Add ruff rules for pandas-vet (#561)

Author: cbornet

examples/evaluation/langchain_trulens_full.ipynb

@@ -468,9 +468,7 @@
     "\n",
     "columns_to_drop = [col for col in df_records.columns if col not in columns_to_keep]\n",
     "\n",
-    "df_records.drop(columns=columns_to_drop, inplace=True)\n",
-    "\n",
-    "df_records"
+    "df_records.drop(columns=columns_to_drop)"
    ]
   },
   {

examples/evaluation/tru_download.py

@@ -22,7 +22,7 @@
     ]
     columns_to_drop = [col for col in df_records.columns if col not in columns_to_keep]
 
-    df_records.drop(columns=columns_to_drop, inplace=True)
+    df_records = df_records.drop(columns=columns_to_drop)
 
     df_records["test"] = app_id.split("#")[0]
     df_records["test_uuid"] = app_id.split("#")[1]

examples/notebooks/langchain_multimodal_gemini.ipynb

@@ -551,8 +551,8 @@
     "        \"https://www.breville.com/content/dam/breville/us/catalog/products/images/sp0/sp0000166/tile.jpg\",\n",
     "    ],\n",
     "}\n",
-    "df = pd.DataFrame(data=d)\n",
-    "df"
+    "products = pd.DataFrame(data=d)\n",
+    "products"
    ]
   },
   {
@@ -599,11 +599,11 @@
     "    collection_name=\"coffee_shop_ecommerce\", dimension=1408\n",
     ")\n",
     "\n",
-    "for i in range(len(df)):\n",
-    "    name = df.loc[i, \"name\"]\n",
-    "    image = df.loc[i, \"image\"]\n",
-    "    price = df.loc[i, \"price\"]\n",
-    "    url = df.loc[i, \"url\"]\n",
+    "for i in range(len(products)):\n",
+    "    name = products.loc[i, \"name\"]\n",
+    "    image = products.loc[i, \"image\"]\n",
+    "    price = products.loc[i, \"price\"]\n",
+    "    url = products.loc[i, \"url\"]\n",
     "\n",
     "    # Download this product's image and save it to the Colab filesystem.\n",
     "    # In a production system this binary data would be stored in Google Cloud Storage\n",

libs/ragulate/ragstack_ragulate/analysis.py

@@ -22,7 +22,7 @@ def get_all_data(self, recipes: List[str]) -> DataFrame:
 
             for app in tru.get_apps():
                 dataset = app["app_id"]
-                df, metrics = tru.get_records_and_feedback([dataset])
+                df_records, metrics = tru.get_records_and_feedback([dataset])
                 all_metrics.extend(metrics)
 
                 columns_to_keep = [
@@ -33,24 +33,24 @@ def get_all_data(self, recipes: List[str]) -> DataFrame:
                     "total_cost",
                 ]
                 columns_to_drop = [
-                    col for col in df.columns if col not in columns_to_keep
+                    col for col in df_records.columns if col not in columns_to_keep
                 ]
 
-                df.drop(columns=columns_to_drop, inplace=True)
-                df["recipe"] = recipe
-                df["dataset"] = dataset
+                df_records = df_records.drop(columns=columns_to_drop)
+                df_records["recipe"] = recipe
+                df_records["dataset"] = dataset
 
                 # set negative values to None
                 for metric in metrics:
-                    df.loc[df[metric] < 0, metric] = None
+                    df_records.loc[df_records[metric] < 0, metric] = None
 
-                df_all = pd.concat([df_all, df], axis=0, ignore_index=True)
+                df_all = pd.concat([df_all, df_records], axis=0, ignore_index=True)
 
             tru.delete_singleton()
 
-        df_all.reset_index(drop=True, inplace=True)
+        reset_df = df_all.reset_index(drop=True)
 
-        return df_all, list(set(all_metrics))
+        return reset_df, list(set(all_metrics))
 
     def calculate_statistics(self, df: pd.DataFrame, metrics: list):
         stats = {}

pyproject.toml

@@ -85,6 +85,7 @@ select = [
     "LOG",
     "N",
     "NPY",
+    "PD",
     "PERF",
     "PIE",
     "PGH",