From eb44623e24af13026377055df2275988a7b6716a Mon Sep 17 00:00:00 2001 From: Sander Roosendaal Date: Thu, 18 Apr 2024 08:31:08 +0200 Subject: [PATCH] some more polars, passing test_uploads --- rowers/dataprep.py | 92 ++++++++++++++------------ rowers/interactiveplots.py | 2 +- rowers/tasks.py | 2 +- rowers/tests/test_imports.py | 2 +- rowers/tests/testdata/testdata.tcx.gz | Bin 4000 -> 4000 bytes rowers/utils.py | 27 ++++---- rowers/views/exportviews.py | 2 +- 7 files changed, 68 insertions(+), 59 deletions(-) diff --git a/rowers/dataprep.py b/rowers/dataprep.py index c5195951..f6de443b 100644 --- a/rowers/dataprep.py +++ b/rowers/dataprep.py @@ -8,6 +8,7 @@ from rowers.datautils import p0 from scipy import optimize from rowers.utils import calculate_age import datetime +import gzip from scipy.signal import savgol_filter from rowers.opaque import encoder from rowers.database import * @@ -27,6 +28,7 @@ from fitparse import FitFile import itertools import numpy as np import pandas as pd +import polars as pl from zipfile import BadZipFile import zipfile import os @@ -219,18 +221,18 @@ def check_marker(workout): ids.append(w.id) gms.append(gmstandard) - df = pd.DataFrame({ + df = pl.DataFrame({ 'id': ids, 'gms': gms, }) - if df.empty: # pragma: no cover + if df.is_empty(): # pragma: no cover workout.ranking = True workout.save() return workout - indexmax = df['gms'].idxmax() - theid = df.loc[indexmax, 'id'] + theid = df.filter(pl.col("gms") == pl.col("gms").max())['id'][0] + wmax = Workout.objects.get(id=theid) # gms_max = wmax.goldmedalstandard @@ -326,7 +328,7 @@ def workout_summary_to_df( goldstandarddurations.append(int(goldstandardduration)) rankingpieces.append(w.rankingpiece) - df = pd.DataFrame({ + df = pl.DataFrame({ 'ID': ids, 'date': startdatetimes, 'name': names, @@ -420,19 +422,21 @@ def calculate_goldmedalstandard(rower, workout, recurrance=True): try: df = pl.read_parquet(cpfile) except: + df = getsmallrowdata_pl(['power'], ids=[workout.id]) background = True if settings.TESTING: background = False - df, delta, cpvalues = setcp(workout, background=background) - if df.empty: - return 0, 0 - df = pl.from_pandas(df) + if recurrance: + df, delta, cpvalues = setcp(workout, background=background) + if df.is_empty(): + return 0, 0 + else: + return 0,0 if df.is_empty() and recurrance: # pragma: no cover df, delta, cpvalues = setcp(workout, recurrance=False, background=True) - if df.empty: + if df.is_empty(): return 0, 0 - df = pl.from_pandas(df) age = calculate_age(rower.birthdate, today=workout.date) @@ -459,7 +463,7 @@ def calculate_goldmedalstandard(rower, workout, recurrance=True): if getrecords: # pragma: no cover durations = [1, 4, 30, 60] distances = [100, 500, 1000, 2000, 5000, 6000, 10000, 21097, 42195] - df2 = pd.DataFrame( + df2 = pl.DataFrame( list( C2WorldClassAgePerformance.objects.filter( sex=rower.sex, @@ -467,7 +471,7 @@ def calculate_goldmedalstandard(rower, workout, recurrance=True): ).values() ) ) - jsondf = df2.to_json() + jsondf = df2.write_json() _ = myqueue(queuelow, handle_getagegrouprecords, jsondf, distances, durations, age, rower.sex, rower.weightcategory) @@ -511,21 +515,21 @@ def calculate_goldmedalstandard(rower, workout, recurrance=True): def setcp(workout, background=False, recurrance=True): try: filename = 'media/cpdata_{id}.parquet.gz'.format(id=workout.id) - df = pd.read_parquet(filename) + df = pl.read_parquet(filename) - if not df.empty: + if not df.is_empty(): # check dts tarr = datautils.getlogarr(4000) if df['delta'][0] in tarr: return(df, df['delta'], df['cp']) - except: + except Exception as e: pass - strokesdf = getsmallrowdata_db( + strokesdf = getsmallrowdata_pl( ['power', 'workoutid', 'time'], ids=[workout.id]) - if strokesdf.empty: - return pd.DataFrame({'delta': [], 'cp': []}), pd.Series(dtype='float'), pd.Series(dtype='float') + if strokesdf.is_empty(): + return pl.DataFrame({'delta': [], 'cp': []}), pd.Series(dtype='float'), pd.Series(dtype='float') totaltime = strokesdf['time'].max() maxt = totaltime/1000. @@ -565,23 +569,27 @@ def setcp(workout, background=False, recurrance=True): dologging('metrics.log', traceback.format_exc()) return pd.DataFrame({'delta': [], 'cp': []}), pd.Series(dtype='float'), pd.Series(dtype='float') - delta = pd.Series(np.array(response.delta)) - cpvalues = pd.Series(np.array(response.power)) + delta = pl.Series(np.array(response.delta)) + cpvalues = pl.Series(np.array(response.power)) powermean = response.avgpower - - - df = pd.DataFrame({ + df = pl.DataFrame({ 'delta': delta, 'cp': cpvalues, 'id': workout.id, }) - df.to_parquet(filename, engine='fastparquet', compression='GZIP') + df = df.drop_nulls() + + with gzip.open(filename, 'w') as f: + df.write_parquet(f) + + + #df.to_parquet(filename, engine='fastparquet', compression='GZIP') if recurrance: goldmedalstandard, goldmedalduration = calculate_goldmedalstandard( - workout.user, workout) + workout.user, workout, recurrance=False) workout.goldmedalstandard = goldmedalstandard workout.goldmedalduration = goldmedalduration workout.save() @@ -737,24 +745,24 @@ def fetchcp_new(rower, workouts): except: # CP data file doesn't exist yet. has to be created df, delta, cpvalues = setcp(workout) - df['workout'] = str(workout) - df['url'] = workout.url() + df = df.with_columns((pl.lit(str(workout))).alias("workout")) + df = df.with_columns((pl.lit(workout.url())).alias("url")) data.append(df) if len(data) == 0: - return pd.Series(dtype='float'), pd.Series(dtype='float'), 0, pd.Series(dtype='float'), pd.Series(dtype='float') + return pl.Series(dtype='float'), pl.Series(dtype='float'), 0, pl.Series(dtype='float'), pl.Series(dtype='float') if len(data) > 1: - df = pd.concat(data, axis=0) + df = pl.concat(data) + + #df = df.to_pandas() try: - df = df[df['cp'] == df.groupby(['delta'])['cp'].transform('max')] - except KeyError: # pragma: no cover + df = df.group_by(pl.col("delta")).agg(pl.max("cp"), pl.max("workout"), pl.max("url")).sort("delta") + except (KeyError, ColumnNotFoundError): # pragma: no cover return pd.Series(dtype='float'), pd.Series(dtype='float'), 0, pd.Series(dtype='float'), pd.Series(dtype='float') - df = df.sort_values(['delta']).reset_index() - df = df[df['cp']>20] - + df = df.filter(pl.col("cp")>20) return df['delta'], df['cp'], 0, df['workout'], df['url'] @@ -810,16 +818,16 @@ def update_rolling_cp(r, types, mode='water', dosend=False): delta, cp, avgpower, workoutnames, urls = fetchcp_new(r, workouts) - powerdf = pd.DataFrame({ + powerdf = pl.DataFrame({ 'Delta': delta, 'CP': cp, }) - powerdf = powerdf[powerdf['CP'] > 0] - powerdf.dropna(axis=0, inplace=True) - powerdf.sort_values(['Delta', 'CP'], ascending=[1, 0], inplace=True) - powerdf.drop_duplicates(subset='Delta', keep='first', inplace=True) - if powerdf.empty: + powerdf = powerdf.filter(pl.col("CP")>0) + powerdf = powerdf.fill_nan(None).drop_nulls().sort(["Delta", "CP"]) + powerdf = powerdf.unique(subset=["Delta"], keep="first") + + if powerdf.is_empty(): return False res2 = datautils.cpfit(powerdf) @@ -1062,7 +1070,7 @@ def checkbreakthrough(w, r): workouttype = w.workouttype if workouttype in rowtypes: cpdf, delta, cpvalues = setcp(w) - if not cpdf.empty: + if not cpdf.is_empty(): if workouttype in otwtypes: try: res, btvalues, res2 = utils.isbreakthrough( diff --git a/rowers/interactiveplots.py b/rowers/interactiveplots.py index 072fe65c..80517b36 100644 --- a/rowers/interactiveplots.py +++ b/rowers/interactiveplots.py @@ -1380,7 +1380,7 @@ def interactive_chart(id=0, promember=0, intervaldata={}): columns = ['time', 'pace', 'hr', 'fpace', 'ftime', 'spm'] datadf = dataprep.getsmallrowdata_pl(columns, ids=[id]) - if datadf.is_eompty(): + if datadf.is_empty(): return "", "No Valid Data Available" datadf = datadf.fill_nan(None).drop_nulls() diff --git a/rowers/tasks.py b/rowers/tasks.py index 1a54e8c3..f2a7355f 100644 --- a/rowers/tasks.py +++ b/rowers/tasks.py @@ -562,7 +562,7 @@ def handle_sporttracks_workout_from_data(user, importid, source, strokedata = pd.DataFrame.from_dict({ key: pd.Series(value, dtype='object') for key, value in data.items() }) - + try: workouttype = data['type'] except KeyError: # pragma: no cover diff --git a/rowers/tests/test_imports.py b/rowers/tests/test_imports.py index 16346575..d5e9470d 100644 --- a/rowers/tests/test_imports.py +++ b/rowers/tests/test_imports.py @@ -1307,7 +1307,7 @@ class STObjects(DjangoTestCase): self.r.sporttrackstoken = '12' - self.r.sporttracksrefreshtoken = '12' + self.r.sporttroacksrefreshtoken = '12' self.r.sporttrackstokenexpirydate = arrow.get(datetime.datetime.now()+datetime.timedelta(days=1)).datetime self.r.save() diff --git a/rowers/tests/testdata/testdata.tcx.gz b/rowers/tests/testdata/testdata.tcx.gz index f0e0d26a9ef6b916c8a782de10c0aa02a2355759..132867e104e5cbf19a16c3afdbd4a693272e648a 100644 GIT binary patch literal 4000 zcmV;R4`1*fiwFqk6(D8;|8!+@bYx+4VJ>uIcmVC4TW=Ic7J%RR6&4T4!zwlBQl~DC zaYP`iMFKVoD4VxcV&?BpuloLcf3bM` z_N_VYE>2f#b98pPNMCM_UM)|%&EopQ7jNk9?iBmoyKgwl{iA;MX4Sv{fJ=Ir8lK>)b;iX@kz0nQq>5qCt z=H2GyUmy7Qu7A*-oL#Jzo5dUa>;3=i`lH?c%gamQw_R!w&`l5B{l>e8clP!UK0JW* zGxEnz8{Y3dJ3qVV@7?`p|7v-0clTxb`q9fn0(3us{6O#=dEeqbITIamguemo0~{>U zP9OJw+@0?|>${8o`Ra6e@9tk6dwHRB=<~C_JIRMVTOOUQUv3VC{lVg^t)C8fxZ1rz zPnPM&Zqj|`d--DHr@cSoJ46fzi_1@+cDSEJtvg=+W0~&$x2|9Qe15upHT(DH>6ct? zezg91+4Y-Wmlsb~>(~AA-~E*zkxuh_ck+69h&Ekh_vfc8EUvl4m9L+4?^dU;PybIh zAnuwQxaJa9zJ9dczN-EE!|hwKOk^*&Tz!x)m+1i?66VLa`}+#Ti`DUQy0vBh>g?s= zlWzUGJ4qBj+j@2H?&!at-TijU`PpjSU)`L)NbkTmyo=l2K0oq{S3-;(eoeLq+>*dm z3go|{up@Cjg3lf>T=RUlJ6w)lo3AeW{MSFec_z4a4FB?s>74)S`aX$GJXq}h`tq8K zbkQ$=-u1VbUxe%MYN<1rB<*4q)R7HN1ABhZ6>%q55%&ZU!370(N_ta4?&0R#sV;f9 zQW1A&h*34lry+t1a*wm&?rcTeW0H3(Uc)_>ly`>NaQ78)x3nGSor1d!h6-}WTX4s! zxCf|+yCm*Os5IKVH*O-_>yr1t6>;|_cyJnh-no)y9MgQb`+B$w7$4AT?9gyotqY+EP1{Kd8&&% zRWrrYJ|t64bfz4ll=R)OosZ13R7BpIn3Q^z%#)ZBb!3eIqDsSyv7|jeC-OE_ zL_X>1OsGn}f><(2XK6;GJyt|MntwZS#CZQCb^}Pp_iNo{wOB)9|J~&kLF)@+y0t0g6U! zW;EK@B=cfTdbU6s@`#hs?OT*~n7lJyFleH*Ojy*OpA-3t?D=e=u~p6U%xL07$YY!5 zuP4tr_m%-kzdzw6!Kz#A*f_tsA$BdkspV=Hz7$pYREfoGJiwzTn~8! zBLRY{Nrva5GVhyiKMHv^Y*AG6eBh!Qo^LWg3VCNRqNrp(+M?duCiCNv-{#(`TAZ3-tx8#b-d}vnPDCE7Nh)(0Zwd{)K z+lMyVk3pVc+Y(Eo+gt03>h__<=SLw=#xX=y6FezPeohsYt|IbctT3zQc~`WSc2?x2 zBJ#l`?y3w-TehN7pvn9gE)pb8_ZHzMdKz z&bhbFBWgx_L70rpH+z1Vd^8jw1{L|(bo=Q?Y=m;R6jhW~7@BTBeVG?)LelM3GH=Uf zUB^~SI|_Mkgmn=ck+S5Q4{rqWHo-RNtt#tUTqA!|^0p%KWP;aKSBb~T#OIsL50e*T z302iq63W)qwH}z`kPn6sRh<(gmaVbzO}8I|Jc5ZxtnuDDz?OSEeVIo#5{NWWT8Gmi z&&`uFNZ#eFYgH#iM_5$1pA-3-M*C<0F=*oRj;UxBwQuqHG05lK+kipS_tse|ntMAZ z@)gGl>3dr^6EJlc{P z-mUWRnua$_o^$t8mG{;sOE9(!%*m4%&^;Z!Fe$azfcPDuo*PQOL7F z()GX$q3Eh*b0S~UJWpbRVANz_hKNNoGiXNhyswCSFp^|mQ)v-_n#@lzJ2t*=xFcS zTGV5Z-{s!w=J^ncT6J?GUy&zAl7+?uD)KQD&B<|Xo*#w0Gri5Z8t}0q^ql5;0@YlP5GI4ZwgJB;sdt6}Nj29a zmMyoL6MC))easqu)Jc8WRSO&XanR>ZsE$>cS(s+5rs2;3z2pI36|zMXDjKbylbTnN zamT@s$Dq>i(KT=0O^v%y5BfBAgUkK#H0XWNKDf8a$b@Fx4WhROfLJH>!A=G}Hsfv( zy-(gS5frb(2s&XnrM|S9RgR>xTEIH8v}h()+6f5ZjiF$ze)Y%(Tj0ji=N81 zN}H3^Lq+I=5g}CrY@}on`i6emE*+mcV_GMoGVUlT2r`lOIoOt^A4G2rdtLjSyfDUO zIwv)|ei*$s6eQ>dd=Etn>#?bMW1z<*z{-LI;Pa}Yfnbh@rtNU`X`rL!9F zsUq~wfF)f-w7Gs1^n@l7fhs|ZX*z3~H6Q*kdTRnJ zt8bXyWSXY6ES+KW-gqml((u`qUFU{c&D|L2bB3j$uD+h7Y&#WdYThX5*$80NNxch` z8St?sXbqxw#-eU5Lu4O|Hcq2v*N=ie8j5Ij8h$7nzTxK08w0(i+&@d_&EpuGnm7Fc z---cI7G7C|$;9hxOUAA!Xls+kh7_X;JyDzMr;lF1MCHQ8Xqk#e?wSvO9Q16WL(-wQ zJelRqnj?1@{kCPISE+ecY;);M-;uj*nTTFD*E^X^f>uk*96%qoEfYbvIKR!SI;w2^PnR1 zcH3G(rRGWDHcE4+oZMUA`GRbw3VAxpj24fviXBlT^-A4Jb+g0L=LFTobo@LSM! z9Q4*WL{xRvkZ7AXZ~Bg0_Qo?=RnXRfn@p-kQ}YJVCp*nL)Me>7bZx+&dTw4q&qgRa zX*B3*IwcsbjnXiBXDsAIOAUGsZNQ(tBiF`+o=8=AKYLZ4T%1`Im%94CXmtzSQko=p(a zMefpzQc}auohTam5zss1*+!L1CpawW(wP-{sR(_{YubWN>OB@6@O>-67z4eKaTbFr zODBY)mA1D+-)ioL(Mztr395|T5Q|pY-fF;ay>y1rhrFh(E^-&4=wfHH((r4^(h=CM zWDV$5j@*bc8J7;UzIlV_w<}re*13yRHfRfNt{(-x&($|d^z0|Ix@DW|2hr!1tQkq= z&5PC-t-hhwH*Xa5yj{dxW!y!HZFS3ud-F1SmkY0y=v^!-^{r9rIOt<8ywchAF+gjc zHgRuWLcd+fS{b>E5nE^M8KAc&&Gf1TPza#)$h|3gKNk8ATR(0%SUm2|FF!mz>zA9u z?@sz3e>qv6F4z6u_uEb1?{)niyz_sindk_X?nQv_j{6@E7P|xP?9TM@H|EcmpH{o? z4<1}3j%ODi_I=u=x98&Q?WSA5>`qR$Pb+;~t&d-JeV5ML?cT0L<Ch_==fjpK-G9<0v$Npp=dX8L ztNpZ42sxO}u)7Z20Zc#$rBc;|l&JZ&t* GfB^v86B%Ow literal 4000 zcmV;R4`1*fiwFp#;U8uK|8!+@bYx+4VJ>uIcmVC4NpBoC7J%>m6@m}RVG#CP__!#t zf^35^U}u8Z$mBMtNFBLnq?XWKUh?mU?6zguRv~`U6u~Oc2SZ(5Th~{Qd^?Z6d2?}o z@M^i)t}fS)9-@JV2j3h&IC{S6R_oRJ?EB@qUv7Tr`tJ8`yF5rcueXmLzUcd_!^PtD z>(}P2+gz;H=H&8Xk-prXyjWgz+r{mNFJ95Z{V5K+H=l8qZ%_Kwt5yH!k8k?rI^W=p z6TI23J~+c`HkX^T1p(mVx2I2@?N0I4YQ62d^~o|FwOt+V&->)^r0aLT_D(l&pg-yr znK#?hzdrEqUH^S|ez{pKw~JT!*Zcp`^(XuNm)DoV&%4whpt~Nr|BVli9~>MUy?p@b zXXKCXH~e|5_haQ|ic`tj*80lKe1z9M*zd}#5IoQaM&!ruT60ge`F zr+525>8=i*_1&g_zPeaGdiYnzUSB93`uwu*&huf6ct@ ze!Tu}+4b9>mz$@n^~-+w@BYS*NT>O=JAb)6Mw>3O|MUG77PnmD#@A1~H>-=67yqXl z5O>QB+;WK&Ue*53F_9Sjc@WBIyTb}Q3hwIU6`^jaW|NPr$&jh!Q;UAwdo%3H^-zTw&M~nSmA768k zF8cA$yZsjPi*OrWEp-Nyq+P6nIPg?r}EUovny_O!98UYq-af^3E_D?!F@KmUiR3Q*gJzP(ki^5AIkM z_W%`fm&82@l}4NQ#!ZBKUGg5dBJSP<4^E@cJ6F<-W10_lUk`Ty;{#fayn8V+5$=_Z z^A*`~!X|=^8ty?#+V69c_fQdcXB-e{^tvcT_4)b8J3>W@okYW|bJHS}CC~RDPj!)} zYNmKPgk;Kz&XhxxlD-?Z^O1R$ipX0NlTxpec@h%}l9w5gj}?(m=A0l0Rd~h~wd37< z$Xizrc?9Dul1k>2@n6z&=jSue3sgj&jb#+pkT2}Ji!&lm6_Ixaoz97a&X=6$Ews^o z6!O7D5~~T%p${efclRRSM*BhXgvJqg)o35elAlk64y7Wij;s+tRB3oImbB;RMBavq z$R|CW3028g5KBhsEX`=N$BM{D6A-*rrQJ31Ga%1oT*Rm%pObS7lAn>^zNXQh#jpdf zndixtC2wa$zM_Z?2Lk|JG~@+JM(6E){Px_ex*_sOeyu}Q8Wu?`>BE~9`HFmamJK6X zO_Y{=(G2@WekAhFfK`$YFd3PzZJw_vN^67h=~YzO^AU`18s60Bc|nszUS-cSK+%ZJ zj7IyKWL~UE&lX5S9&s|deT&i#lXu1o22GTf35(kEb0S}nJ)bQ!wyJra8BLrBd2I9i z?c_P<-ZCI*JUQt_DVcR`UGv+ILS76o1eMGS6^+<5^5c;ACM1bR4SB~+=I=miR| zBtTF#$?#lM=6%!cMgB=55)m z>)2{(MGorgM=%kIHQrkX*m7^DFZ0Mo0+A+4>u@^c zxp{I1$-A6&t?Gp62#f0Wb0S~UXdewA22Fh4F%_+%_ANd?2Kk(O8!%}4-a1P~b8qKF zz9M^`tRaW2n&*WklY7gpetwX=H)xSmqrIb|_WX>2L2m7oF$lM7|>3 zK51(lQMGEB3*4f#GaxSpKvZ4MUE~&}o&IRg-iYWrA$n-!?@GQR-QI#>FDg%tM_W?E zyH_4w)9{ALbMAhs^4|Jn3C5OzIeGG8ymhFF)%m0w6pYmg&d8pxC_Zn4vB;>&$?@rJ zu@fO*n-4G4LtZe~(yG?9@KUsXeoiu9lZGd3ybx-9c;2~^e*0O?^A*|iNq#w!s>a59 z*D5OSNWP*dtwaM%pc?Iin~ZMXytf17g15oSSD3w zMnrDpryj9Mmsax;c@r$dhA(p-7;~w~yWz9qnCP zi+T+5``laIJRd?)t8Px@EAr$>ve1}7MLve2IXSM)^P`Y=CRk!klokQXhH2-7z9LtS z#GG&sL4}^6p}%AQ(`v?hmpk9Os2lGQ7>lOd-v>QZhF&T{FS%g~=mvaClR=NI-}NB+ zU<3(Cg+8fqkLvjTKb+a1p#TslOw7stCQ$EwVhS z&<8gesjmz@yL!+kJIw|Ht3n@rQ8jN)13p%Sp3_`UpqlFu!er3bHsIGJ_0BLLspfjb zvgI~&LeCYUk6FWyI;k(aYGFe^4*J{))v+ow3)76%H2fK$mptIBLbixPMWgj|Qu8V@ z?l>6o7*rZQy5`Nht8o|VL7&EMaJfI82E8xZ2lrkXnb3^8LG;!D5bLBq*vX*BX50;; z_X)kOiAK)mDvF*P`ccqF6RpyvL*R-UchtOjW1vsUdPH5>4N`XeH>saIdNIyx(No!0 zX>*c#s0e*9BBW}7jg%}x-_TFnrQ>sFOzT8c#vLUEK_=2Z2ivmrgXpbcuWO%^7si-O z=cH!W52N>nf&|@w@1ba6JvKFO4D^@;SXpp~6mT*faj~Uk4x(p+PS-UHDHdJ0bXEgC zRfOIdu%ye#W#GxIant7dVf4X7QC8nD+M+I)cSQxf=t0&af2J)z`C>ZKpy_%^L+h8v%?usdr&A z13tC{twHq8Sk$d$i0ot0#%a{-`ccqFLlLb`!w+S{H{85=W1zQ``)BFAc^qR?^QJ%G zTQMNY!Yhj~nRtC|$=DSIZEe!nkYZG!Cu(#3^wA5Ls9d-hEmP6RUGw3OgPu)vNILYE zC$qd+bL0-A-?dEiDmBlFZ7!YZJ92j|6VdDDdMA@f&}wO!1L(u9Wg_U7yg}=w8Fy2+ z>qBmtXhl%v$hDzW-`o{_s0e+?9kg^^HEb*!zTr0DkAWW1FpE=3y+GOJ>UlUs=mSosp-NNp2GGacG7%7?&aRKt(lTd&o=pIyH_t=a z<&2^A(isK4%jl!ZuJ_1AYwn;KccY+>#-lP1-&5HV3}`{yG0-FBhAK*_=dvXj(1Nz3 zpqJcGg>+JHrD)_1nwmEX`e0Zv=%ij;(NvA5=8b}$l5t1M)(YOa=E$A8ODE=*iJ)E6 z)_Xshxt?0kb`X6ovSv_)Z+!60uAc$=Xe3#8I;oF-GIl*RHE#$#lM!om;hO-V+4VC( zFS+mvbm*}t`n*(78}LU!9}F<+`r(J1ygw22)Vz5^=<`a}fI)|z<7Cjc_3MYxvk5}F z$X$9-N^1DI6GcNm0(xgW+o*Es1cxPEI4Z?U()M2HTg}}tddbx{L6wmkV$n+5dky%lm(DQykk_=;MeZULUF>XD8h$NVIs&_u ztO32sksDDa1o=8b}$cZ-;-jJqhYt!_DSZ(c_4a^aN{y^BSqzBNi62Yt+iS30{s258OG zChpBk=yxkwD$?p{iznUH^@pD>`{nld zi}U{LAJ3N;%XNS7%Tr)><@UbKhwM4m_J{CTJ66- z`hJr*UT)s*`%{8B=#7W-cFWW5Kk1U$S#b07r@O7z zX|tF2`|UPdKiZGZfAl=v^zkpt^JTYPCJy=2+edffzK))B>$Ar{9RK)JX0!Wqzy0U* z7XS3};`e2`Gk~uE_WvI(ZhWzS^1H`${9uQAwNB6B`-F12`5`@x-_xa!AN&thJ=+V! GfB^vhXh2{9 diff --git a/rowers/utils.py b/rowers/utils.py index 8ea2225b..7cb1413b 100644 --- a/rowers/utils.py +++ b/rowers/utils.py @@ -6,6 +6,7 @@ from django.utils import timezone import math import numpy as np import pandas as pd +import polars as pl import colorsys from django.conf import settings import collections @@ -342,22 +343,22 @@ def isbreakthrough(delta, cpvalues, p0, p1, p2, p3, ratio): pwr *= ratio - delta = delta.astype(int, errors='ignore').values - cpvalues = cpvalues.astype(int, errors='ignore').values - pwr = pwr.astype(int, errors='ignore').values + delta = delta.cast(pl.Int32) + cpvalues = cpvalues.cast(pl.Int32) + pwr = pwr.cast(pl.Int32) - res = np.sum(cpvalues > pwr+1) - res2 = np.sum(cpvalues > pwr2+1) + btdf = pl.DataFrame({ + 'delta': delta, + 'cpvalues': cpvalues, + 'pwr': pwr, + 'pwr2': pwr2 + }) - btdf = pd.DataFrame( - { - 'delta': delta[cpvalues > pwr], - 'cpvalues': cpvalues[cpvalues > pwr], - 'pwr': pwr[cpvalues > pwr], - } - ) + res = btdf.select(pl.col("cpvalues")>pl.col("pwr")+1)['cpvalues'].sum() + res2 = btdf.select(pl.col("cpvalues")> pl.col("pwr2")+1)['cpvalues'].sum() - btdf.sort_values('delta', axis=0, inplace=True) + btdf = btdf.filter(pl.col("cpvalues")>pl.col("pwr")) + btdf = btdf.sort('delta') return res >= 1, btdf, res2 >= 1 diff --git a/rowers/views/exportviews.py b/rowers/views/exportviews.py index 15bf47f6..593fa23d 100644 --- a/rowers/views/exportviews.py +++ b/rowers/views/exportviews.py @@ -208,7 +208,7 @@ def workouts_summaries_email_view(request): ) df = dataprep.workout_summary_to_df( r, startdate=startdate, enddate=enddate) - df.to_csv(filename, encoding='utf-8') + df.write_csv(filename) _ = myqueue(queuehigh, handle_sendemailsummary, r.user.first_name, r.user.last_name,