update models information

Author: Kazeya27

docs/_sources/user_guide/model.md.txt

@@ -311,6 +311,14 @@ For trajectory next-location prediction:
   Wang S, Wu H, Shi X, et al. Timemixer: Decomposable multiscale mixing for time series forecasting[J]. arXiv preprint arXiv:2405.14616, 2024.
   ```
 
+- **FreTS**
+
+  This paper introduces a new model named FreTS (Frequency-domain MLPs for Time Series forecasting), which applies multi-layer perceptrons (MLPs) in the frequency domain to capture complex spatial-temporal dependencies in time series data more effectively than traditional time-domain or separate encoding methods. FreTS unifies spatial and temporal information within a single transformer-style model, enabling every node at every timestamp to interact with every other node in every other timestamp in just one step through the spatial-temporal correlation matrix. This design allows FreTS to capture global periodic patterns and key features while filtering out noise.
+
+  ```
+  Yi K, Zhang Q, Fan W, et al. Frequency-domain MLPs are more effective learners in time series forecasting[J]. Advances in Neural Information Processing Systems, 2024, 36.
+  ```
+
 #### Traffic Speed Prediction
 
 * **DCRNN**: 

docs/user_guide/model.html

@@ -365,6 +365,12 @@ <h2>Traffic Flow Prediction<a class="headerlink" href="#traffic-flow-prediction"
 </pre></div>
 </div>
 </li>
+<li><p><strong>FreTS</strong></p>
+<p>This paper introduces a new model named FreTS (Frequency-domain MLPs for Time Series forecasting), which applies multi-layer perceptrons (MLPs) in the frequency domain to capture complex spatial-temporal dependencies in time series data more effectively than traditional time-domain or separate encoding methods. FreTS unifies spatial and temporal information within a single transformer-style model, enabling every node at every timestamp to interact with every other node in every other timestamp in just one step through the spatial-temporal correlation matrix. This design allows FreTS to capture global periodic patterns and key features while filtering out noise.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">Yi</span> <span class="n">K</span><span class="p">,</span> <span class="n">Zhang</span> <span class="n">Q</span><span class="p">,</span> <span class="n">Fan</span> <span class="n">W</span><span class="p">,</span> <span class="n">et</span> <span class="n">al</span><span class="o">.</span> <span class="n">Frequency</span><span class="o">-</span><span class="n">domain</span> <span class="n">MLPs</span> <span class="n">are</span> <span class="n">more</span> <span class="n">effective</span> <span class="n">learners</span> <span class="ow">in</span> <span class="n">time</span> <span class="n">series</span> <span class="n">forecasting</span><span class="p">[</span><span class="n">J</span><span class="p">]</span><span class="o">.</span> <span class="n">Advances</span> <span class="ow">in</span> <span class="n">Neural</span> <span class="n">Information</span> <span class="n">Processing</span> <span class="n">Systems</span><span class="p">,</span> <span class="mi">2024</span><span class="p">,</span> <span class="mf">36.</span>
+</pre></div>
+</div>
+</li>
 </ul>
 </section>
 <section id="traffic-speed-prediction">

source/user_guide/model.md

@@ -311,6 +311,14 @@ For trajectory next-location prediction:
   Wang S, Wu H, Shi X, et al. Timemixer: Decomposable multiscale mixing for time series forecasting[J]. arXiv preprint arXiv:2405.14616, 2024.
   ```
 
+- **FreTS**
+
+  This paper introduces a new model named FreTS (Frequency-domain MLPs for Time Series forecasting), which applies multi-layer perceptrons (MLPs) in the frequency domain to capture complex spatial-temporal dependencies in time series data more effectively than traditional time-domain or separate encoding methods. FreTS unifies spatial and temporal information within a single transformer-style model, enabling every node at every timestamp to interact with every other node in every other timestamp in just one step through the spatial-temporal correlation matrix. This design allows FreTS to capture global periodic patterns and key features while filtering out noise.
+
+  ```
+  Yi K, Zhang Q, Fan W, et al. Frequency-domain MLPs are more effective learners in time series forecasting[J]. Advances in Neural Information Processing Systems, 2024, 36.
+  ```
+
 #### Traffic Speed Prediction
 
 * **DCRNN**: