Add dataset creation utilities
Browse files- dataset-creation-utilities/create-year-lists.py +13 -0
- dataset-creation-utilities/lists/2004.txt +0 -0
- dataset-creation-utilities/lists/2005.txt +0 -0
- dataset-creation-utilities/lists/2006.txt +0 -0
- dataset-creation-utilities/lists/2007.txt +0 -0
- dataset-creation-utilities/lists/2008.txt +0 -0
- dataset-creation-utilities/lists/2009.txt +0 -0
- dataset-creation-utilities/lists/2010.txt +0 -0
- dataset-creation-utilities/lists/2011.txt +0 -0
- dataset-creation-utilities/lists/2012.txt +0 -0
- dataset-creation-utilities/lists/2013.txt +0 -0
- dataset-creation-utilities/lists/2014.txt +0 -0
- dataset-creation-utilities/lists/2015.txt +0 -0
- dataset-creation-utilities/lists/2016.txt +0 -0
- dataset-creation-utilities/lists/2017.txt +0 -0
- dataset-creation-utilities/lists/2018.txt +0 -0
- dataset-creation-utilities/move-files.sh +5 -0
dataset-creation-utilities/create-year-lists.py
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import pandas as pd
|
2 |
+
import numpy as np
|
3 |
+
from tqdm import tqdm
|
4 |
+
|
5 |
+
df = pd.read_feather("hupd_metadata_2022-02-22.feather", columns=['application_number', 'filing_date'])
|
6 |
+
assert df['filing_date'].dt.year.isnull().sum() == 2
|
7 |
+
df['year'] = df['filing_date'].dt.year.fillna(2018.0).astype(int)
|
8 |
+
years = df['year'].value_counts()
|
9 |
+
print(years)
|
10 |
+
for year in tqdm(years.index):
|
11 |
+
df_year_numbers = df['application_number'][df['year'] == year]
|
12 |
+
df_year_numbers.to_frame().to_csv(f'lists/{year}.txt', sep=' ', index=False, header=False)
|
13 |
+
print(f'saved {year}')
|
dataset-creation-utilities/lists/2004.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2005.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2006.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2007.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2008.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2009.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2010.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2011.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2012.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2013.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2014.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2015.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2016.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2017.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/lists/2018.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
dataset-creation-utilities/move-files.sh
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
for year in 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 # 2004
|
2 |
+
do
|
3 |
+
mkdir years/${year}
|
4 |
+
cat hupd-subset/lists/$year.txt | xargs -i mv distilled/{}.json years/${year}
|
5 |
+
done
|