Djv;dZddlZddlmZddlmZmZdZd&dZ d Z d'd Z d'd Z d'd Z gd Zedk(rFeddZej$eZdZej*gdZej*ddgddgddgddgddgddgddgddgddgddgddgddgg Zej0edezfZee eedee eedeeeeDcgc]}ej6e|ez |c}e edddZe edddZeeeeeeeezz eeedz ddfe eddeeedze dzdzdfe eddeede dzdfe eededej0edezfZee edddee eedee eedeeedz dddfe eddeeedze dzdzddfe eddeede dzddfe eedeeedz de edddeeedze dzdze edddeede dze eedddlmZeej@jCeej*gddz d ej*gd!Z"ee"e ej$d"ddej*gd#Z#ee#e ej$d"ddej*gd$Z$ee$e ej$d"d%dyycc}w)(ausing scipy signal and numpy correlate to calculate some time series statistics original developer notes see also scikits.timeseries (movstat is partially inspired by it) added 2009-08-29 timeseries moving stats are in c, autocorrelation similar to here I thought I saw moving stats somewhere in python, maybe not) TODO moving statistics - filters do not handle boundary conditions nicely (correctly ?) e.g. minimum order filter uses 0 for out of bounds value -> append and prepend with last resp. first value - enhance for nd arrays, with axis = 0 Note: Equivalence for 1D signals >>> np.all(signal.correlate(x,[1,1,1],'valid')==np.correlate(x,[1,1,1])) True >>> np.all(ndimage.filters.correlate(x,[1,1,1], origin = -1)[:-3+1]==np.correlate(x,[1,1,1])) True # multidimensional, but, it looks like it uses common filter across time series, no VAR ndimage.filters.correlate(np.vstack([x,x]),np.array([[1,1,1],[0,0,0]]), origin = 1) ndimage.filters.correlate(x,[1,1,1],origin = 1)) ndimage.filters.correlate(np.vstack([x,x]),np.array([[0.5,0.5,0.5],[0.5,0.5,0.5]]), origin = 1) >>> np.all(ndimage.filters.correlate(np.vstack([x,x]),np.array([[1,1,1],[0,0,0]]), origin = 1)[0]==ndimage.filters.correlate(x,[1,1,1],origin = 1)) True >>> np.all(ndimage.filters.correlate(np.vstack([x,x]),np.array([[0.5,0.5,0.5],[0.5,0.5,0.5]]), origin = 1)[0]==ndimage.filters.correlate(x,[1,1,1],origin = 1)) update 2009-09-06: cosmetic changes, rearrangements N)signal)assert_array_equalassert_array_almost_equalc|}tj|dk(r|tj|df}tjtj||dz|tj||dzfS)Nr)npndimshaper_ones)xkkadds cC:\Crop_Prediction\Backend\crop-ai-system\venv\Lib\site-packages\statsmodels/sandbox/tsa/movstat.py expandarrr3sd D wwqzQbhhqk!n% 55qt#AbggdmAbE&99 ::laggedcP|dk(r|dz}n|dk(rd}n|dk(r | dzdz}nttj|r|}n'|dk(r |dz dz }n|dk(rd}n|d k(r|dz }ntt||}t j |tj ||||z ||z S) anmoving order statistics Parameters ---------- x : ndarray time series data order : float or 'med', 'min', 'max' which order statistic to calculate windsize : int window size lag : 'lagged', 'centered', or 'leading' location of window relative to current position Returns ------- filtered array rrcenteredrleadingrmedminmax) ValueErrorr isfiniterr order_filterr)rorderwindsizelagleadordxexts rmovorderr&:s, h{  y!|Q {{5 %!|Q % %l Q !D   tBGGH$5c :8D=8TX=IY ZZrc&ddlm}tjdd}t |d}t ||tjddd}t |d}t ||t t |dd dd|ddtj dd tjzd }tj|dz}t |d}|j|j||d ||d |jdt |dd }|j|j||d ||d |jdt |dd }|j|j||d ||d |jdy)zgraphical test for movorderrNr r)r r rr)r r"rz.-zmoving max laggedzmoving max centeredrzmoving max leading) matplotlib.pylabpylabr aranger&rlinspacepisinfigureplottitle)pltrxotts rcheck_movorderr6hsI" !BA !5 !Br1 "QrA !5 !Br1xJ?DaeL Qqwr "B r QA !5 !BJJLHHR$r"T"II!" !5j 1BJJLHHR$r"T"II#$ !5i 0BJJLHHR$r"T"II"#rc t|d||S)amoving window mean Parameters ---------- x : ndarray time series data windsize : int window size lag : 'lagged', 'centered', or 'leading' location of window relative to current position Returns ------- mk : ndarray moving mean, with same shape as x Notes ----- for leading and lagging the data array x is extended by the closest value of the array r windowsizer" movmoment)rr9r"s rmovmeanr<s2 Qjc ::rcNt|d||}t|d||}|||zz S)aGmoving window variance Parameters ---------- x : ndarray time series data windsize : int window size lag : 'lagged', 'centered', or 'leading' location of window relative to current position Returns ------- mk : ndarray moving variance, with same shape as x rr8rr:)rr9r"m1m2s rmovvarr@s2( 1aJC 8B 1aJC 8B 2:rc|}|dk(r d}t|dz xsdd|dz zxsd}nr|dk(r.| dz}t|dz |dzzxsd|dz |dzz xsd}n?|dk(r4| dz}td|dz zdz|zxsdd|dz z|z dzxsd}nttj|t |z }t ||dz }t ||jdk(rtj||z|d |St |jt |dddfjtj||z|dddfd |ddfS) anon-central moment Parameters ---------- x : ndarray time series data windsize : int window size lag : 'lagged', 'centered', or 'leading' location of window relative to current position Returns ------- mk : ndarray k-th moving non-central moment, with same shape as x Notes ----- If data x is 2d, then moving moment is calculated for each column. rrrNrrrfull) slicerr rfloatrprintr correlater r) rrr9r"r!r#slavgkernr%s rr;r;s{4H h HQJ'4XaZ)@D A y!| HQJ! +3txz]8Q;5N5VRV W y!| 1hqj>!#D(0DAxzN44G2H2J2Rd Swwz"5#44G Q #D "I yyA~||D!GWf5b99 djj gafo##$a4&A"Q$GGr)r&r<r@r;__main__z! checkin moving mean and variancer() gUUUUUU??g@@g@@@@ @gUUUUUU!@ rKg#q?g| q?gvWUU?gUUU@rr8rrrz- checking moving moment for 2d (columns only))ndimage)rrrrM)axis)drKg?333333?333333?rL?@ffffff@ @@@@@!@#@%@'@)@+@-@/@0@1@2@3@4@5@6@7@8@9@:@;@<@=@>@?@@@@@@@A@A@@B@B@@C@C@@D@D@@E@E@@F@F@@G@G@@H@H@@I@I@@J@J@@K@K@@L@L@@M@M@@N@N@@O@O@ P@`P@P@P@ Q@`Q@Q@Q@ R@`R@R@R@ S@`S@S@S@ T@`T@T@T@ U@`U@U@U@ V@`V@V@V@ W@`W@W@d)krUrVrLrWrXrYrZr[r\r]r^r_r`rarbrcrdrerfrgrhrirjrkrlrmrnrorprqrrrsrtrurvrwrxryrzr{r|r}r~rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrgW@g X@g9X@g`X@gX@gX@g̬X@gX@gX@)dg@g?@g@@g@@gA@gA@gB@gB@gC@gC@gD@gD@gE@gE@gF@gF@gG@gG@gH@gH@gI@gI@gJ@gJ@gK@gK@gL@gL@gM@gM@gN@gN@gO@gO@gP@g@P@gP@gP@gQ@g@Q@gQ@gQ@gR@g@R@gR@gR@gS@g@S@gS@gS@gT@g@T@gT@gT@gU@g@U@gU@gU@gV@g@V@gV@gV@gW@g@W@gW@g⣋.W@g٧뢋W@gEX@gb\tEX@gw.hX@ )rrr)rr)%__doc__numpyr scipyr numpy.testingrrrr&r6r<r@r;__all____name__rFnobsr,rwsarrayavevac_ave2drangevarr>r?x2drSfilters correlate1dxgxdxc)is0rrsA*XG;,[\$j;609Hf 9 z ./ D $A B "(( C R"-*-*-*-*-*-*-*-*-*-*-"- / 0B EE#qu* E '! 12 &rx 01 eBo 66266!AbD)  67 1aA8 4B 1aA8 4B "I "I "r"u*bAqk1 :<bQsAvax!121 ;=b2#a%l1:<  :; %%1Q3-C )Cqj 9: '#"( 34 & 23bAqk31)<>bQsAvax!1231*=?b2#a%l328<>eBqDElc1 BDeBE2#q&(3c1 CEeFbSUmCBH=? '// % %c8288G+