Fixing errors

tabs/main_page_tabs/dataset_overview.py

@@ -11,16 +11,9 @@ import pandas as pd
 import os
 from pathlib import Path
 from utils.icons import lucide_icon
-from utils.load_data import load_dataset
+from utils.load_data import get_data_directory
 
-def find_data_dir():
-    current = Path(__file__).resolve()
-    for parent in current.parents:
-        if (parent / "review_data").exists():
-            return parent / "review_data"
-    return None
-
-DATA_DIR = find_data_dir()
+DATA_DIR = get_data_directory()
 
 def render():
     st.sidebar.header('Data')

tabs/predictive_model_tabs/pred_model_two.py

@@ -8,7 +8,11 @@ def render():
     model = load_model() # Using the cached function
     df = load_demo_data()
     model_columns = ['lemma_title', 'lemma_text', 'images', 'Review Length', 'Title Length']
-    X = df[model_columns]
+    
+    X = df[model_columns].copy()
+    X['lemma_title'] = X['lemma_title'].fillna("")
+    X['lemma_text'] = X['lemma_text'].fillna("")
+
     y = df['vote']
     st.session_state.demo_probs = model.predict_proba(X)
 

utils/load_data.py

@@ -1,6 +1,7 @@
 import pandas as pd
 import streamlit as st
 from pathlib import Path
+from huggingface_hub import snapshot_download
 import os
 
 @st.cache_data(show_spinner="Loading data...⏳")
@@ -21,4 +22,14 @@ def load_dataset(path: str | Path, category: str | None=None):
     else:
         df = pd.read_parquet(path)
 
-    return df
+    return df
+
+
+@st.cache_resource
+def get_data_directory():
+    # This downloads the whole review_data folder from your Dataset repo
+    data_path = snapshot_download(
+        repo_id="tkbarb10/ads505-review-data", 
+        repo_type="dataset"
+    )
+    return Path(data_path)

utils/topically.py

@@ -10,6 +10,7 @@ from sklearn.decomposition import NMF
 from sklearn.feature_extraction.text import TfidfVectorizer
 from sklearn.pipeline import make_pipeline, Pipeline
 from utils.build_plotly import _build_topic_figure
+from load_data import get_data_directory
 
 import plotly.graph_objects as go # type: ignore
 
@@ -22,17 +23,6 @@ from utils.remove_html import remove_html_tags
 # ROOT = Path(__file__).resolve().parents[1]
 # DEFAULT_DATA_DIR = ROOT / "review_data"
 
-from huggingface_hub import snapshot_download
-
-@st.cache_resource
-def get_data_directory():
-    # This downloads the whole review_data folder from your Dataset repo
-    data_path = snapshot_download(
-        repo_id="tkbarb10/ads505-review-data", 
-        repo_type="dataset"
-    )
-    return Path(data_path) / "review_data"
-
 DEFAULT_DATA_DIR = get_data_directory()
 
 COLOR_WHEEL = {