="100%"/>
+</div>
+
+
+
+## Quickstart
+
+You need to install the following packages:
+
+```shell
+pip install transformers[torch]
+
+pip install mamba-ssm[causal-conv1d]
+
+pip install zuko
+```
+
+For mamba-ssm, there exists a bug about triton. We fix it and release the code at [./flame/mamba/mamba_ssm](./flame/mamba/mamba_ssm), you can compile it from scratch or simply replace the mamba_ssm package under the default path of conda.
+