bug fix

app.py

@@ -114,7 +114,26 @@ def main():
     image = Image.open('data/image.png')
     st.image(image, caption='Coding.Waterkant Festival for AI')
 
+    
     st.markdown(body = """
+        ### Abstract
+        Multi-horizon forecasting often contains a complex mix of inputs – including
+        static (i.e. time-invariant) covariates, known future inputs, and other exogenous
+        time series that are only observed in the past – without any prior information
+        on how they interact with the target. Several deep learning methods have been
+        proposed, but they are typically ‘black-box’ models which do not shed light on
+        how they use the full range of inputs present in practical scenarios. In this pa-
+        per, we introduce the Temporal Fusion Transformer (TFT) – a novel attention-
+        based architecture which combines high-performance multi-horizon forecasting
+        with interpretable insights into temporal dynamics. To learn temporal rela-
+        tionships at different scales, TFT uses recurrent layers for local processing and
+        interpretable self-attention layers for long-term dependencies. TFT utilizes spe-
+        cialized components to select relevant features and a series of gating layers to
+        suppress unnecessary components, enabling high performance in a wide range of
+        scenarios. On a variety of real-world datasets, we demonstrate significant per-
+        formance improvements over existing benchmarks, and showcase three practical
+        interpretability use cases of TFT.
+    
         ### Experiments
         We implemented TFT for sales multi-horizon sales forecast during Coding.Waterkant. 
         Please try our implementation and adjust some of the training data.
@@ -143,29 +162,8 @@ def main():
     st.pyplot(fig)
     
     st.markdown(body = """
-        ### Abstract
-        Multi-horizon forecasting often contains a complex mix of inputs – including
-        static (i.e. time-invariant) covariates, known future inputs, and other exogenous
-        time series that are only observed in the past – without any prior information
-        on how they interact with the target. Several deep learning methods have been
-        proposed, but they are typically ‘black-box’ models which do not shed light on
-        how they use the full range of inputs present in practical scenarios. In this pa-
-        per, we introduce the Temporal Fusion Transformer (TFT) – a novel attention-
-        based architecture which combines high-performance multi-horizon forecasting
-        with interpretable insights into temporal dynamics. To learn temporal rela-
-        tionships at different scales, TFT uses recurrent layers for local processing and
-        interpretable self-attention layers for long-term dependencies. TFT utilizes spe-
-        cialized components to select relevant features and a series of gating layers to
-        suppress unnecessary components, enabling high performance in a wide range of
-        scenarios. On a variety of real-world datasets, we demonstrate significant per-
-        formance improvements over existing benchmarks, and showcase three practical
-        interpretability use cases of TFT.
-    
-        ### Experiments
-        We implemented TFT for sales multi-horizon sales forecast during Coding.Waterkant. 
-        Please try our implementation and adjust some of the training data.
-
-        Adjustments to the model and extention with Quantile forecast are coming soon ;) 
+        Sources: Bryan Lim et al. in Temporal Fusion Transformers (TFT) for Interpretable Multi-horizon Time Series Forecasting
+        Demo created by: <a href=https://github.com/MalteLeuschner</a>
     """)
     
 if __name__ == '__main__':