R generuje histogram 2D ​​z surowych danych

Question

Mam pewne dane surowe w 2D, x, y jak podano poniżej. Chcę wygenerować histogram 2D ​​z danych. Zazwyczaj dzieląc wartości x, y na przedziały o rozmiarze 0,5 i policz liczbę wystąpień w każdym z nich (zarówno dla xi y w tym samym czasie). Czy istnieje jakiś sposób, aby to zrobić?

> df 
      x   y 
1 4.2179611 5.7588577 
2 5.3901279 5.8219784 
3 4.1933089 6.4317645 
4 5.8076411 5.8999598 
5 5.5781166 5.9382342 
6 4.5569735 6.7833469 
7 4.4024492 5.8019719 
8 4.1734975 6.0896355 
9 5.1707871 5.5640962 
10 5.6380258 6.9112775 
11 4.6405353 5.2251746 
12 4.1809004 6.1127144 
13 4.2764079 5.4598799 
14 5.4466446 6.0130047 
15 5.2443804 5.5421851 
16 5.7521515 5.4115965 
17 4.9667564 5.3519795 
18 4.5007141 6.8669231 
19 5.0268273 5.7681888 
20 4.4738948 6.4241168 
21 4.4116357 5.9819519 
22 4.5741988 6.4595129 
23 4.0839075 6.8105259 
24 4.7154364 6.5054761 
25 4.8986785 5.5511226 
26 5.6262397 6.8996480 
27 4.9034275 5.6716375 
28 4.1872928 5.8387641 
29 4.0444855 5.2554446 
30 4.8911393 5.8449165 
31 5.7268887 6.7100432 
32 5.9136374 6.5059128 
33 4.9481286 6.4679917 
34 4.6198987 5.7462047 
35 5.7306916 6.0613158 
36 5.5818586 6.4533566 
37 5.9240267 6.7748290 
38 4.8160926 6.4942865 
39 5.5456258 5.7911897 
40 4.3075173 6.8165520 
41 4.9654533 5.8904734 
42 5.9581820 5.7692468 
43 4.2417172 5.7990554 
44 5.3670112 5.8252479 
45 5.2932098 5.3983672 
46 5.7456521 6.2563828 
47 4.9398795 5.2879065 
48 4.8526884 6.9827555 
49 5.6135753 6.5219431 
50 4.0727956 5.2647714 
51 6.9418969 5.2584325 
52 5.4189039 5.9936456 
53 3.9193741 6.7099562 
54 5.5885252 5.9680734 
55 5.9581279 5.1843804 
56 4.5724421 6.6774004 
57 4.7700303 6.6083613 
58 5.5490254 6.2431170 
59 4.1668548 5.1017475 
60 5.8948947 6.7646917 
61 6.5501872 5.2803433 
62 5.6011444 4.2733087 
63 5.1337226 6.5225780 
64 5.3153358 6.6164809 
65 3.3815056 6.4077659 
66 3.8405670 5.3677008 
67 6.7036350 4.3090214 
68 3.2446588 4.0965275 
69 4.6563593 7.6868628 
70 5.2382914 7.0020874 
71 6.0771605 6.6232541 
72 3.5672511 6.9333691 
73 5.0865233 4.0778233 
74 5.6743559 5.5177734 
75 4.5759146 7.2210012 
76 5.8203140 4.9787148 
77 3.1106176 6.3937707 
78 4.6310679 4.4731806 
79 6.8237641 6.2679791 
80 3.7653803 5.9188107 
81 5.6139040 5.8586176 
82 6.2016662 5.3514293 
83 3.9362048 5.3217560 
84 6.8005236 7.9247371 
85 5.8030101 7.7492432 
86 6.0143418 6.0709249 
87 6.5734089 7.6112815 
88 4.0569383 5.8440535 
89 4.6825752 7.7926235 
90 4.8204027 6.3106798 
91 3.5001675 6.3156079 
92 3.6521280 7.5155810 
93 5.0945236 4.8206873 
94 3.8732946 5.6771599 
95 6.4812309 5.6082170 
96 5.0308355 7.6877289 
97 5.2193389 7.7133717 
98 6.2239631 5.5387684 
99 4.6501488 7.8559335 
100 3.5389389 5.4594034 
101 5.7139486 4.5008182 
102 3.5425132 7.3562487 
103 6.9950663 6.1036549 
104 5.3801845 5.8903123 
105 4.7629191 5.3394552 
106 4.4102815 7.2312852 
107 5.8723641 4.1410996 
108 3.4691208 4.6383708 
109 4.6479362 5.8562699 
110 3.0315732 6.8614265 
111 5.9456145 4.7497545 
112 4.8461189 4.4730002 
113 4.9606723 5.1099093 
114 4.7802659 7.8147864 
115 5.0189229 6.9308301 
116 6.4738074 5.0539666 
117 5.3725075 5.3282273 
118 6.5374505 7.0508875 
119 4.0907139 5.0855075 
120 5.0557532 5.6449829 
121 6.5483249 7.5800015 
122 3.1083616 7.3697234 
123 3.6119548 7.7639486 
124 6.5157691 7.7152933 
125 4.0305622 7.0521419 
126 3.2197769 6.5881246 
127 4.7570419 6.4564400 
128 4.0063007 6.3981942 
129 4.4412649 7.6576221 
130 5.7348769 6.7601804 
131 3.1312551 5.6295996 
132 3.8627964 7.5817083 
133 5.2008281 5.1082509 
134 6.4229161 6.2816475 
135 2.5241894 6.0802138 
136 7.3759753 5.1090478 
137 3.7284166 5.2045976 
138 3.4404286 6.9708127 
139 6.4237399 5.1363851 
140 4.1829368 5.1612791 
141 5.9500285 5.4765621 
142 3.3555182 6.2627360 
143 7.7691356 5.1877095 
144 4.0684189 7.1663495 
145 7.3929140 7.3819058 
146 2.1659981 7.9796005 
147 4.8539955 7.3108966 
148 5.3932658 4.7116979 
149 3.5610560 4.6096759 
150 5.1883331 6.8068501 
151 6.4233558 7.2955388 
152 7.3308739 6.1761356 
153 3.0710449 4.5296235 
154 7.5400128 5.1559900 
155 3.5776389 5.2057676 
156 4.0402288 7.1487121 
157 2.3107258 6.9816127 
158 7.2065591 7.7307439 
159 5.7577620 5.6652052 
160 2.0595554 7.4373547 
161 7.5994468 4.6216856 
162 4.8053745 3.9113634 
163 7.5769460 7.6019067 
164 5.5362034 8.9270974 
165 3.6713241 3.9060205 
166 6.0612046 7.3862080 
167 6.9205755 7.0792392 
168 6.0892821 6.3248315 
169 2.0532905 4.1545875 
170 3.4086310 3.5510909 
171 5.2148895 5.3266145 
172 4.7638780 7.9240988 
173 6.4717329 5.1350172 
174 7.8287022 4.3457324 
175 6.0299681 3.0952274 
176 3.2760103 5.2730464 
177 2.5729991 7.6594251 
178 3.9403251 7.8928014 
179 6.0021556 7.5313493 
180 7.8561727 4.5092728 
181 3.5818174 4.1140876 
182 7.4972295 5.5313987 
183 6.0138287 6.9369784 
184 3.9257191 7.6395296 
185 3.0462106 3.1347680 
186 6.0630447 4.1847229 
187 7.4878528 5.1004141 
188 4.5145570 4.6389011 
189 6.2777996 4.2647980 
190 3.0166336 7.5755042 
191 2.8791041 6.4471746 
192 7.1029767 7.0061048 
193 2.4526181 6.3373793 
194 5.8762775 7.0746223 
195 7.0609100 8.1256569 
196 4.7252400 8.4829780 
197 3.3695501 8.8786640 
198 3.8505741 6.8260398 
199 5.3573846 6.3864944 
200 3.7039072 8.9951078 
201 4.6216933 6.7890198 
202 7.0390643 5.9458624 
203 5.7172605 6.9083246 
204 2.3814644 8.3856125 
205 2.4432566 3.2618192 
206 4.3881965 6.7022219 
207 5.2583749 7.2432485 
208 5.8540367 8.5154705 
209 6.4267791 4.9593757 
210 5.0668461 3.1358129 
211 2.6845736 8.9880143 
212 7.3094761 5.4049133 
213 4.2176252 5.5062193 
214 5.2025716 4.0798478 
215 6.5592571 8.1852765 
216 2.0417939 7.0843906 
217 7.6045374 7.4870940 
218 6.5971789 8.8641329 
219 5.3541694 7.2176914 
220 2.8314803 6.4831720 
221 2.4252467 4.0918736 
222 6.6804732 6.3624739 
223 6.0325285 6.2057468 
224 2.2751047 5.1275412 
225 5.5397481 5.9890834 
226 4.6420585 4.6013327 
227 7.6385642 5.1722194 
228 6.7378078 5.8246169 
229 5.0647686 7.9219705 
230 2.8672731 6.6371082 
231 7.5487359 4.5727898 
232 1.0837662 7.1788146 
233 5.4483746 6.8955122 
234 9.3085746 4.8330044 
235 3.8484225 6.0133789 
236 2.8034987 3.0023096 
237 2.8952626 8.2623788 
238 5.7666136 3.2158710 
239 6.4978214 5.7866574 
240 1.5184268 5.9791716 
241 2.3836147 8.2897188 
242 4.7318649 6.1174515 
243 5.8544588 7.5056688 
244 9.6776416 6.5151695 
245 0.4319531 4.2470331 
246 0.9810053 8.6452087 
247 7.0819634 3.2488110 
248 1.9084265 6.1122130 
249 7.5096342 3.3495096 
250 8.9564496 3.4960564 
251 5.7603943 6.9091760 
252 0.8801204 7.2744429 
253 1.2183581 6.4264214 
254 1.7761613 7.1199729 
255 3.2490662 7.9935963 
256 3.5420375 8.4801333 
257 8.7709382 3.8011487 
258 8.4770868 3.4749692 
259 0.9965042 6.7509705 
260 7.5049457 5.4313474 
261 9.7261151 6.5909553 
262 5.3893371 4.0194548 
263 9.6154510 7.3117416 
264 1.0327841 6.2376586 
265 4.0064715 3.7333634 
266 6.6941050 3.9452152 
267 4.1317951 9.3322756 
268 9.6481471 7.5330023 
269 7.3474233 1.0310166 
270 3.7343864 4.9808341 
271 9.1412231 2.6655861 
272 5.8414100 0.1329439 
273 2.4837309 7.4956203 
274 2.7983337 1.3563719 
275 0.6335727 7.9273816 
276 7.5566740 0.4321263 
277 8.6182079 0.6038505 
278 0.8928523 8.0131172 
279 5.7375090 8.5275545 
280 0.7864533 3.3954255 
281 8.7808839 1.7059789 
282 9.6621659 0.9215045 
283 8.4894688 8.7667948 
284 1.0358920 7.2505891 
285 0.7378660 0.1173287 
286 9.5485481 3.3186128 
287 6.8987508 9.5480887 
288 7.4105831 5.8809522 
289 6.6984457 5.9509037 
290 1.7878216 9.1932955 
291 0.8443295 5.1662902 
292 0.4498266 8.9636923 
293 2.5068754 5.3692908 
294 9.2509052 2.4204235 
295 4.1333742 6.2581851 
296 6.5510938 7.2923688 
297 4.3412873 3.5514825 
298 4.2349765 9.3207514 
299 2.8730785 7.2752405 
300 2.0425362 6.6513146 
301 6.4498432 7.2949259 
302 5.7453188 6.3263712 
303 7.0501276 8.2238207 
304 4.1915008 1.5325379 
305 8.1307954 7.7681944 
306 7.3156552 6.3031412 
307 4.0302052 0.3039900 
308 3.3740358 2.1386235 
309 8.2055657 2.9112215 
310 1.8817856 7.0503046 
311 7.0820523 6.8739097 
312 5.0725238 6.9951556 
313 1.6246224 5.4126084 
314 3.8865553 7.6398192 
315 6.6727672 8.9677947 
316 9.6048687 7.6757966 
317 2.2006018 9.6385351 
318 9.6403802 7.6438900 
319 0.1267512 0.9048408 
320 1.8160829 7.3193066 
321 9.9318386 9.6068456 
322 2.1275892 7.8034724 
323 1.2232242 1.0695030 
324 3.0198057 3.8964732 
325 3.3265773 8.5865587 
326 5.1519605 7.5068253 
327 0.4137485 5.9223826 
328 1.6896445 0.6071874 
329 1.8534083 2.3554291 
330 1.7182264 9.3488597 
331 6.4165456 9.8670765 
332 7.6270001 2.1839607 
333 8.9867227 5.9565743 
334 6.9185079 0.2440980 
335 6.7359209 7.1072908 
336 3.8034763 5.8466404 
337 3.4583027 6.9041502 
338 1.7983897 1.7108336 
339 6.9184406 6.3632716 
340 1.3538600 6.8484462 
341 3.6731748 4.9846946 
342 5.6139620 8.0637827 
343 9.0991782 2.3051189 
344 1.1220448 8.9624365 
345 2.5925265 8.3673795 
346 9.9977377 8.5423564 
347 5.1761187 5.1240824 
348 5.9330451 9.4141322 
349 6.3337224 6.8055697 
350 2.7287418 5.7100024 
351 6.1022411 2.9733360 
352 2.7331869 3.7135612 
353 6.7394034 8.2721572 
354 2.1757932 9.0574057 
355 5.5011486 6.0124142 
356 4.5301911 2.5865048 
357 5.3137001 0.7062267 
358 0.6959286 3.2395043 
359 5.3494169 6.5742589 
360 7.1472046 6.3821916 
361 0.1749855 0.3954287 
362 6.7709760 6.5212015 
363 7.2983482 3.0086604 
364 0.6147726 9.3336870 
365 7.4417342 2.6836695 
366 1.2769881 4.0591093 
367 9.5342317 5.3443613 
368 0.9368862 1.1391497 
369 8.4271193 8.6641296 
370 6.2000851 8.2987486 
371 2.1768279 6.0684896 
372 5.2021222 6.9222675 
373 0.6095874 8.4759464 
374 2.0217473 9.5844241 
375 4.8080163 6.5052801 
376 3.6099334 0.3272768 
377 6.0132712 7.9920535 
378 4.0495344 8.8153621 
379 6.9646704 7.0375214 
380 3.9211171 2.5994333 
381 4.4749268 1.0517360 
382 1.1683429 3.8710614 
383 1.7618115 0.3513996 
384 1.1257639 5.7446745 
385 3.7351688 8.7376011 
386 4.9234662 7.1975462 
387 7.4899861 7.3846309 
388 7.4170082 2.2885060 
389 0.8526702 3.8160722 
390 4.5907512 8.9315418 
391 7.6996179 9.8409051 
392 0.2340987 4.2906009 
393 2.2502736 1.7819172 
394 3.5679969 1.7419479 
395 5.4214908 5.6001803 
396 3.9965213 9.2021549 
397 3.8610336 2.0462740 
398 5.9490575 4.4422382 
399 9.8897791 5.6402915 
400 6.1153192 4.1236797 
401 5.8906384 2.6153750 
402 8.0582664 2.7137804 
403 7.2969209 2.9362187 
404 3.8673527 1.0837191 
405 3.5647339 6.2338014 
406 9.6490210 0.8373270 
407 0.8133243 6.3393130 
408 2.8760565 9.9462423 
409 3.3836457 7.4451869 
410 4.7772609 2.9141127 
411 8.6635971 5.7812494 
412 5.6192160 1.4764255 
413 9.1334625 8.9822399 
414 0.4662385 6.6440937 
415 3.4503559 4.2064800 
416 0.6704780 2.8508758 
417 0.5211872 4.3109175 
418 7.5615411 9.2851454 
419 7.5081906 4.0019450 
420 8.8851669 9.7323717 
421 7.3856288 8.6152906 
422 9.5926351 0.3993818 
423 1.4478981 1.4845263 
424 5.0425560 1.3501638 
425 0.8952120 7.9407680 
426 6.4732584 7.
427 9.6595225 5.2377876 
428 7.2204625 2.0300222 
429 3.5410601 7.3117738 
430 6.7991771 3.6368291 

Tylko dla wyjaśnienia, chcę dostać coś takiego wykresie poniżej (to fabuła nie trzeba robić nic z moich surowych danych, jestem po prostu pokazując je wyjaśnić problem jaśniej! Jeśli użyję Hist (DF $ x) to pokazują rozkład only.)

Answer 1

ggplot jest elegancka i szybka i ładna , jak zwykle. Ale jeśli chcesz użyć grafiki bazowej (image, contour, persp) i wyświetlić rzeczywiste częstotliwości (zamiast wygładzania jądra 2D), musisz najpierw uzyskać same marginesy i utworzyć macierz częstotliwości. Oto niektóre kodu (niekoniecznie elegancki, ale bardzo wytrzymałe), który robi binowania 2D i generuje wykresy nieco podobne do tych powyżej:

require(mvtnorm) 
    xy <- rmvnorm(1000,c(5,10),sigma=rbind(c(3,-2),c(-2,3))) 

    nbins <- 20 
    x.bin <- seq(floor(min(xy[,1])), ceiling(max(xy[,1])), length=nbins) 
    y.bin <- seq(floor(min(xy[,2])), ceiling(max(xy[,2])), length=nbins) 

    freq <- as.data.frame(table(findInterval(xy[,1], x.bin),findInterval(xy[,2], y.bin))) 
    freq[,1] <- as.numeric(freq[,1]) 
    freq[,2] <- as.numeric(freq[,2]) 

    freq2D <- diag(nbins)*0 
    freq2D[cbind(freq[,1], freq[,2])] <- freq[,3] 

    par(mfrow=c(1,2)) 
    image(x.bin, y.bin, freq2D, col=topo.colors(max(freq2D))) 
    contour(x.bin, y.bin, freq2D, add=TRUE, col=rgb(1,1,1,.7)) 

    palette(rainbow(max(freq2D))) 
    cols <- (freq2D[-1,-1] + freq2D[-1,-(nbins-1)] + freq2D[-(nbins-1),-(nbins-1)] + freq2D[-(nbins-1),-1])/4 
    persp(freq2D, col=cols) 

Przez pewien czas naprawdę zabawy, spróbuj interaktywny, zoomable , powierzchnia 3D:

require(rgl) 
surface3d(x.bin,y.bin,freq2D/10, col="red") 

Answer 2

dwuwymiarowe oszacowania gęstości mogą być wykonane z MASS::kde2d lub KernSmooth::bkde2D (zarówno dostarczanej z rozkładem R zasada). Ten ostatni używa algorytmu opartego na szybkiej transformacji Fouriera na siatce punktów i jest bardzo szybki. Wynik można wydrukować za pomocą contour lub persp lub podobnych funkcji w innych pakietach graficznych.

Korzystanie swoje dane:

require(KernSmooth) 
z <- bkde2D(df, .5) 
persp(z$fhat) 

Answer 3

Jeśli chcesz go z 2d konturu, można również użyć ggplot2 pakietu. Niektóre przykład kodu jest pokazany na to pytanie:

gradient breaks in a ggplot stat_bin2d plot nieznacznie skorygowane:

x <- rnorm(10000)+5 
y <- rnorm(10000)+5 
df <- data.frame(x,y) 
require(ggplot2) 
p <- ggplot(df, aes(x, y)) 
p <- p + stat_bin2d(bins = 20) 
p 

Oto wyjście kodu powyżej:

Answer 4

freq <- as.data.frame(table(findInterval(xy[,1], x.bin),findInterval(xy[,2], y.bin))) 
    freq[,1] <- as.numeric(freq[,1]) 
    freq[,2] <- as.numeric(freq[,2]) 

Jest to prawdopodobnie błędne, ponieważ niszczy oryginalne wskaźniki.

Answer 5

Aby uzyskać kompletność, można również użyć funkcji hist2d{gplots}.To wydaje się być najprostszym do 2D działki:

library(gplots) 

# data is in variable df 

# define bin sizes 
bin_size <- 0.5 
xbins <- (max(df$x) - min(df$x))/bin_size 
ybins <- (max(df$y) - min(df$y))/bin_size 

# create plot 
hist2d(df, same.scale=TRUE, nbins=c(xbins, ybins)) 

# if you want to retrieve the data for other purposes 
df.hist2d <- hist2d(df, same.scale=TRUE, nbins=c(xbins, ybins), show=FALSE) 
df.hist2d$counts 

Answer 6

doszedłem do tej strony http://www.r-bloggers.com/5-ways-to-do-2d-histograms-in-r/ który wymienia jedną z powyższych odpowiedzi. Udostępnia próbki kodu w sumie 5 metod:

hist2d from the library gplots 
hexbin,hexbinplot from the library hexbin 
stat_bin2d from the library ggplot2 
kde2d from the library MASS 
the "hard way" solution listed above.