From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (from majordomo@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id XAA02253; Mon, 13 May 2002 23:46:21 +0200 (MET DST) X-Authentication-Warning: pauillac.inria.fr: majordomo set sender to owner-caml-list@pauillac.inria.fr using -f Received: from nez-perce.inria.fr (nez-perce.inria.fr [192.93.2.78]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id XAA02248 for ; Mon, 13 May 2002 23:46:20 +0200 (MET DST) X-SPAM-Warning: Sending machine is listed in blackholes.five-ten-sg.com Received: from gogol.zorgol (Mix-Montsouris-109-4-70.abo.wanadoo.fr [80.9.126.70]) by nez-perce.inria.fr (8.11.1/8.11.1) with SMTP id g4DLkFr12359 for ; Mon, 13 May 2002 23:46:15 +0200 (MET DST) Received: (qmail 13928 invoked by uid 1001); 13 May 2002 21:47:09 -0000 Date: Mon, 13 May 2002 23:47:09 +0200 From: Berke Durak To: Markus Mottl , caml-list@inria.fr Subject: Re: [Caml-list] Turning off type-checking Message-ID: <20020513234709.A3920@gogol.zorgol> References: <20020513133102.GB9777@kiefer.ai.univie.ac.at> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20020513133102.GB9777@kiefer.ai.univie.ac.at>; from markus@oefai.at on Mon, May 13, 2002 at 03:31:02PM +0200 Sender: owner-caml-list@pauillac.inria.fr Precedence: bulk On Mon, May 13, 2002 at 03:31:02PM +0200, Markus Mottl wrote: > Hello, Hi >... > > Needless to say that I get into trouble when I actually want to interpret > or compile these models to evaluate them on new data. Type checking just > takes an awful amount of time, in fact much longer than the learning > algorithm needs for model generation... I'm conservatively assuming that your code uses just classical types (guessing that type-checking objects etc. can be more costly). Could the type experts on this list confirm or deny that : - For ``classical'' types, type-checking amounts to term unification. - Term unification can be done in worst-case linear-time. - Ocaml's unification algorithm is worst-case suboptimal. - This explains the ``awful amount of time'' needed for type checking. If these are true, a fifth question arises : - Is it worth to make Ocaml's type checking worst-case optimal ? How will this affect real average-case performance ? -- Berke Durak ------------------- To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/ Beginner's list: http://groups.yahoo.com/group/ocaml_beginners