ENH: Added a min_count keyword to stat funcs (#18876) · pandas-dev/pandas@dbec3c9 (original) (raw)

`@@ -36,7 +36,8 @@ def get_dispatch(dtypes):

36

`def group_add_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

37

` ndarray[int64_t] counts,

38

` ndarray[{{c_type}}, ndim=2] values,

39

ndarray[int64_t] labels):

39

ndarray[int64_t] labels,

40

Py_ssize_t min_count=1):

40

41

` """

41

42

` Only aggregates on axis=0

42

43

` """

`@@ -88,7 +89,7 @@ def group_add_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

88

89

90

` for i in range(ncounts):

90

91

` for j in range(K):

91

if nobs[i, j] == 0:

92

if nobs[i, j] < min_count:

92

93

` out[i, j] = NAN

93

94

` else:

94

95

` out[i, j] = sumx[i, j]

`@@ -99,7 +100,8 @@ def group_add_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

99

100

`def group_prod_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

100

101

` ndarray[int64_t] counts,

101

102

` ndarray[{{c_type}}, ndim=2] values,

102

ndarray[int64_t] labels):

103

ndarray[int64_t] labels,

104

Py_ssize_t min_count=1):

103

105

` """

104

106

` Only aggregates on axis=0

105

107

` """

`@@ -147,7 +149,7 @@ def group_prod_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

147

149

148

150

` for i in range(ncounts):

149

151

` for j in range(K):

150

if nobs[i, j] == 0:

152

if nobs[i, j] < min_count:

151

153

` out[i, j] = NAN

152

154

` else:

153

155

` out[i, j] = prodx[i, j]

`@@ -159,12 +161,15 @@ def group_prod_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

159

161

`def group_var_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

160

162

` ndarray[int64_t] counts,

161

163

` ndarray[{{dest_type2}}, ndim=2] values,

162

ndarray[int64_t] labels):

164

ndarray[int64_t] labels,

165

Py_ssize_t min_count=-1):

163

166

` cdef:

164

167

` Py_ssize_t i, j, N, K, lab, ncounts = len(counts)

165

168

` {{dest_type2}} val, ct, oldmean

166

169

` ndarray[{{dest_type2}}, ndim=2] nobs, mean

167

170

171

assert min_count == -1, "'min_count' only used in add and prod"

172

+

168

173

` if not len(values) == len(labels):

169

174

` raise AssertionError("len(index) != len(labels)")

170

175

`@@ -208,12 +213,15 @@ def group_var_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

208

213

`def group_mean_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

209

214

` ndarray[int64_t] counts,

210

215

` ndarray[{{dest_type2}}, ndim=2] values,

211

ndarray[int64_t] labels):

216

ndarray[int64_t] labels,

217

Py_ssize_t min_count=-1):

212

218

` cdef:

213

219

` Py_ssize_t i, j, N, K, lab, ncounts = len(counts)

214

220

` {{dest_type2}} val, count

215

221

` ndarray[{{dest_type2}}, ndim=2] sumx, nobs

216

222

223

assert min_count == -1, "'min_count' only used in add and prod"

224

+

217

225

` if not len(values) == len(labels):

218

226

` raise AssertionError("len(index) != len(labels)")

219

227

`@@ -263,7 +271,8 @@ def group_mean_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

263

271

`def group_ohlc_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

264

272

` ndarray[int64_t] counts,

265

273

` ndarray[{{dest_type2}}, ndim=2] values,

266

ndarray[int64_t] labels):

274

ndarray[int64_t] labels,

275

Py_ssize_t min_count=-1):

267

276

` """

268

277

` Only aggregates on axis=0

269

278

` """

`@@ -272,6 +281,8 @@ def group_ohlc_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

272

281

` {{dest_type2}} val, count

273

282

` Py_ssize_t ngroups = len(counts)

274

283

284

assert min_count == -1, "'min_count' only used in add and prod"

285

+

275

286

` if len(labels) == 0:

276

287

` return

277

288

`@@ -332,7 +343,8 @@ def get_dispatch(dtypes):

332

343

`def group_last_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

333

344

` ndarray[int64_t] counts,

334

345

` ndarray[{{c_type}}, ndim=2] values,

335

ndarray[int64_t] labels):

346

ndarray[int64_t] labels,

347

Py_ssize_t min_count=-1):

336

348

` """

337

349

` Only aggregates on axis=0

338

350

` """

`@@ -342,6 +354,8 @@ def group_last_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

342

354

` ndarray[{{dest_type2}}, ndim=2] resx

343

355

` ndarray[int64_t, ndim=2] nobs

344

356

357

assert min_count == -1, "'min_count' only used in add and prod"

358

+

345

359

` if not len(values) == len(labels):

346

360

` raise AssertionError("len(index) != len(labels)")

347

361

`@@ -382,7 +396,8 @@ def group_last_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

382

396

`def group_nth_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

383

397

` ndarray[int64_t] counts,

384

398

` ndarray[{{c_type}}, ndim=2] values,

385

ndarray[int64_t] labels, int64_t rank):

399

ndarray[int64_t] labels, int64_t rank,

400

Py_ssize_t min_count=-1):

386

401

` """

387

402

` Only aggregates on axis=0

388

403

` """

`@@ -392,6 +407,8 @@ def group_nth_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

392

407

` ndarray[{{dest_type2}}, ndim=2] resx

393

408

` ndarray[int64_t, ndim=2] nobs

394

409

410

assert min_count == -1, "'min_count' only used in add and prod"

411

+

395

412

` if not len(values) == len(labels):

396

413

` raise AssertionError("len(index) != len(labels)")

397

414

`@@ -455,7 +472,8 @@ def get_dispatch(dtypes):

455

472

`def group_max_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

456

473

` ndarray[int64_t] counts,

457

474

` ndarray[{{dest_type2}}, ndim=2] values,

458

ndarray[int64_t] labels):

475

ndarray[int64_t] labels,

476

Py_ssize_t min_count=-1):

459

477

` """

460

478

` Only aggregates on axis=0

461

479

` """

`@@ -464,6 +482,8 @@ def group_max_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

464

482

` {{dest_type2}} val, count

465

483

` ndarray[{{dest_type2}}, ndim=2] maxx, nobs

466

484

485

assert min_count == -1, "'min_count' only used in add and prod"

486

+

467

487

` if not len(values) == len(labels):

468

488

` raise AssertionError("len(index) != len(labels)")

469

489

`@@ -526,7 +546,8 @@ def group_max_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

526

546

`def group_min_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

527

547

` ndarray[int64_t] counts,

528

548

` ndarray[{{dest_type2}}, ndim=2] values,

529

ndarray[int64_t] labels):

549

ndarray[int64_t] labels,

550

Py_ssize_t min_count=-1):

530

551

` """

531

552

` Only aggregates on axis=0

532

553

` """

`@@ -535,6 +556,8 @@ def group_min_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

535

556

` {{dest_type2}} val, count

536

557

` ndarray[{{dest_type2}}, ndim=2] minx, nobs

537

558

559

assert min_count == -1, "'min_count' only used in add and prod"

560

+

538

561

` if not len(values) == len(labels):

539

562

` raise AssertionError("len(index) != len(labels)")

540

563

`@@ -686,7 +709,8 @@ def group_cummax_{{name}}(ndarray[{{dest_type2}}, ndim=2] out,

686

709

`def group_median_float64(ndarray[float64_t, ndim=2] out,

687

710

` ndarray[int64_t] counts,

688

711

` ndarray[float64_t, ndim=2] values,

689

ndarray[int64_t] labels):

712

ndarray[int64_t] labels,

713

Py_ssize_t min_count=-1):

690

714

` """

691

715

` Only aggregates on axis=0

692

716

` """

`@@ -695,6 +719,9 @@ def group_median_float64(ndarray[float64_t, ndim=2] out,

695

719

` ndarray[int64_t] _counts

696

720

` ndarray data

697

721

` float64_t* ptr

722

+

723

assert min_count == -1, "'min_count' only used in add and prod"

724

+

698

725

` ngroups = len(counts)

699

726

` N, K = ( values).shape

700

727