Cosmetic changes to cleanup code

.gitignore

@@ -6,7 +6,6 @@
 **/*.pyc
 **/*.egg-info
 **/_build/
-**/*tfevents*
 tmd/storage/output/cached_files
 tmd/storage/output/tax_expenditures
 !tmd/storage/input/*.csv
@@ -16,15 +15,8 @@ tmd/storage/output/tax_expenditures
 **demographics_2015.csv
 **puf_2015.csv
 *.DS_STORE
-.Rproj.user
 
-# Local Netlify folder
-.netlify
-
-# Quarto output directories
-**/.quarto/
-
-# Claude AI assistant files
+# Claude Code AI assistant files
 claude/
 
 # local environment

setup.py

@@ -11,7 +11,6 @@
         "scikit-learn",
         "scipy",
         "xlrd",
-        "jax",
         "black>=26.1.0",
         "pycodestyle>=2.14.0",
         "pylint>=3.3.8",

tests/conftest.py

@@ -6,7 +6,7 @@
 import pytest
 import numpy as np
 import pandas as pd
-import taxcalc as tc
+import taxcalc
 from tmd.storage import STORAGE_FOLDER
 from tmd.imputation_assumptions import TAXYEAR
 
@@ -18,13 +18,15 @@ def create_tmd_records(
     data_path, weights_path, growfactors_path, exact_calculations=True
 ):
     """
-    Create tc.Records with start_year=TAXYEAR.
+    Create taxcalc.Records with start_year=TAXYEAR.
     Bypasses tmd_constructor() which hardcodes start_year=2021.
     """
-    return tc.Records(
+    return taxcalc.Records(
         data=pd.read_csv(data_path),
         start_year=TAXYEAR,
-        gfactors=tc.GrowFactors(growfactors_filename=str(growfactors_path)),
+        gfactors=taxcalc.GrowFactors(
+            growfactors_filename=str(growfactors_path)
+        ),
         weights=pd.read_csv(weights_path),
         adjust_ratios=None,
         exact_calculations=exact_calculations,

tests/test_imputed_variables.py

@@ -6,7 +6,7 @@
 
 import numpy as np
 import pytest
-import taxcalc as tc
+import taxcalc
 from tmd.imputation_assumptions import TAXYEAR, CREDIT_CLAIMING
 from tests.conftest import create_tmd_records
 
@@ -123,9 +123,9 @@ def actual_results(rdf, bdf):
         growfactors_path=tmd_growfactors_path,
     )
     # create baseline_sim Calculator object for simyear and get its output
-    pol = tc.Policy()
+    pol = taxcalc.Policy()
     pol.implement_reform(CREDIT_CLAIMING)
-    baseline_sim = tc.Calculator(policy=pol, records=recs)
+    baseline_sim = taxcalc.Calculator(policy=pol, records=recs)
     baseline_sim.advance_to_year(simyear)
     baseline_sim.calc_all()
     bdf = baseline_sim.dataframe(output_variables)
@@ -140,10 +140,10 @@ def actual_results(rdf, bdf):
     tolerance_scale = 1.0
     for ded, info in deductions.items():
         # create reform Calculator object for simyear
-        reform_policy = tc.Policy()
+        reform_policy = taxcalc.Policy()
         reform_policy.implement_reform(CREDIT_CLAIMING)
         reform_policy.implement_reform(info["reform_dict"])
-        reform_sim = tc.Calculator(policy=reform_policy, records=recs)
+        reform_sim = taxcalc.Calculator(policy=reform_policy, records=recs)
         reform_sim.advance_to_year(simyear)
         # get reform Calculator object's output
         reform_sim.calc_all()

tests/test_misc.py

@@ -3,7 +3,7 @@
 """
 
 import pytest
-import taxcalc as tc
+import taxcalc
 from tmd.storage import STORAGE_FOLDER
 from tmd.imputation_assumptions import TAXYEAR, CREDIT_CLAIMING
 from tests.conftest import create_tmd_records
@@ -52,14 +52,14 @@ def compare(name, act, exp, tol):
         ), f"{name}:act,exp,tol= {act} {exp} {tol}"
 
     # use national tmd files to compute various TAXYEAR income tax statistics
-    pol = tc.Policy()
+    pol = taxcalc.Policy()
     pol.implement_reform(CREDIT_CLAIMING)
     rec = create_tmd_records(
         data_path=STORAGE_FOLDER / "output" / "tmd.csv.gz",
         weights_path=STORAGE_FOLDER / "output" / "tmd_weights.csv.gz",
         growfactors_path=STORAGE_FOLDER / "output" / "tmd_growfactors.csv",
     )
-    sim = tc.Calculator(policy=pol, records=rec)
+    sim = taxcalc.Calculator(policy=pol, records=rec)
     sim.advance_to_year(TAXYEAR)
     sim.calc_all()
     wght = sim.array("s006")

tests/test_tax_expenditures.py

@@ -1,13 +1,13 @@
 """
-Test 2023 tax expenditures calculated using tmd files
+Test 2023 tax expenditures calculated using TMD files
 against expected tax expenditure values in the tests folder.
 """
 
 import pytest
 import numpy as np
 from tmd.storage import STORAGE_FOLDER
 from tmd.imputation_assumptions import TAXYEAR
-from tmd.utils.taxcalc_utils import get_tax_expenditure_results
+from tmd.utils.tax_expenditures import get_tax_expenditure_results
 
 
 @pytest.mark.taxexp

tests/test_tax_revenue.py

@@ -5,7 +5,7 @@
 import yaml
 import numpy as np
 import pytest
-import taxcalc as tc
+import taxcalc
 from tmd.imputation_assumptions import TAXYEAR, CREDIT_CLAIMING
 
 FIRST_CYR = 2023
@@ -45,11 +45,11 @@ def test_tax_revenue(
         exp_itax[year] = round(fy2cy(fy_itax[year], fy_itax[year + 1]), 3)
         exp_ptax[year] = round(fy2cy(fy_ptax[year], fy_ptax[year + 1]), 3)
     # calculate actual tax revenues for each calendar year
-    pol = tc.Policy()
+    pol = taxcalc.Policy()
     pol.implement_reform(CREDIT_CLAIMING)
     wghts = str(tmd_weights_path)
-    growf = tc.GrowFactors(growfactors_filename=str(tmd_growfactors_path))
-    input_data = tc.Records(
+    growf = taxcalc.GrowFactors(growfactors_filename=str(tmd_growfactors_path))
+    input_data = taxcalc.Records(
         data=tmd_variables,
         start_year=TAXYEAR,
         weights=wghts,
@@ -58,7 +58,7 @@ def test_tax_revenue(
         exact_calculations=True,
         weights_scale=1.0,
     )
-    sim = tc.Calculator(policy=pol, records=input_data)
+    sim = taxcalc.Calculator(policy=pol, records=input_data)
     act_itax = {}
     act_ptax = {}
     for year in range(FIRST_CYR, LAST_CYR + 1):

tmd/create_taxcalc_cached_files.py

@@ -4,7 +4,7 @@
 
 import numpy as np
 import pandas as pd
-import taxcalc as tc
+import taxcalc
 from tmd.storage import STORAGE_FOLDER, CACHED_TAXCALC_VARIABLES
 from tmd.imputation_assumptions import TAXYEAR, CREDIT_CLAIMING
 
@@ -21,18 +21,18 @@ def create_cached_files():
     # calculate all Tax-Calculator variables for TAXYEAR
     # Construct Records directly (bypassing tmd_constructor which
     # hardcodes start_year=2021 in the taxcalc library).
-    pol = tc.Policy()
+    pol = taxcalc.Policy()
     pol.implement_reform(CREDIT_CLAIMING)
-    rec = tc.Records(
+    rec = taxcalc.Records(
         data=pd.read_csv(INFILE_PATH),
         start_year=TAXYEAR,
-        gfactors=tc.GrowFactors(growfactors_filename=str(GFFILE_PATH)),
+        gfactors=taxcalc.GrowFactors(growfactors_filename=str(GFFILE_PATH)),
         weights=pd.read_csv(WTFILE_PATH),
         adjust_ratios=None,
         exact_calculations=True,
         weights_scale=1.0,
     )
-    calc = tc.Calculator(policy=pol, records=rec)
+    calc = taxcalc.Calculator(policy=pol, records=rec)
     calc.advance_to_year(TAXYEAR)
     calc.calc_all()
 

tmd/create_taxcalc_input_variables.py

@@ -2,7 +2,7 @@
 Construct tmd.csv, a Tax-Calculator-style input variable file for TAXYEAR.
 """
 
-import taxcalc as tc
+import taxcalc
 from tmd.datasets.tmd import create_tmd_dataframe
 from tmd.imputation_assumptions import (
     TAXYEAR,
@@ -41,7 +41,7 @@ def create_variable_file(write_file=True):
         vdf.to_csv(fname, index=False)
     # streamline dataframe so that it includes only input variables
     print("Removing output variables from PUF+CPS dataframe...")
-    rec = tc.Records(
+    rec = taxcalc.Records(
         data=vdf,
         start_year=TAXYEAR,
         gfactors=None,

tmd/datasets/cps.py

@@ -138,7 +138,7 @@
     "WKSWORK",
 ]
 
-TC_CPS_AGED_RNG = np.random.default_rng(seed=374651932)
+TAXCALC_CPS_AGED_RNG = np.random.default_rng(seed=374651932)
 
 CPS_URL_BY_YEAR = {
     2018: (
@@ -305,7 +305,7 @@ def _derive_age(person: pd.DataFrame) -> np.ndarray:
     """
     return np.where(
         person.A_AGE == 80,
-        TC_CPS_AGED_RNG.integers(
+        TAXCALC_CPS_AGED_RNG.integers(
             low=80, high=85, endpoint=False, size=len(person)
         ),
         person.A_AGE.values,
@@ -373,7 +373,7 @@ def _is_tax_filer(tcdf: pd.DataFrame, taxyear: int) -> pd.Series:
     return filer
 
 
-def create_tc_cps(taxyear: int) -> (pd.DataFrame, pd.Series):
+def create_taxcalc_cps(taxyear: int) -> (pd.DataFrame, pd.Series):
     """
     Create a Tax-Calculator-compatible CPS DataFrame for the given taxyear
     directly from the Census raw CPS data.
@@ -583,7 +583,7 @@ def map_spouse(values):
     var["f2441"] = sum_all((age < 13) * dep_flag).astype(int)
 
     # variables with no CPS source are set to zero
-    zero_tc_names = [
+    zero_taxcalc_names = [
         "a_lineno",
         "agi_bin",
         "h_seq",
@@ -646,7 +646,7 @@ def map_spouse(values):
         "e09700",  # recapture of investment credit
         "e09800",  # unreported payroll tax
     ]
-    for tcname in zero_tc_names:
+    for tcname in zero_taxcalc_names:
         if tcname not in var:
             var[tcname] = zeros
 

tmd/datasets/puf.py

@@ -170,7 +170,7 @@ def preprocess_puf(puf: pd.DataFrame) -> pd.DataFrame:
     return puf
 
 
-def create_tc_puf(taxyear: int) -> pd.DataFrame:
+def create_taxcalc_puf(taxyear: int) -> pd.DataFrame:
     """
     Create a Tax-Calculator-compatible PUF DataFrame for
     the given taxyear directly from raw PUF data.
@@ -309,7 +309,7 @@ def _decode_dep_ages(agerange_vals, rng):
     # mapping from TC variable name to PE-named column in pre-processed PUF:
     #  for person-level variables, the tax-unit total is scaled by
     #  person_scale (= head_frac + (1-head_frac)*is_joint)
-    tc_to_pe = {
+    taxcalc_to_policyengine = {
         "RECID": "household_id",
         "S006": "household_weight",
         "E03500": "alimony_expense",
@@ -413,7 +413,7 @@ def _decode_dep_ages(agerange_vals, rng):
     # create dictionary that will be used to create Tax-Calculator DataFrame
     NO_SCALE = {"RECID", "S006", "XTOT"}
     var = {}
-    for tcname, pename in tc_to_pe.items():
+    for tcname, pename in taxcalc_to_policyengine.items():
         if tcname in NO_SCALE:
             var[tcname] = puf[pename].values
         elif pename in PERSON_LEVEL_VARS:

tmd/datasets/tmd.py

@@ -1,27 +1,27 @@
 import numpy as np
 import pandas as pd
 from tmd.imputation_assumptions import TAXYEAR, CPS_WEIGHTS_SCALE
-from tmd.datasets.puf import create_tc_puf
-from tmd.datasets.cps import create_tc_cps
-from tmd.utils.taxcalc_utils import add_taxcalc_outputs
+from tmd.datasets.puf import create_taxcalc_puf
+from tmd.datasets.cps import create_taxcalc_cps
+from tmd.utils.taxcalc_output import add_taxcalc_outputs
 from tmd.utils.reweight import reweight
 
 
 def create_tmd_dataframe(taxyear: int) -> pd.DataFrame:
     """
     Create DataFrame for given taxyear containing PUF filers and CPS nonfilers.
     """
-    # always call create_tc_puf and create_tc_cps
+    # always call create_taxcalc_puf and create_taxcalc_cps
     # (because imputation assumptions may have changed)
-    tc_puf = create_tc_puf(taxyear)
-    tc_cps, nonfiler = create_tc_cps(taxyear)
-    tc_cps = tc_cps[nonfiler].reset_index(drop=True)
+    taxcalc_puf = create_taxcalc_puf(taxyear)
+    taxcalc_cps, nonfiler = create_taxcalc_cps(taxyear)
+    taxcalc_cps = taxcalc_cps[nonfiler].reset_index(drop=True)
 
     # scale CPS weights to get sensible combined population count
-    tc_cps["s006"] *= CPS_WEIGHTS_SCALE[TAXYEAR]
+    taxcalc_cps["s006"] *= CPS_WEIGHTS_SCALE[TAXYEAR]
 
     print("Combining PUF filers and CPS nonfilers...")
-    combined = pd.concat([tc_puf, tc_cps], ignore_index=True)
+    combined = pd.concat([taxcalc_puf, taxcalc_cps], ignore_index=True)
 
     # ensure RECID values are unique
     combined["RECID"] = np.arange(1, len(combined) + 1, dtype=int)

tmd/examination/2022/bootstrap_sampling.py

@@ -6,7 +6,7 @@
 import numpy as np
 import pandas as pd
 
-USAGE = "USAGE: python bootstrap_sampling.py tc_dump_output_csv_file_name\n"
+USAGE = "USAGE: python bootstrap_sampling.py taxcalc_dump_output_csv_fname\n"
 
 SS_FRAC = 1.00  # 0.1
 SS_RNSEED = 902345678