app changed

app.py

@@ -10,7 +10,6 @@ from safetensors import safe_open
 from sentence_transformers import SentenceTransformer
 
 from semantic_search import predict
-from utils.process_data import add_coor, load_data, merge_geoemtry
 
 HF_TOKEN = os.environ.get("HF_TOKEN")
 CITIES_ENRICHED = os.path.join("data", "cities_enriched_manually.csv")
@@ -151,8 +150,9 @@ fig = go.Figure(
 fig.update_layout(margin=dict(t=50, l=25, r=25, b=25))
 fig.update_layout(height=1000, width=1000, template="plotly")
 
+
 # load data ready to plot for local testing
-germany = pd.read_csv(MAP_PATH)
+
 # germany.drop(columns=["lat", "lon"], inplace=True)
 
 # # or generate it directly in this script
@@ -162,15 +162,60 @@ germany = pd.read_csv(MAP_PATH)
 # germany.to_csv(MAP_PATH_WITH_COORD, index=False)
 
 # # germany need columns with lat and lon as well as hover data
-fig_map = px.scatter_mapbox(
+
+
+tensors = {}
+with safe_open("corpus_embeddings.pt", framework="pt", device="cpu") as f:
+    for k in f.keys():
+        tensors[k] = f.get_tensor(k)
+
+model = SentenceTransformer(
+    model_name_or_path="and-effect/musterdatenkatalog_clf",
+    device="cpu",
+    use_auth_token=HF_TOKEN,
+)
+
+
+st.set_page_config(layout="wide")
+
+
+@st.cache_data
+def load_data() -> pd.DataFrame:
+    germany = pd.read_csv(MAP_PATH)
+    return germany
+
+
+# germany = load_data()
+# germany["lat"] = pd.to_numeric(germany["lat"])
+# germany["lon"] = pd.to_numeric(germany["lon"])
+germany = pd.DataFrame(
+    {
+        "ORG": [
+            "Berlin",
+            "Hamburg",
+            "München",
+            "Köln",
+            "Frankfurt am Main",
+            "Stuttgart",
+            "Düsseldorf",
+            "Leipzig",
+            "Dortmund",
+            "Essen",
+        ],
+        "lat": [52.52, 53.55, 48.14, 50.94, 50.11, 48.78, 51.22, 51.34, 51.51, 51.45],
+        "lon": [13.41, 9.99, 11.58, 6.96, 8.68, 9.18, 6.77, 12.37, 7.46, 7.01],
+        "Count": [1000, 800, 600, 500, 400, 300, 200, 150, 100, 50],
+    }
+)
+
+
+fig_map = px.scatter_geo(
     germany,
     lat="lat",
     lon="lon",
     hover_name="ORG",
     custom_data=["Count"],
-    # color_discrete_map=["magenta"],
-    zoom=5,
-    height=700,
+    scope="europe",
 )
 # Custom hover template
 fig_map.update_traces(
@@ -181,22 +226,24 @@ fig_map.update_traces(
         ]
     )
 )
-fig_map.update_layout(mapbox_style="carto-positron")
-
-tensors = {}
-with safe_open("corpus_embeddings.pt", framework="pt", device="cpu") as f:
-    for k in f.keys():
-        tensors[k] = f.get_tensor(k)
-
-model = SentenceTransformer(
-    model_name_or_path="and-effect/musterdatenkatalog_clf",
-    device="cpu",
-    use_auth_token=HF_TOKEN,
+fig_map.update_layout(
+    geo=dict(
+        showland=True,
+        landcolor="LightGray",
+        showocean=True,
+        oceancolor="LightBlue",
+        # showcountries=True,
+        # countrycolor="Gray",
+        showsubunits=True,
+        # subunitcolor="Gray",
+        fitbounds="locations",  # Fit the map bounds to the locations
+        lataxis=dict(range=[47, 55]),  # Approximate latitude range for Germany
+        lonaxis=dict(range=[5, 16]),  # Approximate longitude range for Germany
+    ),
+    mapbox_style="carto-positron",
+    height=700,
 )
 
-
-st.set_page_config(layout="wide")
-
 st.title("Musterdatenkatalog (MDK)")
 
 st.markdown(
@@ -356,5 +403,5 @@ st.markdown(
    """,
     unsafe_allow_html=True,
 )
-st.table(germany.head())
+
 st.plotly_chart(fig_map)