Update README.md

README.md

@@ -79,7 +79,7 @@ import pandas as pd
 from datetime import datetime
 import gc
 
-def load_csvs_from_huggingface(start_date, end_date, columns_to_keep=None):
+def load_temporal_optimized(start_date, end_date, columns_to_keep=None):
     """
     Load only the necessary CSV files from a Hugging Face dataset repository.
 
@@ -113,6 +113,8 @@ def load_csvs_from_huggingface(start_date, end_date, columns_to_keep=None):
         "patent_number": "Int64",
     }
 
+    category_cols = {"country", "kind", "lang", "status"}
+
     dataset_years = [str(year) for year in range(1978, 2006)]
 
     start_date_int = int(datetime.strptime(start_date, "%Y-%m-%d").strftime("%Y%m%d"))
@@ -126,20 +128,20 @@ def load_csvs_from_huggingface(start_date, end_date, columns_to_keep=None):
         raise ValueError(f"No matching CSV files found for {start_date} to {end_date}")
 
     df_list = []
-    
+
     for year in matching_years:
         filepath = f"data/years/{year}/clefip2011_en_classification_{year}_validated.csv"
 
         try:
             # 1. Load the dataset (This stays on disk as an Arrow memory-map, NOT in RAM)
             dataset = load_dataset(
-                huggingface_dataset_name, 
-                data_files=filepath, 
+                huggingface_dataset_name,
+                data_files=filepath,
                 split="train",
                 sep=";",
                 on_bad_lines="skip"
             )
-            
+
             # Select Columns Before Chunking
             if columns_to_keep is not None:
                 # Safety check: Only select columns that actually exist in this specific file
@@ -149,14 +151,14 @@ def load_csvs_from_huggingface(start_date, end_date, columns_to_keep=None):
 
             # 2. Tell HuggingFace to output Pandas dataframes when sliced
             dataset = dataset.with_format("pandas")
-            
-            # 3. CHUNKING: Process exactly 10,000 rows at a time
+
+            # 3. CHUNKING: Process exactly 40,000 rows at a time
             chunk_size = 10000
             for i in range(0, len(dataset), chunk_size):
-                # Only these 10,000 rows are loaded into RAM
+                # Only these 40,000 rows are loaded into RAM
                 df_chunk = dataset[i : i + chunk_size]
                 df_chunk.columns = df_chunk.columns.str.strip()
-                
+
                 # Filter this specific chunk
                 if "date" in df_chunk.columns:
                     temp_dates = pd.to_numeric(df_chunk["date"], errors="coerce")
@@ -164,13 +166,22 @@ def load_csvs_from_huggingface(start_date, end_date, columns_to_keep=None):
                     df_filtered = df_chunk[mask].copy()
                 else:
                     df_filtered = df_chunk.copy()
-                
+
                 # Apply types and append
                 if not df_filtered.empty:
-                    valid_column_types = {col: dtype for col, dtype in column_types.items() if col in df_filtered.columns}
+                    # Step 1: Cast category columns via string[pyarrow] first to handle None/mixed types
+                    for col in category_cols:
+                        if col in df_filtered.columns:
+                            df_filtered[col] = pd.Categorical(df_filtered[col])
+
+                    # Step 2: Apply remaining column types, skipping category cols already handled
+                    valid_column_types = {
+                        col: dtype for col, dtype in column_types.items()
+                        if col in df_filtered.columns and col not in category_cols
+                    }
                     df_filtered = df_filtered.astype(valid_column_types)
                     df_list.append(df_filtered)
-                
+
                 # Clear chunk from memory immediately
                 del df_chunk, df_filtered
                 gc.collect()
@@ -182,17 +193,18 @@ def load_csvs_from_huggingface(start_date, end_date, columns_to_keep=None):
         except Exception as e:
             print(f"Error processing {filepath}: {e}")
 
-
     if not df_list:
         return pd.DataFrame()
-    
+
     # Combine chunks
     final_df = pd.concat(df_list, ignore_index=True)
-    
+    for col in category_cols:
+        if col in final_df.columns:
+            final_df[col] = pd.Categorical(final_df[col])
     # Destroy the list
     del df_list
     gc.collect()
-    
+
     return final_df
 
 
@@ -205,7 +217,7 @@ Load All Columns
 start_date = "1985-03-01"
 end_date = "1985-04-30"
 
-df = load_csvs_from_huggingface(start_date, end_date)
+df = load_temporal_optimized(start_date, end_date)
 
 
 ```
@@ -223,7 +235,7 @@ columns_to_keep = [
 start_date = "1985-03-01"
 end_date = "1985-04-30"
 
-df = load_csvs_from_huggingface(start_date, end_date, columns_to_keep)
+df = load_temporal_optimized(start_date, end_date, columns_to_keep)
 
 ```