mendeley/mrec

A recommender systems development and evaluation package by Mendeley

mrec's Issues

install error

when i install mrec with git, there is an error i cant deal with:
ImportError: cannot import name 'izip' from 'itertools' (unknown location)

Citations format in Bibliography - removing et al.

I need to edit my own citation style so that it does not put et al. in the bibliography but cite all author names. Using the CSL editor, it seems simple to change the et al usage in-text but not in the bibliography, where my citations with over 5 Authors get et al. after the first name.

About scripts

Hello,

I'm using mrec now, however, I can't find where mrec_prepare, mrec_train and mrec_predict scripts are. May I ask you where they are? And how do I use them?

thx!

Sincerely yours,
David

Reference style for Journal of Industrial and Biotechnology

Hi,

Could anyone please share CSL file for the Journal of Industrial and Biotechnology.

https://www.springer.com/journal/10295/submission-guidelines#Instructions%20for%20Authors_References

I need to use Mendeley to insert the references according to this Journal.

Many thanks
nabiyogesh

`mrec_train` error: `AttributeError: 'list' object has no attribute 'get'`

When I executed mrec_train -n4 --input_format tsv --train "splits/u.data.train.*" --outdir models in here, the error appeared.

$ mrec_train -n4 --input_format tsv --train "splits/u.data.train.*" --outdir models
/Users/nzw/.pyenv/versions/miniconda2-latest/lib/python2.7/site-packages/ipython-4.0.0-py2.7.egg/IPython/parallel.py:13: ShimWarning: The `IPython.parallel` package has been deprecated. You should import from ipyparallel instead.
  "You should import from ipyparallel instead.", ShimWarning)
Traceback (most recent call last):
  File "/Users/nzw/.pyenv/versions/miniconda2-latest/bin/mrec_train", line 11, in <module>
    load_entry_point('mrec==0.3.1', 'console_scripts', 'mrec_train')()
  File "build/bdist.macosx-10.7-x86_64/egg/mrec/examples/train.py", line 22, in main
ImportError: cannot import name Client

To run correctly, I installed ipyparallel.

But, there is the other error:

$ mrec_train -n4 --input_format tsv --train "splits/u.data.train.*" --outdir models
...
File "/Users/nzw/.pyenv/versions/miniconda2-latest/bin/mrec_train", line 11, in <module>
    load_entry_point('mrec==0.3.1', 'console_scripts', 'mrec_train')()
File "build/bdist.macosx-10.7-x86_64/egg/mrec/examples/train.py", line 137, in main
File "build/bdist.macosx-10.7-x86_64/egg/mrec/parallel/item_similarity.py", line 39, in run
AttributeError: 'list' object has no attribute 'get'

I also tried to use only ipython. So, I installed some old ipythons (4.1.1 and 3.2.1) , but both failed.

Add --validation_users option to mrec_prepare

It would be useful to have the option of splitting out ratings for some number/proportion of users upfront into a fixed validation set.

How can I sync the pdf reading progress

I read pdf on PC and phone, and I want sync the reading progress, how can I reach it.

missing C source code: mrec/mf/model/warp_fast.c

Hi,

when i try to build mrec, i got this issue: no such file or directory: 'mrec/mf/model/warp_fast.c'

creating build/temp.macosx-10.8-intel-2.7/mrec/mf/model
clang -fno-strict-aliasing -fno-common -dynamic -g -Os -pipe -fno-common -fno-strict-aliasing -fwrapv -mno-fused-madd -DENABLE_DTRACE -DMACOSX -DNDEBUG -Wall -Wstrict-prototypes -Wshorten-64-to-32 -DNDEBUG -g -Os -Wall -Wstrict-prototypes -DENABLE_DTRACE -arch i386 -arch x86_64 -pipe -I/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/numpy/core/include -I/System/Library/Frameworks/Python.framework/Versions/2.7/include/python2.7 -c mrec/mf/model/warp_fast.c -o build/temp.macosx-10.8-intel-2.7/mrec/mf/model/warp_fast.o
clang: error: no such file or directory: 'mrec/mf/model/warp_fast.c'
clang: error: no input files
error: command 'clang' failed with exit status 1

Thanks.
jm

Add a save() method to recommenders

Extend test coverage to all base class methods and helpers

train joint model?

Hi all,

I'm trying to build the joint model using WARP as described in the documentation but the training always hangs...it never produces a model? Any ideas what could be the problem?

Thx in advance

is mendeley plugin/extension available for libreoffice (?) ..

ValueError: numpy.ufunc has the wrong size, try recompiling while training the data.

ValueError: floating-point under-/overflow occurred.

Issuing:
mrec_train -n8 --train=filtered_p1_u40_t50.txt.train.matrix.idx --input_format=tsv --outdir p1u40t50

Yields:
[2014-06-02 13:44:20,204] INFO: 96556 users and 30924 items
[2014-06-02 13:44:20,658] INFO: running 8 tasks in parallel across ipython engines...
Traceback (most recent call last):
File "/usr/local/anaconda/bin/mrec_train", line 9, in
load_entry_point('mrec==0.3.1', 'console_scripts', 'mrec_train')()
File "build/bdist.linux-x86_64/egg/mrec/examples/train.py", line 136, in main
File "build/bdist.linux-x86_64/egg/mrec/parallel/item_similarity.py", line 39, in run
File "/usr/local/anaconda/lib/python2.7/site-packages/IPython/parallel/client/asyncresult.py", line 118, in get
raise self._exception
IPython.parallel.error.CompositeError: one or more exceptions from call to method: process
[3:apply]: ValueError: floating-point under-/overflow occurred.
[1:apply]: ValueError: floating-point under-/overflow occurred.
[0:apply]: ValueError: floating-point under-/overflow occurred.
[5:apply]: ValueError: floating-point under-/overflow occurred.

Usually this happens due to too large regularization learning rate, but there seems to be no option to change the learning rate for SLIM.