From 126d0ba86fd5206aed4da2aa37baaa95a36a5ea9 Mon Sep 17 00:00:00 2001 From: Umachandar Jayachandran Date: Fri, 2 Nov 2018 12:58:38 -0700 Subject: [PATCH] Modified ML scripts to use external table based views. --- .../data-pool/data-ingestion-sql.sql | 22 ++++++++++-------- .../book-click-prediction-partitioned-py.sql | Bin 7932 -> 11690 bytes .../book-click-prediction-partitioned-r.sql | Bin 7454 -> 11194 bytes .../sql/book-click-prediction-py.sql | Bin 6382 -> 10118 bytes .../sql/book-click-prediction-r.sql | Bin 6260 -> 9984 bytes 5 files changed, 12 insertions(+), 10 deletions(-) diff --git a/samples/features/sql-big-data-cluster/data-pool/data-ingestion-sql.sql b/samples/features/sql-big-data-cluster/data-pool/data-ingestion-sql.sql index 09c77076..e2843491 100644 --- a/samples/features/sql-big-data-cluster/data-pool/data-ingestion-sql.sql +++ b/samples/features/sql-big-data-cluster/data-pool/data-ingestion-sql.sql @@ -15,11 +15,13 @@ IF NOT EXISTS(SELECT * FROM sys.external_tables WHERE name = 'web_clickstream_cl DISTRIBUTION = ROUND_ROBIN ); GO + -- Currently the create external table operation is asynchronous and there is no -- way to determine completion of the operation. To prevent failures of the insert -- into the external table, wait for few minutes. WAITFOR DELAY '00:02:00'; GO + -- Insert results of a SELECT statement into the external table created on the data pool. -- Store summary results for quick access instead of going to the source tables always. -- @@ -46,16 +48,16 @@ SELECT TOP 10 * FROM [dbo].[web_clickstream_clicks_data_pool] -- SELECT TOP (100) w.wcs_user_sk, - SUM( CASE WHEN i.i_category = 'Books' THEN 1 ELSE 0 END) AS book_category_clicks, - SUM( CASE WHEN w.i_category_id = 1 THEN 1 ELSE 0 END) AS [Home & Kitchen], - SUM( CASE WHEN w.i_category_id = 2 THEN 1 ELSE 0 END) AS [Music], - SUM( CASE WHEN w.i_category_id = 3 THEN 1 ELSE 0 END) AS [Books], - SUM( CASE WHEN w.i_category_id = 4 THEN 1 ELSE 0 END) AS [Clothing & Accessories], - SUM( CASE WHEN w.i_category_id = 5 THEN 1 ELSE 0 END) AS [Electronics], - SUM( CASE WHEN w.i_category_id = 6 THEN 1 ELSE 0 END) AS [Tools & Home Improvement], - SUM( CASE WHEN w.i_category_id = 7 THEN 1 ELSE 0 END) AS [Toys & Games], - SUM( CASE WHEN w.i_category_id = 8 THEN 1 ELSE 0 END) AS [Movies & TV], - SUM( CASE WHEN w.i_category_id = 9 THEN 1 ELSE 0 END) AS [Sports & Outdoors] + SUM( CASE WHEN i.i_category = 'Books' THEN w.clicks ELSE 0 END) AS book_category_clicks, + SUM( CASE WHEN w.i_category_id = 1 THEN w.clicks ELSE 0 END) AS [Home & Kitchen], + SUM( CASE WHEN w.i_category_id = 2 THEN w.clicks ELSE 0 END) AS [Music], + SUM( CASE WHEN w.i_category_id = 3 THEN w.clicks ELSE 0 END) AS [Books], + SUM( CASE WHEN w.i_category_id = 4 THEN w.clicks ELSE 0 END) AS [Clothing & Accessories], + SUM( CASE WHEN w.i_category_id = 5 THEN w.clicks ELSE 0 END) AS [Electronics], + SUM( CASE WHEN w.i_category_id = 6 THEN w.clicks ELSE 0 END) AS [Tools & Home Improvement], + SUM( CASE WHEN w.i_category_id = 7 THEN w.clicks ELSE 0 END) AS [Toys & Games], + SUM( CASE WHEN w.i_category_id = 8 THEN w.clicks ELSE 0 END) AS [Movies & TV], + SUM( CASE WHEN w.i_category_id = 9 THEN w.clicks ELSE 0 END) AS [Sports & Outdoors] FROM [dbo].[web_clickstream_clicks_data_pool] as w INNER JOIN (SELECT DISTINCT i_category_id, i_category FROM item) as i ON i.i_category_id = w.i_category_id diff --git a/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-partitioned-py.sql b/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-partitioned-py.sql index bfad474cbbf900c0be5d459940ef957e4946de21..3339cad36b26008b858e14ed2e649ec207c4f697 100644 GIT binary patch delta 2834 zcmbtWTTc@~6h5_XVnGZ>8U#{V>_Y{)Snf!wDOj)25}+a>gw#@`QQFcLiN37*=97~7 z1;#huNZkeoa}*&>`t`Kqa)c z(b}Or74cuB1$=q5NIWl%R_Q3R-uaXsQH<6g+`xk8n7&EFv>}cHL2VZ!JK|XD32ssw zwNR8+f6y9%&Uv{ehSnNA#QZo~X=okhulV8h8(T1{ zOewr?S*_(=FMH!w6kErHg2H&}i6M+L8w5;1utHP7%Bz>`Ms$hB0Xm z0byrN+-@x9C?{YN@5c(-F+`7}o6>^4Dn?1f@Fm_nk|PH|b~DA%7N0Km`Km7P4z`9Z z)sS+3w5~r^*B`IzPt^4%>-tkQ{Xi|cF=I<%<394_5WbWUJeAT#v`X-Wv(ZQQX&Ig` z!e8BybI#nf#(mIpnl8(KWc77Qn(9tU+$>aN#e1JWqV{eM0nNjhQiFTQVaE{?(Uwj= zC_@h525_U-{8L$EkIEd*&9k${fjDAsEFa&q{j9~~BjXOwPMl`&#$4msiP;Pu6NWqg z?8I*dZ^9*B*MDWpq)W8U5&PIS$xRCIC#(y}O&!NGfGO;xc=W2K73sRhKq6ejUBY)K;)^{SjzrZMm>S=jer z4Bt>vRG?jX&Eh+aWfqgSGR$xS7qK9N!d#Yy@cn?*IQOf1Wx{B!d{##20n~EH`5nT; z8r^CBe~M&=@h*vaP= kB{x4&Hei~(NNLXI9coRClb`5lZC;_Vff=m0LdS>^09TkBHvj+t diff --git a/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-partitioned-r.sql b/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-partitioned-r.sql index 48c02f3eb3cea95c392187a6b462b5855dcedab8..65b247acd8ce7cd8b2c181d53f0edf8c2533b6f7 100644 GIT binary patch literal 11194 zcmd6tZBHD@5y$&2k@6izobFgB7j*d9$tA)Djw8pDV96!2EVMj^fPq^U6X!)fdXoRI zYl`j}W_HKEjS`_{XJ@*rs;leaU)A&Pf9-|cuo8-J69(ZT9P9g+;TK^wY{hS`f1|%5 zEQO(-4#G(Pi|`?w>HAJsx0-Prj&yYrt~90y8)04lxjT&G4s~^-J0txbXx?S~oW?Pu zILGE+>-kk+^pU8-4em=}NAkT6{V3&@eiiz)9)ENFr?93cKkDkO<{gN_yD+coJIy}` z`=WOteuv>)G9UqKc(3dI@K=pLiC4_J(unHlcai^V=?6JxqMp{pYe~HAP9d&8iCfNN zKduf+rmIs?9)^$M>uk=FR$yBHF?=tIo1$|lX`pl@DNF-K;Y{-{br0#Gh`tN0nK`{M zCrU@s@<3xub46Iz{L{G7V?F;`=p6cUwpX%91jclItxaH|1-S6May`Wz! z+7DaLH}Th|WMG$e_5ZNKxoK^mq8X&W)QYjKBK%&TIoaEszPn5GUfMzvRlZx%s#YZP zlGf9!?8jl$=8<-Gq@BQuN2Lv$cK0OHxjt92x=Treh6Zt;Y(K}%pKM7B6-nFd?Qy43 zPdbfy+G*6YPNSZ88ug;nsF$5ay?QX}PP>W6TdW#JZ#pmi@jFG7E!j{{dzSVBYj$7v z`!m|zd3~B_xh)YT_DX6^UVbJDXX0Ov4sOeDWJ5%vn~pY(rMaJ_I5#N}g0uy!yEkB`=zWL{Fw5QS#VnNJPZ{xjIVGVHy%Wn}S3s5=}#*=Tnd< zMZIZA^kNDUr3gC>iC#`Yq7==iA-NOMS+fuC_{ymuDRx2Od1v7$PzGqxk_8k!2-m6ry63YC1Xa+F_v{=2B3hjrJQ5ao?zI2J`Lij`gKdm((Izd{^AomRC`sj<>3dg)8j zZQ&c6!H)b&pU<=J=1Qqe_G-;Ik{+oY|2uWycC<~f`?3%iHDkRZOnmoao#+)`wNBfQ zHS10EC@Q!8sC&NKQ9Lz)Rg1l_826mLz0jOXSqv6U#Y|*4ln(e#yJ7z2RpuMXy&7YJ zS?scYFUbpEmn+JtJIw>h3LmP>p12Ld-|xM4Mp(FLkcs-`Lnc=j8gRLmcVXQY;ow;4 zZhdXdGUWQdiP}69J#2b11+-y(h+Ro-eqh#=q@orl?wVBIjYwP>d8#N;s9V5}+(`%2 z#NNXw!b^G7m->CBz5sS=9FC+>YVY|ng5h06$i=AlCXEo4H-+HaeGyK0; z(%jYFuWD|ub6+7v*xWa(Hf^?KwqW&R-VJD~$?G$`X0G3F>U=Deddm4dtW6$w8ug^p zsHdGqJ?k{;d8bh?I*oeSX_TXmTwip7+T)Gw!ZaheXEw{ybHO6|OzCrd`Bb%Y2YRXc zFy4aIxi^t5hl)h&@}^UH3QtbtpY*ASK6JIXL;eEy?8qlS=z3Srtv>?~@D#nfLF`-6 z@!HT_G;<~X_$hQ5NnD`c$bWo-JXyq9m+_u?M3F^jA4A`fjvI8L*6v4iv)qz<;{WJA zCV*f)Dx`=*=0{eKb9r3&B67Obvx`_$Lm6~I<~fP7AhGKS9AP=-)1}5*E;QeEVN2`# zn|Oa4ySmjK>Wy9dKQ(J|H?Ii4);NzB{PLLLXf)}No(xfWKXwzC*YxbM?zcUo!ft!^ zbj&lV&Tf)X4YqNhE^B*6h1B+pT7|sqnZt`B; zE&fBci}zM7n|?4D)v^;c0+x-g%`X`5ybd%QnpoCA7C>(l&dKHs9k|UfdS3DjltJ+V zJT8}}_rTg%>!whd)f11(=u+ijSSjI>WS{2QIfspVZ)oOaUH%KL=S-_Ojxhx+X9!?R zTDXYF1%K=-dhE$}4TmAqR>U}_xn->qtsny$WQOZ)K1MIYci=Af&>ZzKo*2g79V_QI zk3&DR`f$!N8}}GVozGi&2O|zOdshF)=Z0CW9p1;=80MXIi0~0On8oc3l3(E{EE2Y`KbEjM9wc zrS1W#U~9Mf=0wh-2^G&yS@n3tS=27o5zwM6#eJih3#)3gSh42GjgBTo@H%gg8k|}uX`8Ljdjb#p40Wo>hU&U#rfjy% zS@m*h*LklBt$Tfr715#kLU}9QPc=k}9eFgisOYEV7I6|P_sEi@{_Q<&c#P|l?5&{1n6J&F>TkP%*_&cV{e%HK0V(+3q*Eex@LDzH zxESbu?wzW9toLMy(0MeltbsIE{~$+~4hBBY>A{jLBfT}ndrrKWX7Sl*Th~9Q?y^*S z8}^RITfH@9CtOANN*sQO9*td6?{uL>FT`nW#27-?s>{qL*~~YR!rn6JGLt@v^IVcX zt`tka-ylF+$7{)8Z~4)aqs}Idb)iik#%ZsJww-cOYw(sJXKr1n(4%oGea&Up*g=yePybLf23i9UUvoc_mq>Z{5RQ0gWpJt3#FAt~)GN|&^+cNx2^mhO>L+0}~B zxK*yK*SZ%+Q48cSqlhP_$ii6}=VWWLa4MM%&9chLv2}B#!%vsRyRQ!vi87X*KrPAg z$u-^IvviNXQ*@yoTNC~7bw#%$_wqU_@`y02u_w_k837IR-|L7Fs})}NDD!RjgD9o* j7>gZgjo!Jeqcy8UE%JF6&gPEq+qmZ2^O!|mM@RnwAGF^4 delta 52 zcmdlLKF?}`#pWEo8rI2djAENPM0W5@Hd2|iIZE*Y(`F+T5vIu>w6r#FQQN=_mT%EA GV*~)`wh^2F diff --git a/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-py.sql b/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-py.sql index c2cbfbae51bba93749e0c4bcbb65b279907f4fb0..3d1f6b74cf43b051234499abb6fd7b0eb6ff5edf 100644 GIT binary patch literal 10118 zcmcJVX>S|J5r+G#0_=Ync###IY=re~=YnlXQJm<4M6DAT3xd|Qgd~znQPw&i^3#*N zPZ!JGGn}EuYX$gqb)s^NeCa}*B4W+=jS=!eU2s_(DDufkT?i{D)TPJcyM z4-<{{!&Lu^@F5KJeXFY*tvC%Qx;hIZ%_+i8*w%lZPU5^{U0v(RRQG+Y8^+IhoHLDU zZ2ef{qrmJFQH2}a*TaG2dlkA-%01l`y4#NTTz?H1dd--n*) zU5ejvxR4A;z#gu2-3vcz{#m?Y)krhSvww*EUr9g6u@d#PEne&5ZBGhu{Yl(%9=*6b zD4DL#MR^iFhHtYq>)L^7{cZTGDDH~Rv7~|0iKH+M7=?k>5A_V`p@_ZzUTV*)5jY7sG0R^9yW~L*YoZtxgY3mQ|Dp( z`7Yk=N(N5pQ2&ohoa^@XIhsNGp>~XI72!|%tjgY2_1#{gD`^W&l=<#OtJ;*z>)KDJ zbRNf1nV}dA4fW$W*?BHnf43z)C`sC6Zx36|demyx z<5shtw3_v_)vRZ&W<75;>&4uxTb(8zZ?S3;z3F1+kAG1_*^>=*bY|%+ux9slf4-vA zUDKzImfI3h;%G*#&dUd)FcAN0bZ}dKEgK>dUAMGpEY1Ba#ksp;&@Yt|xX4!^(K z=N8MCEPELe5wE^fqU1%(km%78BuXB;42g*NU#g=N9hM={lO;%$BGED=db$LOQq)_9 zM9-EWQHrq3km&gmBudeI84|r%fP)9(eFVbBM&`lk#)RV=Ka}_D7B|OqSCsXG1V0 zcx_jFyYc%-8E{)zCS}E~m@ALMzQ$jNf9iiTJzJIH|4%bI_4_}LxtVI=q>q}z&~Z+5w% z`*r!+tC`$6b*K4W62XUBvLkN&@bf3%ToK|eYGk5H`H;!gh6Y^5@*!;4;ustYJ+1Do zUWQ!D*V4p9GK@v74F$Ae-G)O+ZPso!lcb{JCc2tb-g8J?nR%{QP^c%siQI}mm9KXy zitt>1^K0FYRQ+R(#^FR7rNUmD#VLGbHVwrK*^O5D&)nIXWW#&U&`AE6n2IeUHB-W1=xp$D`CWrA)9qC z@MqCw(Ruw_^Y88(7U8#==P`>rqnu8go)7t;7rPbAt{Z!(=S^c&@J(Zn7mQKWE|P~D zvxzR%Ueg$LSJRk97}GIl>lw|f1p1~}XFuwR-gOai1J>hK2HL?(t$&PN+iLQdMHh6O z?AowT!WsBV8=LAhSd-Z#J^IV&N1!V6N36+mdr$HWHI9YDk2nV+FpQOCv(emZ)}(#5`cP8ZsMFnEAlKT^StVUn32@yD<5ci6*rLxYt|D1R_+5cwvN5qpy(hMk=>VfS4vp!TuRGN0r)kJ$j+vTWd`(Dv%5D*>=s=CoP z9eC3$*eBneAt5SGPVuGqR-@=k@%Efl&(Bt(J4ak@v*yE`cM~FWnHST@(Yi-=_&J~I zF07)rcSdR7?GZY4A}6sQ|05eaNLE_)AVb;pNWDy}Drur=a^_shE6bWn>#DMtwwuTE zou%mD_~TI4V%3LZ>~#9}#xGIEt}XXh^SXB_svcoTB=;A?3jV&0><3nH-u6Dw>N?Ax zE2eMLh%;cEP~ zx6*5+*ynqke-pI0|4?@t2B@ckgt4;N@kV zSN-0zADKA){$L{Sp#}hpTWqoTVDty#YE{Wpr(ikH>;Ghn{HDR;gXeCG#aJ0ps45%3 zfZ%CeoWYx|Cky>~d#pbVx5%dGPdx6#*pc2%CT}hKeBR-!=Z5U<&i?Z~!6QT2BCDN9 zbYnQw{@>M2CO(Z-y}7m2d5>F|*T(r=uL2F_COfn^f;5I*kfTlgVcp<&m0wrzdkAo* zS+dc#zF|(?WhvJv_Ad9GeqCXm@FIMpJ^wx05^)Ly+J+W2jnmqR)=Qn-X62)76+q#kmf>%xqqn0wuq*vBi$V7~^U3jm(SgWAxho^sloqHW!9DiwZBL~p(g6{;?$ z($!k}V|W3wC5^vPRwd7jB6=aZ($Q~2pUN$#|M3&`Ek!0MwUd)7&*|($IXu=NX;-#U^$~o(d$NkD>=Z2!J&huMm~aq1b^7UVWZ_*!o*m&7!%U7a>mv((x}hC( z^?@RB>&t(SqYMi1%7WgnOEM7VVM@112*69nb3FJQlW(GJgzz7NykD Qwb(K5-POG25t7{bKW9xguK)l5 delta 54 zcmZqke`h$sVsZ}Op3MgQc`TEssmX0VCj5wdvXY$0WCaD5$qou^lh4UlY<5!6VcNV& K=@v6k$p-*48Wd#! diff --git a/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-r.sql b/samples/features/sql-big-data-cluster/machine-learning/sql/book-click-prediction-r.sql index f7771de2726e85b465c340370c2557324c1683cd..298438da8d83c3769881e7a5c78406ede76b5ae0 100644 GIT binary patch literal 9984 zcmcJVX>S`x5{CP$0^~mo5+tH`vqF5^*<>MGQWRx%L8Oumf`FiPt=6)nP_z~2Lw@@1 zzE2lR-8~#qW19g%a*pn*>gqb)s+s@&cRzH)TByQR7=*KMr2mh@M`1ne#D9+ercV`C z!ccbyVWi(Gya}iJzt+)f%{U5&Iyw#)8dHU>u&H0J4&%539bM_lNaq91JCEO!IA#>* z*!)Y~zX*&z6jivveI@KkzGtBqrQFe3rL)a=&hZanLwA1E(a)OqLKJ=pOFF*R{1>4w zdS~Ky5MD|KBw!7%bleaB*7)Oi#HkYge0X;Zvb#M`b^;`*z&4Zg5 zI+T_N8e^KP!m8$<#FZZD{udgzD48#$2ln?wU%&c5_ugt`^PGEA?%vV$-i_RMb+)Gc zu=RWy&$cB4yVTY1LBY9cZSSHPq(9e+v8^h6qwk{ZZBhU2C3+=op^1|3PPD2u$-JWV zbc+2rh}t~V&W^MbSn+6V!=~MR$@Eg+3t8Q{q(MW2xKFm9v*zD!Np}lLr`g-RR-^8> z8ug&nsE4gaJ!&=TajQ{JT8(--G3r{oiN{;48b)t=IrhhI6;XC%Lmll|+6%1Neci7w zXm^+NZKCD2M3mScQ)}|_Q&Bh-|9W(ATYfGZA`)G-v}r8O{Vc_~+hWlBN(o&eJ${GZ z-|lm>-?2KZv=RYT(rv+c|!6 zJ%D?fbFXRPAfjh7E9**>Oj9OJJKp_>FJDO(Q1l>Ns#2hW0C&4oWI@}XobNE=#B*^+ z4p1so^1aGYetG}zQ9qM)*M<<~m1H;)MJ$SyUF!dC_)?!r96_B{wNa_D(h|M&Bu*>`iLG)?xhW*kY6)Q*2n9k?596YRb!1V+tRuLu+0eykI{;w$U4eymxq zqDN7=f%mNv?1bgGpX}S2uAj>quvV&Jy#7Gz<3DNIe8(%u=aOkX`ny@T#H_CtaP=$wq_Y}-Cju( zL&=ZkCdyqRu9!npECtNL(3tq6ko_6TpsKi$C?Q_b00GMBeh5 z&M#E+V}r)wP#UGKUK&RyJcsDF9QEF$5hCuk(D`&fgA-o1D_VQ8n%s^wyW0D8&F!@A z8Kmf$`(0_TW=m!ZRvYG(ep9_&pW*dz{d!YZzve zSpG&&iD*IxiYw$AFv^~M?2V4Qx^KM{cz}KAvkhX8iSE>v<`N4o#2>$dE+g@CdV~C8 z0X{92w6IhTe+Wq8acfd z*RxpvLK!4LmN<^GAhGL-oNqbe-KE9{&otk5VMpuxk9dC_JFw*rb!M*p56!y0^Hznw zYMlE&XC4C_IVK&_KOxHYW984hrhE5vecC;$;c53C%(zE|*iJGkw`m-xp{CuVnwfTw zx`I6GnZhwQT}XC)FwVzl4mK1B1JLhPqm0&kzWU|#Rh&CvNI0W)}t$%vsW}V=Zy)j0?lMN@9 zD@32m;^^6;nTyRtPmYD$5-koGmHd9fX)ft(8aa;kmC88-J1T{gHG@t1j=PXPVB^^>TixwKS^kb9E~9`}z6|A>@A2mAzPv;P#y? z!Jft?&RDiZqI&lA4no~03@_v!V;S9_g;{h(IEgG>_x=R8P#+CbhOTus?76!qK8GL>*ZOwQ{bN;;o-%}IhkDGlXT(cI;CO|j!c{00RZ)Y$PJ-@Xe~_xsWg>peg;LeBHgZGBv#&!v#0+enkAxLtXDnveBLtS>6L z^;x7q8bc|_(WY*}#^~>r?=|oy0`HEP{TglS-R0C>ma?L;r>H;cy$0(+SK)JU_-C{r zA`!T<4J|4lr?nMrmRhyV%=_8Q=aRyn1Gkw;_2fKPBBICV+=E!hGs$4@;ZTW#uJM~T zw5h(F_L^v0ADWt;_ju@vx1mC1=2UudZD(-@ku7O_Lpc-sxQN(xJ~D7BN5O3>+SML9geAZbsTx{Wl{j83Ji6`}FbjYK=LijnM;9VTWPMLaQ~9J<+b zvNvSmJ;j?Xp%6nvmIIeH&pIzyaf(0q>8g15^o1hP>ciiJa*w`NUIk}ui2ipv>V!{P tMTAc@_Bh%lc?BBgzn2jel*p}pl=(XRLzGfi)ndn_ch|F-M@TZ}{{gu{807!} delta 37 rcmZqh`(iM`Vsj3k2J2)tMzPIXgiZJ+Pg2v`9Hz8^86+}UP3;x{^WF?S