[
https://issues.apache.org/jira/browse/MAHOUT-184?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sean Owen updated MAHOUT-184:
-
Attachment: Tweaks_to__df__.patch
> Code tweaks for .df.* code
> --
>
>
Code tweaks for .df.* code
--
Key: MAHOUT-184
URL: https://issues.apache.org/jira/browse/MAHOUT-184
Project: Mahout
Issue Type: Improvement
Reporter: Sean Owen
Assignee: Sean Owen
P
Yes. It is just expectations.
x += 1= or x++ or x-- are all common. Other operations are uncommon and %
in particular has the deviant property of other meanings in other languages
(it is often a comment character).
On Thu, Oct 1, 2009 at 12:21 PM, Sean Owen wrote:
> Is it
> merely a question
It's so small that I can't feel strongly about it either way. Is it
merely a question of what one expects to read? I am more accustomed to
the latter form. The language allows it and I'm used to seeing it in
various forms like "i++". It can only be faster, at runtime, since it
might avoid an additi
Colt did a nice job of this. Basically, their idea was to take various
general functional patterns and allow the functions to be plugged in.
Common patterns that are reasonable to include in such a framework include:
a) dot product as the aggregration of a pairwise function application
(normal d
On Thu, Oct 1, 2009 at 10:10 AM, Ted Dunning wrote:
> Btw... the other think that the HashVector does better is inserts. The
> sorted vector could do much better on average if it deferred sorting until
> an access or iteration was done. Even iteration doesn't necessarily need
> sorting, but it
Btw... the other think that the HashVector does better is inserts. The
sorted vector could do much better on average if it deferred sorting until
an access or iteration was done. Even iteration doesn't necessarily need
sorting, but it could by seen as part of the contract.
On Thu, Oct 1, 2009 at
Yeah, I added the "trying to find..." part of the debug output because I
couldn't
figure out what IntDoubleHash was "impossibly confused" about.
Unfortunately, seeing what it was confused about only confused *me* about
why it was impossible.
On Thu, Oct 1, 2009 at 9:12 AM, Ted Dunning wrote:
>
It indicates a bug in the code or in the writer's head.
You are correct about the intent. The default value (which should probably
just be 0) should be returned if the value is missing.
On Thu, Oct 1, 2009 at 5:41 AM, Grant Ingersoll (JIRA) wrote:
> I don't think this is "impossible confusion",
On Thu, Oct 1, 2009 at 2:29 AM, Sean Owen wrote:
> This doesn't make sense when the args are ints:
> Math.floor(numTrees / numMaps)
> The result is already an int, rounded 'down'
>
Since other languages have incoherent rounding rules for this division, this
idiom can actually help readability.
The MTJ committers were willing to relicense under Apache and donate the
entire package.
On Wed, Sep 30, 2009 at 9:17 PM, Jake Mannix wrote:
> On Wed, Sep 30, 2009 at 8:26 PM, Ted Dunning
> wrote:
>
> > No motion. I was pushing that integration because it looked like MTJ was
> > integrating wi
[
https://issues.apache.org/jira/browse/MAHOUT-182?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Grant Ingersoll reassigned MAHOUT-182:
--
Assignee: Grant Ingersoll
> New helper methods for Matrix: times(Vector), timesSquared(
[
https://issues.apache.org/jira/browse/MAHOUT-165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Grant Ingersoll updated MAHOUT-165:
---
Attachment: mahout-165.patch
This gets the VectorTest testEquals to pass. Also fixes an inst
[
https://issues.apache.org/jira/browse/MAHOUT-165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12761201#action_12761201
]
Grant Ingersoll commented on MAHOUT-165:
The exception in the test is:
{quote}
java
[
https://issues.apache.org/jira/browse/MAHOUT-183?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Olivier Grisel updated MAHOUT-183:
--
Description:
The Wikipedia XML splitter inner loop erronously detects end of the
line-iterator
[
https://issues.apache.org/jira/browse/MAHOUT-183?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Olivier Grisel updated MAHOUT-183:
--
Status: Patch Available (was: Open)
> WikipediaXmlSplitter spits one chunk per line
>
[
https://issues.apache.org/jira/browse/MAHOUT-183?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Olivier Grisel updated MAHOUT-183:
--
Attachment: MAHOUT-183-wikipedia-xml-splitter.patch
> WikipediaXmlSplitter spits one chunk per
WikipediaXmlSplitter spits one chunk per line
-
Key: MAHOUT-183
URL: https://issues.apache.org/jira/browse/MAHOUT-183
Project: Mahout
Issue Type: Bug
Components: Classification
Affect
(PS yeah that was my fault for misreading the original message.)
On Thu, Oct 1, 2009 at 11:39 AM, Grant Ingersoll wrote:
>
> On Sep 30, 2009, at 4:34 PM, Jake Mannix wrote:
>
>> I didn't say that equals() should ignore name, I said the opposite -
>> equals
>> and
>> hashCode() should *only* take
On Oct 1, 2009, at 12:17 AM, Jake Mannix wrote:
On Wed, Sep 30, 2009 at 8:26 PM, Ted Dunning
wrote:
So why do we really need vectors to be Writable? I see the appeal,
it's
nice and makes the code nicely integrated, but the way I ended up
going,
so that you could use decomposer either
On Sep 30, 2009, at 4:34 PM, Jake Mannix wrote:
I didn't say that equals() should ignore name, I said the opposite -
equals
and
hashCode() should *only* take into account the contents and the
name, and
not
implementation (which means that hashCode() needs to stay in one
place and
not ge
May I suggest we don't static import classes unless it makes a real
difference in readability, but certainly, may I suggest we don't
static-import a single static method or field from a class? finding a
reference to it in the code becomes confusing because one expects the
method or field exists in
[
https://issues.apache.org/jira/browse/MAHOUT-157?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robin Anil updated MAHOUT-157:
--
Attachment: MAHOUT-157-Oct-1.patch
Finished Sequential version of FPGrowth. May need some more document
23 matches
Mail list logo
