BUG: crosstab(dropna=False) did not keep np.nan in result (#53205) · pandas-dev/pandas@3cfd868 (original) (raw)

`@@ -286,24 +286,29 @@ def test_margin_dropna4(self):

286

`# GH 12642

287

`# _add_margins raises KeyError: Level None not found

288

`# when margins=True and dropna=False

289

GH: 10772: Keep np.nan in result with dropna=False

289

290

`df = DataFrame({"a": [1, 2, 2, 2, 2, np.nan], "b": [3, 3, 4, 4, 4, 4]})

290

291

`actual = crosstab(df.a, df.b, margins=True, dropna=False)

291

expected = DataFrame([[1, 0, 1], [1, 3, 4], [2, 4, 6]])

292

expected.index = Index([1.0, 2.0, "All"], name="a")

292

expected = DataFrame([[1, 0, 1.0], [1, 3, 4.0], [0, 1, np.nan], [2, 4, 6.0]])

293

expected.index = Index([1.0, 2.0, np.nan, "All"], name="a")

293

294

`expected.columns = Index([3, 4, "All"], name="b")

294

295

`tm.assert_frame_equal(actual, expected)

295

296

297

`def test_margin_dropna5(self):

298

GH: 10772: Keep np.nan in result with dropna=False

297

299

`df = DataFrame(

298

300

` {"a": [1, np.nan, np.nan, np.nan, 2, np.nan], "b": [3, np.nan, 4, 4, 4, 4]}

299

301

` )

300

302

`actual = crosstab(df.a, df.b, margins=True, dropna=False)

301

expected = DataFrame([[1, 0, 1], [0, 1, 1], [1, 4, 6]])

302

expected.index = Index([1.0, 2.0, "All"], name="a")

303

expected.columns = Index([3.0, 4.0, "All"], name="b")

303

expected = DataFrame(

304

[[1, 0, 0, 1.0], [0, 1, 0, 1.0], [0, 3, 1, np.nan], [1, 4, 0, 6.0]]

305

)

306

expected.index = Index([1.0, 2.0, np.nan, "All"], name="a")

307

expected.columns = Index([3.0, 4.0, np.nan, "All"], name="b")

304

308

`tm.assert_frame_equal(actual, expected)

305

309

306

310

`def test_margin_dropna6(self):

311

GH: 10772: Keep np.nan in result with dropna=False

307

312

`a = np.array(["foo", "foo", "foo", "bar", "bar", "foo", "foo"], dtype=object)

308

313

`b = np.array(["one", "one", "two", "one", "two", np.nan, "two"], dtype=object)

309

314

`c = np.array(

`@@ -315,13 +320,14 @@ def test_margin_dropna6(self):

315

320

` )

316

321

`m = MultiIndex.from_arrays(

317

322

` [

318

["one", "one", "two", "two", "All"],

319

["dull", "shiny", "dull", "shiny", ""],

323

["one", "one", "two", "two", np.nan, np.nan, "All"],

324

["dull", "shiny", "dull", "shiny", "dull", "shiny", ""],

320

325

` ],

321

326

`names=["b", "c"],

322

327

` )

323

328

`expected = DataFrame(

324

[[1, 0, 1, 0, 2], [2, 0, 1, 1, 5], [3, 0, 2, 1, 7]], columns=m

329

[[1, 0, 1, 0, 0, 0, 2], [2, 0, 1, 1, 0, 1, 5], [3, 0, 2, 1, 0, 0, 7]],

330

columns=m,

325

331

` )

326

332

`expected.index = Index(["bar", "foo", "All"], name="a")

327

333

`tm.assert_frame_equal(actual, expected)

`@@ -330,11 +336,23 @@ def test_margin_dropna6(self):

330

336

` [a, b], c, rownames=["a", "b"], colnames=["c"], margins=True, dropna=False

331

337

` )

332

338

`m = MultiIndex.from_arrays(

333

[["bar", "bar", "foo", "foo", "All"], ["one", "two", "one", "two", ""]],

339

[

340

["bar", "bar", "bar", "foo", "foo", "foo", "All"],

341

["one", "two", np.nan, "one", "two", np.nan, ""],

342

334

343

`names=["a", "b"],

335

344

` )

336

345

`expected = DataFrame(

337

[[1, 0, 1], [1, 0, 1], [2, 0, 2], [1, 1, 2], [5, 2, 7]], index=m

346

[

347

[1, 0, 1.0],

348

[1, 0, 1.0],

349

[0, 0, np.nan],

350

[2, 0, 2.0],

351

[1, 1, 2.0],

352

[0, 1, np.nan],

353

[5, 2, 7.0],

354

355

index=m,

338

356

` )

339

357

`expected.columns = Index(["dull", "shiny", "All"], name="c")

340

358

`tm.assert_frame_equal(actual, expected)