You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spark.apache.org by Hector Yee <he...@gmail.com> on 2015/04/01 20:35:40 UTC
Re: Storing large data for MLlib machine learning
I use Thrift and then base64 encode the binary and save it as text file
lines that are snappy or gzip encoded.
It makes it very easy to copy small chunks locally and play with subsets of
the data and not have dependencies on HDFS / hadoop for server stuff for
example.
On Thu, Mar 26, 2015 at 2:51 PM, Ulanov, Alexander <al...@hp.com>
wrote:
> Thanks, Evan. What do you think about Protobuf? Twitter has a library to
> manage protobuf files in hdfs https://github.com/twitter/elephant-bird
>
>
> From: Evan R. Sparks [mailto:evan.sparks@gmail.com]
> Sent: Thursday, March 26, 2015 2:34 PM
> To: Stephen Boesch
> Cc: Ulanov, Alexander; dev@spark.apache.org
> Subject: Re: Storing large data for MLlib machine learning
>
> On binary file formats - I looked at HDF5+Spark a couple of years ago and
> found it barely JVM-friendly and very Hadoop-unfriendly (e.g. the APIs
> needed filenames as input, you couldn't pass it anything like an
> InputStream). I don't know if it has gotten any better.
>
> Parquet plays much more nicely and there are lots of spark-related
> projects using it already. Keep in mind that it's column-oriented which
> might impact performance - but basically you're going to want your features
> in a byte array and deser should be pretty straightforward.
>
> On Thu, Mar 26, 2015 at 2:26 PM, Stephen Boesch <javadba@gmail.com<mailto:
> javadba@gmail.com>> wrote:
> There are some convenience methods you might consider including:
>
> MLUtils.loadLibSVMFile
>
> and MLUtils.loadLabeledPoint
>
> 2015-03-26 14:16 GMT-07:00 Ulanov, Alexander <alexander.ulanov@hp.com
> <ma...@hp.com>>:
>
> > Hi,
> >
> > Could you suggest what would be the reasonable file format to store
> > feature vector data for machine learning in Spark MLlib? Are there any
> best
> > practices for Spark?
> >
> > My data is dense feature vectors with labels. Some of the requirements
> are
> > that the format should be easy loaded/serialized, randomly accessible,
> with
> > a small footprint (binary). I am considering Parquet, hdf5, protocol
> buffer
> > (protobuf), but I have little to no experience with them, so any
> > suggestions would be really appreciated.
> >
> > Best regards, Alexander
> >
>
>
--
Yee Yang Li Hector <http://google.com/+HectorYee>
*google.com/+HectorYee <http://google.com/+HectorYee>*
RE: Storing large data for MLlib machine learning
Posted by "Ulanov, Alexander" <al...@hp.com>.
Jeremy, thanks for explanation!
What if instead you've used Parquet file format? You can still write a number of small files as you do, but you don't have to implement a writer/reader, because they are available for Parquet in various languages.
From: Jeremy Freeman [mailto:freeman.jeremy@gmail.com]
Sent: Wednesday, April 01, 2015 1:37 PM
To: Hector Yee
Cc: Ulanov, Alexander; Evan R. Sparks; Stephen Boesch; dev@spark.apache.org
Subject: Re: Storing large data for MLlib machine learning
@Alexander, re: using flat binary and metadata, you raise excellent points! At least in our case, we decided on a specific endianness, but do end up storing some extremely minimal specification in a JSON file, and have written importers and exporters within our library to parse it. While it does feel a little like reinvention, it's fast, direct, and scalable, and seems pretty sensible if you know your data will be dense arrays of numerical features.
-------------------------
jeremyfreeman.net<http://jeremyfreeman.net>
@thefreemanlab
On Apr 1, 2015, at 3:52 PM, Hector Yee <he...@gmail.com>> wrote:
Just using sc.textfile then a .map(decode)
Yes by default it is multiple files .. our training data is 1TB gzipped
into 5000 shards.
On Wed, Apr 1, 2015 at 12:32 PM, Ulanov, Alexander <al...@hp.com>>
wrote:
Thanks, sounds interesting! How do you load files to Spark? Did you
consider having multiple files instead of file lines?
*From:* Hector Yee [mailto:hector.yee@gmail.com]
*Sent:* Wednesday, April 01, 2015 11:36 AM
*To:* Ulanov, Alexander
*Cc:* Evan R. Sparks; Stephen Boesch; dev@spark.apache.org<ma...@spark.apache.org>
*Subject:* Re: Storing large data for MLlib machine learning
I use Thrift and then base64 encode the binary and save it as text file
lines that are snappy or gzip encoded.
It makes it very easy to copy small chunks locally and play with subsets
of the data and not have dependencies on HDFS / hadoop for server stuff for
example.
On Thu, Mar 26, 2015 at 2:51 PM, Ulanov, Alexander <
alexander.ulanov@hp.com<ma...@hp.com>> wrote:
Thanks, Evan. What do you think about Protobuf? Twitter has a library to
manage protobuf files in hdfshttps://github.com/twitter/elephant-bird
From: Evan R. Sparks [mailto:evan.sparks@gmail.com]
Sent: Thursday, March 26, 2015 2:34 PM
To: Stephen Boesch
Cc: Ulanov, Alexander; dev@spark.apache.org<ma...@spark.apache.org>
Subject: Re: Storing large data for MLlib machine learning
On binary file formats - I looked at HDF5+Spark a couple of years ago and
found it barely JVM-friendly and very Hadoop-unfriendly (e.g. the APIs
needed filenames as input, you couldn't pass it anything like an
InputStream). I don't know if it has gotten any better.
Parquet plays much more nicely and there are lots of spark-related
projects using it already. Keep in mind that it's column-oriented which
might impact performance - but basically you're going to want your features
in a byte array and deser should be pretty straightforward.
On Thu, Mar 26, 2015 at 2:26 PM, Stephen Boesch <ja...@gmail.com><mailto:
javadba@gmail.com<ma...@gmail.com>>> wrote:
There are some convenience methods you might consider including:
MLUtils.loadLibSVMFile
and MLUtils.loadLabeledPoint
2015-03-26 14:16 GMT-07:00 Ulanov, Alexander <al...@hp.com>
<ma...@hp.com>>:
Hi,
Could you suggest what would be the reasonable file format to store
feature vector data for machine learning in Spark MLlib? Are there any
best
practices for Spark?
My data is dense feature vectors with labels. Some of the requirements
are
that the format should be easy loaded/serialized, randomly accessible,
with
a small footprint (binary). I am considering Parquet, hdf5, protocol
buffer
(protobuf), but I have little to no experience with them, so any
suggestions would be really appreciated.
Best regards, Alexander
--
Yee Yang Li Hector <http://google.com/+HectorYee>
*google.com/+HectorYee<http://google.com/+HectorYee> <http://google.com/+HectorYee>*
--
Yee Yang Li Hector <http://google.com/+HectorYee>
*google.com/+HectorYee<http://google.com/+HectorYee> <http://google.com/+HectorYee>*
Re: Storing large data for MLlib machine learning
Posted by Jeremy Freeman <fr...@gmail.com>.
@Alexander, re: using flat binary and metadata, you raise excellent points! At least in our case, we decided on a specific endianness, but do end up storing some extremely minimal specification in a JSON file, and have written importers and exporters within our library to parse it. While it does feel a little like reinvention, it’s fast, direct, and scalable, and seems pretty sensible if you know your data will be dense arrays of numerical features.
-------------------------
jeremyfreeman.net
@thefreemanlab
On Apr 1, 2015, at 3:52 PM, Hector Yee <he...@gmail.com> wrote:
> Just using sc.textfile then a .map(decode)
> Yes by default it is multiple files .. our training data is 1TB gzipped
> into 5000 shards.
>
> On Wed, Apr 1, 2015 at 12:32 PM, Ulanov, Alexander <al...@hp.com>
> wrote:
>
>> Thanks, sounds interesting! How do you load files to Spark? Did you
>> consider having multiple files instead of file lines?
>>
>>
>>
>> *From:* Hector Yee [mailto:hector.yee@gmail.com]
>> *Sent:* Wednesday, April 01, 2015 11:36 AM
>> *To:* Ulanov, Alexander
>> *Cc:* Evan R. Sparks; Stephen Boesch; dev@spark.apache.org
>>
>> *Subject:* Re: Storing large data for MLlib machine learning
>>
>>
>>
>> I use Thrift and then base64 encode the binary and save it as text file
>> lines that are snappy or gzip encoded.
>>
>>
>>
>> It makes it very easy to copy small chunks locally and play with subsets
>> of the data and not have dependencies on HDFS / hadoop for server stuff for
>> example.
>>
>>
>>
>>
>>
>> On Thu, Mar 26, 2015 at 2:51 PM, Ulanov, Alexander <
>> alexander.ulanov@hp.com> wrote:
>>
>> Thanks, Evan. What do you think about Protobuf? Twitter has a library to
>> manage protobuf files in hdfshttps://github.com/twitter/elephant-bird
>>
>>
>> From: Evan R. Sparks [mailto:evan.sparks@gmail.com]
>> Sent: Thursday, March 26, 2015 2:34 PM
>> To: Stephen Boesch
>> Cc: Ulanov, Alexander; dev@spark.apache.org
>> Subject: Re: Storing large data for MLlib machine learning
>>
>> On binary file formats - I looked at HDF5+Spark a couple of years ago and
>> found it barely JVM-friendly and very Hadoop-unfriendly (e.g. the APIs
>> needed filenames as input, you couldn't pass it anything like an
>> InputStream). I don't know if it has gotten any better.
>>
>> Parquet plays much more nicely and there are lots of spark-related
>> projects using it already. Keep in mind that it's column-oriented which
>> might impact performance - but basically you're going to want your features
>> in a byte array and deser should be pretty straightforward.
>>
>> On Thu, Mar 26, 2015 at 2:26 PM, Stephen Boesch <javadba@gmail.com<mailto:
>> javadba@gmail.com>> wrote:
>> There are some convenience methods you might consider including:
>>
>> MLUtils.loadLibSVMFile
>>
>> and MLUtils.loadLabeledPoint
>>
>> 2015-03-26 14:16 GMT-07:00 Ulanov, Alexander <alexander.ulanov@hp.com
>> <ma...@hp.com>>:
>>
>>
>>> Hi,
>>>
>>> Could you suggest what would be the reasonable file format to store
>>> feature vector data for machine learning in Spark MLlib? Are there any
>> best
>>> practices for Spark?
>>>
>>> My data is dense feature vectors with labels. Some of the requirements
>> are
>>> that the format should be easy loaded/serialized, randomly accessible,
>> with
>>> a small footprint (binary). I am considering Parquet, hdf5, protocol
>> buffer
>>> (protobuf), but I have little to no experience with them, so any
>>> suggestions would be really appreciated.
>>>
>>> Best regards, Alexander
>>>
>>
>>
>>
>>
>>
>> --
>>
>> Yee Yang Li Hector <http://google.com/+HectorYee>
>>
>> *google.com/+HectorYee <http://google.com/+HectorYee>*
>>
>
>
>
> --
> Yee Yang Li Hector <http://google.com/+HectorYee>
> *google.com/+HectorYee <http://google.com/+HectorYee>*
Re: Storing large data for MLlib machine learning
Posted by Hector Yee <he...@gmail.com>.
Just using sc.textfile then a .map(decode)
Yes by default it is multiple files .. our training data is 1TB gzipped
into 5000 shards.
On Wed, Apr 1, 2015 at 12:32 PM, Ulanov, Alexander <al...@hp.com>
wrote:
> Thanks, sounds interesting! How do you load files to Spark? Did you
> consider having multiple files instead of file lines?
>
>
>
> *From:* Hector Yee [mailto:hector.yee@gmail.com]
> *Sent:* Wednesday, April 01, 2015 11:36 AM
> *To:* Ulanov, Alexander
> *Cc:* Evan R. Sparks; Stephen Boesch; dev@spark.apache.org
>
> *Subject:* Re: Storing large data for MLlib machine learning
>
>
>
> I use Thrift and then base64 encode the binary and save it as text file
> lines that are snappy or gzip encoded.
>
>
>
> It makes it very easy to copy small chunks locally and play with subsets
> of the data and not have dependencies on HDFS / hadoop for server stuff for
> example.
>
>
>
>
>
> On Thu, Mar 26, 2015 at 2:51 PM, Ulanov, Alexander <
> alexander.ulanov@hp.com> wrote:
>
> Thanks, Evan. What do you think about Protobuf? Twitter has a library to
> manage protobuf files in hdfs https://github.com/twitter/elephant-bird
>
>
> From: Evan R. Sparks [mailto:evan.sparks@gmail.com]
> Sent: Thursday, March 26, 2015 2:34 PM
> To: Stephen Boesch
> Cc: Ulanov, Alexander; dev@spark.apache.org
> Subject: Re: Storing large data for MLlib machine learning
>
> On binary file formats - I looked at HDF5+Spark a couple of years ago and
> found it barely JVM-friendly and very Hadoop-unfriendly (e.g. the APIs
> needed filenames as input, you couldn't pass it anything like an
> InputStream). I don't know if it has gotten any better.
>
> Parquet plays much more nicely and there are lots of spark-related
> projects using it already. Keep in mind that it's column-oriented which
> might impact performance - but basically you're going to want your features
> in a byte array and deser should be pretty straightforward.
>
> On Thu, Mar 26, 2015 at 2:26 PM, Stephen Boesch <javadba@gmail.com<mailto:
> javadba@gmail.com>> wrote:
> There are some convenience methods you might consider including:
>
> MLUtils.loadLibSVMFile
>
> and MLUtils.loadLabeledPoint
>
> 2015-03-26 14:16 GMT-07:00 Ulanov, Alexander <alexander.ulanov@hp.com
> <ma...@hp.com>>:
>
>
> > Hi,
> >
> > Could you suggest what would be the reasonable file format to store
> > feature vector data for machine learning in Spark MLlib? Are there any
> best
> > practices for Spark?
> >
> > My data is dense feature vectors with labels. Some of the requirements
> are
> > that the format should be easy loaded/serialized, randomly accessible,
> with
> > a small footprint (binary). I am considering Parquet, hdf5, protocol
> buffer
> > (protobuf), but I have little to no experience with them, so any
> > suggestions would be really appreciated.
> >
> > Best regards, Alexander
> >
>
>
>
>
>
> --
>
> Yee Yang Li Hector <http://google.com/+HectorYee>
>
> *google.com/+HectorYee <http://google.com/+HectorYee>*
>
--
Yee Yang Li Hector <http://google.com/+HectorYee>
*google.com/+HectorYee <http://google.com/+HectorYee>*
RE: Storing large data for MLlib machine learning
Posted by "Ulanov, Alexander" <al...@hp.com>.
Thanks, sounds interesting! How do you load files to Spark? Did you consider having multiple files instead of file lines?
From: Hector Yee [mailto:hector.yee@gmail.com]
Sent: Wednesday, April 01, 2015 11:36 AM
To: Ulanov, Alexander
Cc: Evan R. Sparks; Stephen Boesch; dev@spark.apache.org
Subject: Re: Storing large data for MLlib machine learning
I use Thrift and then base64 encode the binary and save it as text file lines that are snappy or gzip encoded.
It makes it very easy to copy small chunks locally and play with subsets of the data and not have dependencies on HDFS / hadoop for server stuff for example.
On Thu, Mar 26, 2015 at 2:51 PM, Ulanov, Alexander <al...@hp.com>> wrote:
Thanks, Evan. What do you think about Protobuf? Twitter has a library to manage protobuf files in hdfs https://github.com/twitter/elephant-bird
From: Evan R. Sparks [mailto:evan.sparks@gmail.com<ma...@gmail.com>]
Sent: Thursday, March 26, 2015 2:34 PM
To: Stephen Boesch
Cc: Ulanov, Alexander; dev@spark.apache.org<ma...@spark.apache.org>
Subject: Re: Storing large data for MLlib machine learning
On binary file formats - I looked at HDF5+Spark a couple of years ago and found it barely JVM-friendly and very Hadoop-unfriendly (e.g. the APIs needed filenames as input, you couldn't pass it anything like an InputStream). I don't know if it has gotten any better.
Parquet plays much more nicely and there are lots of spark-related projects using it already. Keep in mind that it's column-oriented which might impact performance - but basically you're going to want your features in a byte array and deser should be pretty straightforward.
On Thu, Mar 26, 2015 at 2:26 PM, Stephen Boesch <ja...@gmail.com>>> wrote:
There are some convenience methods you might consider including:
MLUtils.loadLibSVMFile
and MLUtils.loadLabeledPoint
2015-03-26 14:16 GMT-07:00 Ulanov, Alexander <al...@hp.com>>>:
> Hi,
>
> Could you suggest what would be the reasonable file format to store
> feature vector data for machine learning in Spark MLlib? Are there any best
> practices for Spark?
>
> My data is dense feature vectors with labels. Some of the requirements are
> that the format should be easy loaded/serialized, randomly accessible, with
> a small footprint (binary). I am considering Parquet, hdf5, protocol buffer
> (protobuf), but I have little to no experience with them, so any
> suggestions would be really appreciated.
>
> Best regards, Alexander
>
--
Yee Yang Li Hector<http://google.com/+HectorYee>
google.com/+HectorYee<http://google.com/+HectorYee>