DVC for cleaned data

namansnghl · Jun 7, 2024 · c82f48a · c82f48a
1 parent b7380a7
commit c82f48a
Show file tree

Hide file tree

Showing 7 changed files with 40 additions and 1 deletion.
diff --git a/data/processed_data/.gitignore b/data/processed_data/.gitignore
@@ -0,0 +1 @@
+
diff --git a/data/processed_data/filtered_data.xlsx b/data/processed_data/filtered_data.xlsx
diff --git a/data/processed_data/filtered_data.xlsx.dvc b/data/processed_data/filtered_data.xlsx.dvc
@@ -0,0 +1,5 @@
+outs:
+- md5: 9192a7625bfba4d0057ad48c48e4dd45
+  size: 1306033
+  hash: md5
+  path: filtered_data.xlsx
diff --git a/data/raw_data/.gitignore b/data/raw_data/.gitignore
diff --git a/data/raw_data/countries_to_drop.csv b/data/raw_data/countries_to_drop.csv
@@ -0,0 +1,29 @@
+Country
+Andorra
+Antigua and Barbuda
+Aruba
+Ecuador
+Iraq
+Kiribati
+Kosovo
+Macao SAR
+Marshall Islands
+Micronesia
+Nauru
+Palau
+Papua New Guinea
+Puerto Rico
+Qatar
+Republic of Congo
+San Marino
+Somalia
+South Sudan
+St. Kitts and Nevis
+St. Lucia
+St. Vincent and the Grenadines
+Timor-Leste
+Tuvalu
+Uruguay
+Venezuela
+West Bank and Gaza
+Zimbabwe
diff --git a/data/raw_data/countries_to_drop.csv.dvc b/data/raw_data/countries_to_drop.csv.dvc
@@ -0,0 +1,5 @@
+outs:
+- md5: 36b16aeb184ec2880c9e0910ed7e8b5d
+  size: 361
+  hash: md5
+  path: countries_to_drop.csv
diff --git a/tests/test_transform.py b/tests/test_transform.py
@@ -50,7 +50,6 @@ def test_transform_data():
 
     # Remove the output files after the test
     os.remove(output_excel_path)
-    os.remove(transformed_pickle_path)
 
 def test_melt_dataframe():
     """