tryout no plot

streamlit_simulation/app.py

@@ -142,29 +142,29 @@ def load_lightgbm_model():
 def load_transformer_model_and_dataset():
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
-    try:
-        with st.spinner("🔄 Loading transformer model..."):
-            # Load model
-            model = load_moment_model()
-            checkpoint_path = hf_hub_download(
-            repo_id="dlaj/energy-forecasting-files",
-            filename="transformer_model/model_final.pth",
-            repo_type="dataset"
-            )
-            model.load_state_dict(torch.load(checkpoint_path, map_location=device))
-            model.to(device)
-            model.eval()
+    model = load_moment_model()
+    checkpoint_path = hf_hub_download(
+    repo_id="dlaj/energy-forecasting-files",
+    filename="transformer_model/model_final.pth",
+    repo_type="dataset"
+    )
+    model.load_state_dict(torch.load(checkpoint_path, map_location=device))
+    model.to(device)
+    model.eval()
 
-            # Datasets
-            train_dataset = InformerDataset(data_split="train", forecast_horizon=FORECAST_HORIZON, random_seed=13)
-            test_dataset = InformerDataset(data_split="test", forecast_horizon=FORECAST_HORIZON, random_seed=13)
-            test_dataset.scaler = train_dataset.scaler
+    csv_path = hf_hub_download(
+    repo_id="dlaj/energy-forecasting-files",
+    filename="data/processed/energy_consumption_aggregated_cleaned.csv",
+    repo_type="dataset"
+    )
+
+    # Datasets
+    train_dataset = InformerDataset(data_split="train", forecast_horizon=FORECAST_HORIZON, random_seed=13, csv_path=csv_path)
+    test_dataset = InformerDataset(data_split="test", forecast_horizon=FORECAST_HORIZON, random_seed=13, csv_path=csv_path)
+    test_dataset.scaler = train_dataset.scaler
+
+    return model, test_dataset, device
 
-            return model, test_dataset, device
-        
-    except Exception as e:
-        st.error(f"❌ Fehler beim Laden des Transformer-Modells: {e}")
-        raise e
 
 @st.cache_data
 def load_data():

transformer_model/scripts/utils/informer_dataset_class.py

@@ -18,6 +18,7 @@ class InformerDataset:
         data_stride_len: int = 1,
         task_name: str = "forecasting",
         random_seed: int = 42,
+        csv_path=None
     ):
         """
         Parameters
@@ -36,7 +37,7 @@ class InformerDataset:
 
         self.seq_len = SEQ_LEN
         self.forecast_horizon = forecast_horizon
-        self.full_file_path_and_name = DATA_PATH
+        self.full_file_path_and_name = csv_path if csv_path is not None else DATA_PATH
         self.data_split = data_split
         self.data_stride_len = data_stride_len
         self.task_name = task_name