Skip to content

Commit ea39940

Browse files
gagolewsgagolews
committed
more
1 parent 50b5b5d commit ea39940

24 files changed

+246
-778
lines changed

‎.devel/sphinx/weave/suite-v1.Rmd

+2-6
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,8 @@ def summarise_battery(battery):
1111
"~/Projects/clustering-data-v1/.catalogue/%s.csv" % battery
1212
)
1313
x = x.iloc[:, 1:] # skip battery
14-
x["g"] = np.round(x["g"], 2)
14+
x.loc[:, "g"] = np.round(x.loc[:, "g"], 2)
15+
x = x.astype(str)
1516
x = x.rename(columns=dict(
1617
noise="noise points",
1718
k="*k*",
@@ -21,11 +22,6 @@ def summarise_battery(battery):
2122
labels="reference labels",
2223
))
2324
24-
25-
x.iloc[:, 0] = x.iloc[:, 0].astype(str)
26-
x.iloc[:, 1] = x.iloc[:, 1].astype(str)
27-
x.iloc[:, 2] = x.iloc[:, 2].astype(str)
28-
2925
idx = [""]*x.shape[0]
3026
j = 1
3127
for i in range(x.shape[0]):

‎.devel/sphinx/weave/suite-v1.md

-264
Original file line numberDiff line numberDiff line change
@@ -108,58 +108,6 @@ and
108108
Adam Wawrzeńczyk.
109109
Thanks!
110110

111-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 4000
112-
1 2000
113-
2 2500
114-
3 9000
115-
4 3546
116-
5 300
117-
6 1000
118-
7 600
119-
8 1500
120-
9 5000
121-
10 1000
122-
11 1000
123-
12 5000
124-
13 10000
125-
14 5000
126-
15 400
127-
16 2977
128-
17 120
129-
18 120
130-
19 120
131-
20 185
132-
21 185
133-
22 192
134-
23 900
135-
24 1000
136-
Name: *n*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
137-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 2
138-
1 2
139-
2 2
140-
3 2
141-
4 2
142-
5 2
143-
6 2
144-
7 3
145-
8 3
146-
9 2
147-
10 2
148-
11 2
149-
12 2
150-
13 2
151-
14 3
152-
15 2
153-
16 2
154-
17 2
155-
18 2
156-
19 2
157-
20 2
158-
21 2
159-
22 2
160-
23 2
161-
24 2
162-
Name: *d*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
163111
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
164112
|:---|:-------------------|:------|:------|:-------------------|------:|---------------:|------:|
165113
| 1 | **circles** | 4000 | 2 | labels0 | 4 | 0 | 0 |
@@ -220,64 +168,6 @@ see {ref}`sec:battery-g2mg` for an alternative.
220168
We excluded the `DIM`-sets as they are too easy for most algorithms.
221169

222170

223-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 3000
224-
1 5250
225-
2 7500
226-
3 788
227-
4 100000
228-
5 100000
229-
6 399
230-
7 399
231-
8 399
232-
9 399
233-
10 399
234-
11 3100
235-
12 240
236-
13 240
237-
14 373
238-
15 300
239-
16 300
240-
17 600
241-
18 600
242-
19 600
243-
20 5000
244-
21 5000
245-
22 5000
246-
23 5000
247-
24 312
248-
25 6500
249-
26 105600
250-
27 105000
251-
Name: *n*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
252-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 2
253-
1 2
254-
2 2
255-
3 2
256-
4 2
257-
5 2
258-
6 2
259-
7 2
260-
8 2
261-
9 2
262-
10 2
263-
11 2
264-
12 2
265-
13 2
266-
14 2
267-
15 2
268-
16 2
269-
17 2
270-
18 2
271-
19 2
272-
20 2
273-
21 2
274-
22 2
275-
23 2
276-
24 2
277-
25 2
278-
26 2
279-
27 64
280-
Name: *d*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
281171
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
282172
|:---|:----------------|:-------|:------|:-------------------|------:|---------------:|------:|
283173
| 1 | **a1** | 3000 | 2 | labels0 | 20 | 0 | 0 |
@@ -327,30 +217,6 @@ can be accessed, e.g., via the R package
327217
[*FCPS*](https://CRAN.R-project.org/package=FCPS);
328218
see also {cite}`ThrunUltsch2020:fcps`.
329219

330-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 800
331-
1 1000
332-
2 4096
333-
3 4096
334-
4 212
335-
5 400
336-
6 770
337-
7 770
338-
8 400
339-
9 800
340-
10 1016
341-
Name: *n*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
342-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 3
343-
1 3
344-
2 2
345-
3 2
346-
4 3
347-
5 2
348-
6 2
349-
7 2
350-
8 3
351-
9 2
352-
10 2
353-
Name: *d*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
354220
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
355221
|:---|:----------------|:------|:------|:-------------------|------:|---------------:|------:|
356222
| 1 | **atom** | 800 | 3 | labels0 | 2 | 0 | 0 |
@@ -378,46 +244,6 @@ in {cite}`graves`.
378244
The dataset consist of 200–1050 observations in 2 dimensions.
379245
They came with no reference labels, so we had to create them ourselves.
380246

381-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 200
382-
1 1000
383-
2 1000
384-
3 1000
385-
4 1000
386-
5 1000
387-
6 250
388-
7 1000
389-
8 1000
390-
9 1000
391-
10 1050
392-
11 1030
393-
12 1030
394-
13 250
395-
14 250
396-
15 300
397-
16 300
398-
17 280
399-
18 280
400-
Name: *n*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
401-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 2
402-
1 2
403-
2 2
404-
3 2
405-
4 2
406-
5 2
407-
6 2
408-
7 2
409-
8 2
410-
9 2
411-
10 2
412-
11 2
413-
12 2
414-
13 2
415-
14 2
416-
15 2
417-
16 2
418-
17 2
419-
18 2
420-
Name: *d*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
421247
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
422248
|:---|:--------------------|:------|:------|:-------------------|------:|---------------:|------:|
423249
| 1 | **dense** | 200 | 2 | labels0 | 2 | 0 | 0 |
@@ -474,24 +300,6 @@ Datasets from multiple sources:
474300
* `square` – a dataset of unknown/unconfirmed origin
475301
(🚧 help needed 🚧).
476302

477-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 8000
478-
1 8000
479-
2 10000
480-
3 8000
481-
4 2309
482-
5 150
483-
6 105
484-
7 1000
485-
Name: *n*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
486-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 2
487-
1 2
488-
2 2
489-
3 2
490-
4 2
491-
5 4
492-
6 4
493-
7 2
494-
Name: *d*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
495303
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
496304
|---:|:---------------------|------:|------:|:-------------------|------:|---------------:|------:|
497305
| 1 | **chameleon_t4_8k** | 8000 | 2 | labels0 | 6 | 761 | 0.25 |
@@ -519,24 +327,6 @@ not clustering algorithms. Most clustering algorithms find them
519327
problematic; due to their being high-dimensional, it is difficult
520328
to verify the sensibleness of the reference labels.
521329

522-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 336
523-
1 214
524-
2 351
525-
3 208
526-
4 2310
527-
5 569
528-
6 178
529-
7 1484
530-
Name: *n*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
531-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 7
532-
1 9
533-
2 34
534-
3 60
535-
4 19
536-
5 30
537-
6 13
538-
7 8
539-
Name: *d*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
540330
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
541331
|---:|:---------------|------:|------:|:-------------------|------:|---------------:|------:|
542332
| 1 | **ecoli** | 336 | 7 | labels0 | 8 | 0 | 0.65 |
@@ -569,12 +359,6 @@ This battery features two large, high-dimensional datasets:
569359
Both datasets consist of 70,000 flattened 28x28 greyscale images
570360
(train and test samples combined).
571361

572-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 70000
573-
1 70000
574-
Name: *n*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
575-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 784
576-
1 784
577-
Name: *d*, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
578362
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
579363
|---:|:------------|------:|------:|:-------------------|------:|---------------:|------:|
580364
| 1 | **digits** | 70000 | 784 | labels0 | 10 | 0 | 0.03 |
@@ -612,30 +396,6 @@ Gaussian blobs (*k*-means, expectation-maximisation (EM)
612396
for Gaussian mixtures) will naturally perform better thereon than
613397
the non-parametric approaches.
614398

615-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 2048
616-
1 2048
617-
2 2048
618-
3 2048
619-
4 2048
620-
...
621-
139 2048
622-
140 2048
623-
141 2048
624-
142 2048
625-
143 2048
626-
Name: *n*, Length: 144, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
627-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 1
628-
1 1
629-
2 1
630-
3 1
631-
4 1
632-
...
633-
139 128
634-
140 128
635-
141 128
636-
142 128
637-
143 128
638-
Name: *d*, Length: 144, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
639399
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
640400
|:---|:----------------|:------|:------|:-------------------|------:|---------------:|------:|
641401
| 1 | **g2mg_1_10** | 2048 | 1 | labels0 | 2 | 0 | 0 |
@@ -801,30 +561,6 @@ Generator: <https://github.com/gagolews/clustering-data-v1/blob/master/.devel/ge
801561
Just like in the case of {ref}`sec:battery-g2mg`, we recommend
802562
that these datasets be studied separately from other batteries.
803563

804-
<string>:19: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 2048
805-
1 2048
806-
2 2048
807-
3 2048
808-
4 2048
809-
...
810-
139 2048
811-
140 2048
812-
141 2048
813-
142 2048
814-
143 2048
815-
Name: *n*, Length: 144, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
816-
<string>:20: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '0 1
817-
1 1
818-
2 1
819-
3 1
820-
4 1
821-
...
822-
139 128
823-
140 128
824-
141 128
825-
142 128
826-
143 128
827-
Name: *d*, Length: 144, dtype: object' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.
828564
| | dataset | *n* | *d* | reference labels | *k* | noise points | *g* |
829565
|:---|:----------------|:------|:------|:-------------------|------:|---------------:|------:|
830566
| 1 | **h2mg_1_10** | 2048 | 1 | labels0 | 2 | 0 | 0 |

0 commit comments

Comments
 (0)