Fix Pipeline

docs/notebooks/sklearn_wrapper.ipynb

@@ -464,7 +464,7 @@
    ],
    "source": [
     "model=SklearnWrapperLdaModel(num_topics=15,id2word=id2word,iterations=50, random_state=37)\n",
-    "clf=linear_model.LogisticRegression(penalty='l2', C=0.1) #l1 penalty used\n",
+    "clf=linear_model.LogisticRegression(penalty='l2', C=0.1) #l2 penalty used\n",
     "pipe = Pipeline((('features', model,), ('classifier', clf)))\n",
     "pipe.fit(corpus, data.target)\n",
     "print_features_pipe(pipe, id2word.values())\n",

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102590

@@ -0,0 +1,26 @@
+Path: cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!fs7.ece.cmu.edu!europa.eng.gtefsd.com!gatech!news-feed-1.peachnet.edu!umn.edu!msus1.msus.edu!mhd.moorhead.msus.edu!HOLSEND
+Newsgroups: rec.sport.baseball
+Subject: Re: DESIGNATED HITTER RULE
+Message-ID: <1993Apr5.135452.2952@msus1.msus.edu>
+From: holsend@mhd.moorhead.msus.edu
+Date: 5 Apr 93 13:54:51 -0600
+Reply-To: holsend@mhd.moorhead.msus.edu
+References: <1993Apr4.221228.17577@bsu-ucs>,<ekdfc.14.0@ttacs1.ttu.edu>
+Organization: Moorhead State University, Moorhead, MN
+Nntp-Posting-Host: 134.29.97.2
+Lines: 14
+
+
+
+
+In article <ekdfc.14.0@ttacs1.ttu.edu>, ekdfc@ttacs1.ttu.edu (David Coons) writes:
+>In article <1993Apr4.221228.17577@bsu-ucs> 00ecgillespi@leo.bsuvc.bsu.edu 
+>writes:
+>>I AM DOING A POSTITION PAPER ON THE DESIGNATED HITTER RULE. ANY INFORMATION
+>>OR EVEN OPINIONS WOULD BE GREATLY APPRECITATED. 00ECGILLESPIE "MAGIC"
+>
+>Should be rescinded.  The rules say baseball is a game between two teams of 
+>nine players each.  Let's keep it that way.
+
+Last weeks Sports Illustrated has a couple of big articles on the designated
+hitter.  It is the 1993 baseball issue.                                                                            Th                        is weeks Sports Illustrated

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102605

@@ -0,0 +1,26 @@
+Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!ogicse!uwm.edu!ux1.cso.uiuc.edu!usenet.ucs.indiana.edu!news
+From: dwarner@journalism.indiana.edu (David J.)
+Newsgroups: rec.sport.baseball
+Subject: Re: 1993 NL East Champion PHILLIES
+Message-ID: <C51J5C.AMx@usenet.ucs.indiana.edu>
+Date: 6 Apr 93 02:34:24 GMT
+Article-I.D.: usenet.C51J5C.AMx
+References: <C4v88x.5Gw@news.udel.edu>
+Sender: news@usenet.ucs.indiana.edu (USENET News System)
+Reply-To: dwarner@journalism.indiana.edu
+Organization: Indiana University
+Lines: 12
+Nntp-Posting-Host: poppy.journalism.indiana.edu
+
+Robert C Hite writes
+> Here are the projected lineups, benches, rotation and bullpen for
+> the 1993 National League East Champion Philadelphia Phillies:
+
+I think the only Phillies in effect here are Philly Blunts.  Of course, if this  
+all becomes true, I'll be the first to smoke one myself.
+
+--
+David J.(dwarner@journalism.indiana.edu)*****Blue Riddle Productions 1993
+*-------------------------------It's on.--------------------------------*
+***"THE RAP IS AN ART EP" is coming out on tape -- this time for real.***
+*------------------------E-mail me for the 411.-------------------------*

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102606

@@ -0,0 +1,38 @@
+Newsgroups: rec.sport.baseball
+Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech!howland.reston.ans.net!newsserver.jvnc.net!siemens!princeton!crux!roger
+From: roger@crux.Princeton.EDU (Roger Lustig)
+Subject: Re: Mike Francesa's 1993 Predictions
+Message-ID: <1993Apr5.162835.1003@Princeton.EDU>
+Originator: news@nimaster
+Sender: news@Princeton.EDU (USENET News System)
+Nntp-Posting-Host: crux.princeton.edu
+Reply-To: roger@astro.princeton.edu (Roger Lustig)
+Organization: Princeton University
+References: <1993Apr2.133703.28131@porthos.cc.bellcore.com> <1993Apr2.171819.1@ulkyvx.louisville.edu> <1993Apr5.123904.17806@porthos.cc.bellcore.com>
+Date: Mon, 5 Apr 1993 16:28:35 GMT
+Lines: 24
+
+In article <1993Apr5.123904.17806@porthos.cc.bellcore.com> dick1@herahera.cc.bellcore.com (vaughn,richard) writes:
+>In article <1993Apr2.171819.1@ulkyvx.louisville.edu>, pjtier01@ulkyvx.louisville.edu writes:
+>> In article <1993Apr2.133703.28131@porthos.cc.bellcore.com>, dick1@herahera.cc.bellcore.com (vaughn,richard) writes:
+
+>>> Mike Francesa mentioned his '93 baseball picks in
+>>> passing on the radio yesterday.  Being that this is Francesa
+>>> talkin', the opinions are well worth consideration.
+
+>> Indeed!  After all, he was the wizard foresaw the Year of the Big East in the
+>> NCAA Tournament!  Unfortunatly, none of those teams made it into the Sweet
+>> Sixteen, much less New Orleans.
+
+Neither did he!
+
+>Overall Mike Francesa has an *outstanding* prediction record.
+
+Overall?  How do you figure?
+
+>Ignore him at your own peril.
+
+So far my radio hasn't exploded from not being tuned to 660...
+
+
+Roger

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102612

@@ -0,0 +1,45 @@
+Path: cantaloupe.srv.cs.cmu.edu!rochester!cornell!batcomputer!caen!saimiri.primate.wisc.edu!aplcen.apl.jhu.edu!jhunix.hcf.jhu.edu!jhunix.hcf.jhu.edu!not-for-mail
+From: admiral@jhunix.hcf.jhu.edu (Steve C Liu)
+Newsgroups: rec.sport.baseball
+Subject: Opening Day of 1990?
+Date: 5 Apr 1993 16:57:47 -0400
+Organization: Homewood Academic Computing, Johns Hopkins University, Baltimore, Md, USA
+Lines: 30
+Sender: admiral@jhunix.hcf.jhu.edu
+Distribution: usa
+Expires: 5/9/93
+Message-ID: <1pq6gbINNc7o@jhunix.hcf.jhu.edu>
+References: <khansen.15.734037238@staff.tc.umn.edu> <C5109u.7C0@ucdavis.edu> <C5125x.Jqt@boi.hp.com>
+NNTP-Posting-Host: jhunix.hcf.jhu.edu
+Summary: Baltimore O's not looking stellar
+
+The O's just lost to the Rangers a few minutes ago I was not too happy about
+the pitching of Rick Sutcliffe (6 runs in 6 innings, 5 in the 3?) This puts
+me in remembering the 1990 O's season. After '89 we didn't do much over
+winter and we wound up in 5th. Now I know that Mussina, McDonald, and Rhodes
+are better pitching prospects than Ballard and Milacki but are any other
+Oriole fans scared out there?
+
+Admiral Steve C. Liu
+
+P.S. Other scores as of now.
+Cincy over Montreal, 2-1 I think
+FLORIDA IS LEADING LA 6-3 IN THE 8TH!
+Braves vs. Cubs, 1-0 in the 8th. Futility of Cubs batting haunting them.
+Yanks beat the Tribe I believe.
+____________________________________________________________________________
+|Admiral Steve C. Liu          Internet Address: admiral@jhunix.hcf.jhu.edu|
+|Commander-In-Chief of the Security Division of the Pi Club - Earth Chapter|
+|    President of the Earth Chapter of the Pi Club - Founded April 1990    |
+|1993 World Champions  - Baltimore Orioles - Why Not? - Series in the Yards|
+|         1992-1993 Stanley Cup Champions -  Washington Capitals           |
+| "Committee for the Liberation and Intergration of Terrifying Organisms   |
+|   and their Rehabilitation Into Society, the only problem is that the    |
+|   abbreviation is CLITORIS." from the "Polymorph" episode of Red Dwarf   |
+|*****The Bangles are the greatest female rock band that ever existed!*****|
+|   This sig has been brought to you by... Frungy! The Sport of Kings!     |
+|"My God man, drilling holes through his head is not the answer!" Dr. McCoy|
+|"You know, Susanna Hoffs has a really nice ass." - comment by M. Flanagan |
+|  The Pi Club - Creating the largest .signatures for the past nine months | 
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102622

@@ -0,0 +1,41 @@
+Newsgroups: rec.sport.baseball
+Path: cantaloupe.srv.cs.cmu.edu!rochester!cornell!batcomputer!caen!zaphod.mps.ohio-state.edu!cs.utexas.edu!asuvax!ukma!bogus.sura.net!news-feed-1.peachnet.edu!umn.edu!myria!lsmith
+From: lsmith@myria.cs.umn.edu (Lance "Squiddie" Smith)
+Subject: Re: quick way to tell if your local beat writer is dumb.
+Message-ID: <C5188A.K34@news2.cis.umn.edu>
+Sender: news@news2.cis.umn.edu (Usenet News Administration)
+Nntp-Posting-Host: myria.cs.umn.edu
+Organization: The Little Projective Squids of the Poor
+References: <Apr.5.16.22.48.1993.11331@pilot.njin.net>
+Distribution: na
+Date: Mon, 5 Apr 1993 22:38:32 GMT
+Lines: 28
+
+In article <Apr.5.16.22.48.1993.11331@pilot.njin.net> gajarsky@pilot.njin.net (Bob Gajarsky - Hobokenite) writes:
+>
+>jayson stark (i think that's him) fits perfectly in this category.
+
+Could be. There is a Jayson Stark that writes weekly for some press syndicate 
+and also for Baseball America.
+
+>anyone who writes "dean palmer has 2 homers  - at this pace, he'll
+> have 324 home runs!" should be shot.
+>
+>if, at the end of april, he has 11, and anyone writes "at this
+> pace, he'll have 100+ homers!" they shouldbe shot too.
+
+Stark has done this sort of thing, but he has never been serious about it.
+He usually states that this sort of projection is useless at the top of 
+such columns. I think he did it one season and some manager was "projected" 
+to be thrown out of 60 games and some hitter could expect to be plunked 
+150 times.
+
+Stark does some really funny stuff. His weekly baseball reviews are good 
+collections of strange things that happened during the previous week. He 
+also regularly prints Kinerisms.
+
+=============================
+  Lance "Cr2O3.2H2O" Smith  | "Moments after being named manager for the
+    (lsmith@cs.umn.edu)     |  Oakland A's, Mr Peanut was crushed by a 
+  Special Limited Edition   |  red headed loner wielding an aluminum bat." 
+     r.s.bb  .signature     |                _Murder at the Mausoleum_

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102625

@@ -0,0 +1,25 @@
+Path: cantaloupe.srv.cs.cmu.edu!rochester!cornell!batcomputer!caen!nigel.msen.com!sdd.hp.com!zaphod.mps.ohio-state.edu!malgudi.oar.net!ucbeh.san.uc.edu!ucunix.san.uc.edu!nest.ece.uc.edu!babbage.ece.uc.edu!flash.ece.uc.edu!rbd
+Newsgroups: rec.sport.baseball
+Subject: Re: quick way to tell if your local beat writer is dumb.
+Message-ID: <C517Fu.BFH@babbage.ece.uc.edu>
+From: rbd@flash.ece.uc.edu (Bobby Davis)
+Date: Mon, 5 Apr 1993 22:21:29 GMT
+Sender: root@babbage.ece.uc.edu (Operator)
+References: <Apr.5.16.22.48.1993.11331@pilot.njin.net>
+Organization: University of Cincinnati
+Nntp-Posting-Host: flash.ece.uc.edu
+Lines: 13
+
+Bob Gajarsky - Hobokenite writes:
+>jayson stark (i trhink that's him) fits perfectly in this category.
+>
+>anyone who writes "dean palmer has 2 homers  - at this pace, he'll
+> have 324 home runs!" should be shot.
+
+Bob, I think that Stark does this sort of thing as a joke, not as a
+serious prediction.  I don't really see why we should shoot him for
+that.  The guys who ought to be shot are the ones who keep claiming
+how great the Royals' chances are in the AL West, since all evidence
+indicates that they aren't joking and actually believe it.
+
+Bob Davis	rbd@thor.ece.uc.edu

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102631

@@ -0,0 +1,50 @@
+Newsgroups: rec.sport.baseball
+Path: cantaloupe.srv.cs.cmu.edu!rochester!cornell!batcomputer!caen!zaphod.mps.ohio-state.edu!sdd.hp.com!decwrl!netcomsv!orchard.la.locus.com!prodnet.la.locus.com!cygnus.la.locus.com!aardvark
+From: aardvark@cygnus.la.locus.com (Warren Usui)
+Subject: rec.sport.baseball.fantasy
+Message-ID: <1993Apr05.224438.101525@locus.com>
+Sender: news@locus.com (Netnews)
+Organization: Locus Computing Corporation, Los Angeles, California
+Date: Mon, 05 Apr 93 22:44:38 GMT
+Lines: 40
+
+
+
+							April 1, 1993
+
+
+I am participating in an NL-league that uses standard Rotisserie rules
+except that the following catagories are used:
+
+For position players:
+	lowest batting average
+	strike-outs
+	caught-stealing
+	errors
+
+For pitchers:
+	losses
+	blown saves
+	higest ERA
+	'taters allowed
+
+This is the fifth year that I've participated in this Blowtisserie
+league.  Last year I won the pennant due primarily to the fact that
+I had terrible pitching.  I would like to lower my batting average
+which is rather high because I do have Jose Offerman (who made up
+for this by helping me lock first place in errors).  Anyway,
+someone offered:
+
+Andres Gallaraga for Bud Black
+
+I can afford to give up Bud Black because I still have Kyle Abbott.
+However, I am afraid of Andres actually doing well this season.
+Should I make the trade or not?
+
+Your comments will be appreciated.
+
+-- 
+Warren Usui
+
+I'm one with the Universe -- on a scale from 1 to 10.
+

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102641

@@ -0,0 +1,24 @@
+Newsgroups: rec.sport.baseball
+Path: cantaloupe.srv.cs.cmu.edu!rochester!news.crd.ge.com!uunet!europa.eng.gtefsd.com!howland.reston.ans.net!usc!sdd.hp.com!saimiri.primate.wisc.edu!caen!batcomputer!cornell!uw-beaver!cs.ubc.ca!newsserver.sfu.ca!sfu.ca!rauser
+From: rauser@fraser.sfu.ca (Richard John Rauser)
+Subject: Quick easy question!
+Message-ID: <rauser.734062227@sfu.ca>
+Sender: news@sfu.ca
+Organization: Simon Fraser University, Burnaby, B.C., Canada
+Date: Tue, 6 Apr 1993 02:10:27 GMT
+Lines: 14
+
+
+
+   Here's an easy question for someone who knows nothing about baseball...
+
+   What city do the California Angels play out of?
+
+
+
+-- 
+Richard J. Rauser        "You have no idea what you're doing."
+rauser@sfu.ca            "Oh, don't worry about that. We're professional
+WNI                          outlaws - we do this for a living."
+-----------------
+"Remember, no matter where you go, there you are." -Dr.Banzai

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102649

@@ -0,0 +1,66 @@
+Path: cantaloupe.srv.cs.cmu.edu!rochester!news.crd.ge.com!uunet!gatech!ukma!netnews.louisville.edu!wkuvx1!cabanrf
+From: cabanrf@wkuvx1.bitnet
+Newsgroups: rec.sport.baseball
+Subject: Re: My Belated Predictions (NL)
+Message-ID: <1993Apr5.235446.6723@wkuvx1.bitnet>
+Date: 5 Apr 93 23:54:46 CDT
+References: <12645@news.duke.edu> <1993Apr3.213116.10442@mnemosyne.cs.du.edu> <Apr.4.19.42.08.1993.12176@pilot.njin.net> <mssC4zyo8.JsC@netcom.com>
+Organization: Western Kentucky University, Bowling Green, KY
+Lines: 56
+
+In article <mssC4zyo8.JsC@netcom.com>, mss@netcom.com (Mark Singer) writes:
+> In article <Apr.4.19.42.08.1993.12176@pilot.njin.net> gajarsky@pilot.njin.net (Bob Gajarsky - Hobokenite) writes:
+>>i've said the braves would improve by injury as well.  here's how.
+>>
+>>javier lopez is a better catcher than greg olson.
+>>ryan klasko is a better firstbaseman than bream.
+>> chipper jones is a better shortstop than anyone the braves
+>> put out there.
+>>
+>>mel nieves is better than nixon/sanders.
+>>
+>>that's how. it FORCES them to play the young guys.
+>>
+>>- bob gaj
+> 
+> I continue to be amazed at these comments.  While Lopez might *some
+> day* be a better catcher than Olson, I find it totally amazing for
+> you to suggest that this 22 year-old with three seasons of professional
+> baseball is *now* better than Olson, a five-year MLB veteran who is
+> noted for his ability to call a game, and who has a better-than-average
+> arm.  Oh, perhaps you are talking about hitting.  Well, sure, Lopez
+> *might* hit better.  Perhaps he *probably* will.
+> 
+> But has there ever in the history of baseball been a 22-year-old (or
+> younger) *rookie* catcher who compared favorably among all league
+> catchers in terms of defense and brought a .247 bat?  Wasn't it 
+
+Yes, Ivan Rodriguez, last year.  Batted .260 and threw out 51% of the
+baserunners.  Not too shabby for a rookie from AA.  20 years old last
+year.
+
+> Sandy Alomar who was supposed to be that good in his rookie year?
+> Not.  Wasn't it Benito Santiago who was supposed to be that good
+> in his rookie year?  Not.
+> 
+> I can continue this thread with the others mentioned, but you get
+> the point.  You and others seem to be so quick to dismiss the 
+> seasoned veterans in favor of the hot *young* rookies.  Perhaps -
+> just perhaps - the management team of the pennant-winning Braves
+> knows something more than you do.  And perhaps what they know is
+> that very, very few 21- and 22-year old rookies come up to the majors
+> and make an impact. 
+> 
+> 
+> --	The Beastmaster
+> 
+> 
+>  
+> -- 
+> Mark Singer    
+> mss@netcom.com
+-- 
+Roy F. Cabaniss......................*Wait till Tommy meets the Lord and
+Western Kentucky University..........*finds out that He's wearing pinstripes.
+All opinions contained herein........*Gaylord Perry (talking about Lasorda)
+Are all mine own, and that's the sin.*Baseball, what a way to spend a day!!

gensim/test/test_data/mini_newsgroups/rec.sport.baseball/102655

@@ -0,0 +1,30 @@
+Newsgroups: rec.sport.baseball
+Path: cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!fs7.ece.cmu.edu!europa.eng.gtefsd.com!howland.reston.ans.net!wupost!uwm.edu!linac!att!princeton!phoenix.Princeton.EDU!niepornt
+From: niepornt@phoenix.Princeton.EDU (David Marc Nieporent)
+Subject: Re: DESIGNATED HITTER RULE
+Message-ID: <1993Apr6.081636.1503@Princeton.EDU>
+Originator: news@nimaster
+Sender: news@Princeton.EDU (USENET News System)
+Nntp-Posting-Host: phoenix.princeton.edu
+Organization: Princeton University
+References: <1993Apr4.221228.17577@bsu-ucs> <ekdfc.14.0@ttacs1.ttu.edu>
+Date: Tue, 6 Apr 1993 08:16:36 GMT
+Lines: 17
+
+In <ekdfc.14.0@ttacs1.ttu.edu> ekdfc@ttacs1.ttu.edu (David Coons) writes:
+>In <1993Apr4.221228.17577@bsu-ucs> 00ecgillespi@leo.bsuvc.bsu.edu writes:
+
+>>I AM DOING A POSTITION PAPER ON THE DESIGNATED HITTER RULE. ANY INFORMATION
+>>OR EVEN OPINIONS WOULD BE GREATLY APPRECITATED. 00ECGILLESPIE "MAGIC"
+
+>Should be rescinded.  The rules say baseball is a game between two teams of 
+>nine players each.  Let's keep it that way.
+
+Not any more the rules don't say that.  So that's a pretty dumb argument.
+
+
+-- 
+David M. Nieporent   |  "Only one thing wrong with theory...
+niepornt@phoenix.    |    Is stupid!  Is stupidest theory I ever heard!"
+   princeton.edu     |   --------------------- 
+Baltimore Orioles 93 |  Who's the dangerous cult -- the BDs or the BATF?