Fast Python Collaborative Filtering for Implicit Feedback Datasets

Last update: Dec 31, 2022

Overview

Implicit

Fast Python Collaborative Filtering for Implicit Datasets.

This project provides fast Python implementations of several different popular recommendation algorithms for implicit feedback datasets:

Alternating Least Squares as described in the papers Collaborative Filtering for Implicit Feedback Datasets and Applications of the Conjugate Gradient Method for Implicit Feedback Collaborative Filtering.
Bayesian Personalized Ranking.
Logistic Matrix Factorization
Item-Item Nearest Neighbour models using Cosine, TFIDF or BM25 as a distance metric.

All models have multi-threaded training routines, using Cython and OpenMP to fit the models in parallel among all available CPU cores. In addition, the ALS and BPR models both have custom CUDA kernels - enabling fitting on compatible GPU's. Approximate nearest neighbours libraries such as Annoy, NMSLIB and Faiss can also be used by Implicit to speed up making recommendations.

Installation

There are binary packages on conda-forge for Linux, Windows and OSX. These can be installed with:

conda install -c conda-forge implicit

There are also GPU enabled packages on conda-forge for x86_64 Linux systems using either CUDA 9.2, 10.0, 10.1 or 10.2. The GPU packages can be installed with:

conda install -c conda-forge implicit implicit-proc=*=gpu

There is also an sdist package on PyPi. This package can be installed with:

pip install implicit

Note that installing with pip requires a C++ compiler to be installed on your system, since this method will build implicit from source.

Basic Usage

import implicit

# initialize a model
model = implicit.als.AlternatingLeastSquares(factors=50)

# train the model on a sparse matrix of item/user/confidence weights
model.fit(item_user_data)

# recommend items for a user
user_items = item_user_data.T.tocsr()
recommendations = model.recommend(userid, user_items)

# find related items
related = model.similar_items(itemid)

The examples folder has a program showing how to use this to compute similar artists on the last.fm dataset.

For more information see the documentation.

Articles about Implicit

These blog posts describe the algorithms that power this library:

There are also several other blog posts about using Implicit to build recommendation systems:

Requirements

This library requires SciPy version 0.16 or later. Running on OSX requires an OpenMP compiler, which can be installed with homebrew: brew install gcc. Running on Windows requires Python 3.5+.

GPU Support requires at least version 9 of the NVidia CUDA Toolkit. The build will use the nvcc compiler that is found on the path, but this can be overriden by setting the CUDAHOME enviroment variable to point to your cuda installation.

This library has been tested with Python 2.7, 3.5, 3.6 and 3.7 on Ubuntu and OSX, and tested with Python 3.5 and 3.6 on Windows.

Benchmarks

Simple benchmarks comparing the ALS fitting time versus Spark and QMF can be found here.

Optimal Configuration

I'd recommend configuring SciPy to use Intel's MKL matrix libraries. One easy way of doing this is by installing the Anaconda Python distribution.

For systems using OpenBLAS, I highly recommend setting 'export OPENBLAS_NUM_THREADS=1'. This disables its internal multithreading ability, which leads to substantial speedups for this package. Likewise for Intel MKL, setting 'export MKL_NUM_THREADS=1' should also be set.

Released under the MIT License

Comments

error: Microsoft Visual C++ 14.0 is required. Get it with "Microsoft Visual C++ Build Tools": http://landinghub.visualstudio.com/visual-cpp-build-tools

Hello, I tried pip installing implicit and I get the error below. Is there any work around? running install running build running build_py creating build creating build\lib.win-amd64-3.6 creating build\lib.win-amd64-3.6\implicit copying implicit\als.py -> build\lib.win-amd64-3.6\implicit copying implicit\approximate_als.py -> build\lib.win-amd64-3.6\implicit copying implicit\nearest_neighbours.py -> build\lib.win-amd64-3.6\implicit copying implicit\recommender_base.py -> build\lib.win-amd64-3.6\implicit copying implicit\utils.py -> build\lib.win-amd64-3.6\implicit copying implicit_init_.py -> build\lib.win-amd64-3.6\implicit running build_ext building 'implicit._als' extension error: Microsoft Visual C++ 14.0 is required. Get it with "Microsoft Visual C++ Build Tools": http://landinghub.visualstudio.com/visual-cpp-build-tools

opened by opokualbert 99
Feature/add_bias_terms_to_als

Added optional user and item bias terms to als.py as suggested by ita9naiwa in issue #176 and already implemented in LMF. Set use_bias=True to train with bias terms (not implemented for GPU yet).

opened by howtodowtle 21
Adding user biases and item biases for implicit matrix factorization
Hi @benfred ,

I was wondering whether adding user and item biases would improve the model and recommendations. Did you consider this at some point or experiment with it? Would you expect better performance?

Intuitively, I fear that without biases very popular items will dominate the recommendations more strongly than I want.

Literature:

The original paper only talks about biases in the context of item-based neighbourhood models (section 3.1 in http://yifanhu.net/PUB/cf.pdf) but not in the context of the objective function.

There is this (somewhat obscure) article which claims that performance is better with biases: http://activisiongamescience.github.io/2016/01/11/Implicit-Recommender-Systems-Biased-Matrix-Factorization/

Well, most implementations of implicit matrix factorization (e.g. Spark) don't use biases.

Would love to hear your take on it! Best Simon
opened by SimonCW 15
Difficulty with bm25_weight in Linux install

On Ubuntu 16.04, when I try to import the bm25_weight function I get the following error:

import implicit File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/implicit/init.py", line 3, in from . import nearest_neighbours File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/implicit/nearest_neighbours.py", line 7, in from ._nearest_neighbours import all_pairs_knn ImportError: /home/ubuntu/anaconda3/lib/python3.6/site-packages/implicit/_nearest_neighbours.cpython-36m-x86_64-linux-gnu.so: undefined symbol: _ZdlPvm

opened by angusturner 15
issues while installing pyaudio

when i type command : pip install pyaudio

i have tried several times and installed VS Build tools,but no use....

it shows the error like-

(c) 2019 Microsoft Corporation. All rights reserved.

(venv) C:\Users\Administrator\untitled2>pip install pyaudio Collecting pyaudio Using cached PyAudio-0.2.11.tar.gz (37 kB) Using legacy setup.py install for pyaudio, since package 'wheel' is not installed. Installing collected packages: pyaudio Running setup.py install for pyaudio ... error ERROR: Command errored out with exit status 1: command: 'c:\users\administrator\untitled2\venv\scripts\python.exe' -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'C:\Users\ADMINI~1\AppData\ Local\Temp\pip-install-ikm8s3da\pyaudio\setup.py'"'"'; file='"'"'C:\Users\ADMINI~1\AppData\Local\Temp\pip-install-ikm8s3da\pyaudio\setup.py'"'"';f= getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' inst all --record 'C:\Users\ADMINI~1\AppData\Local\Temp\pip-record-ds3k0l5c\install-record.txt' --single-version-externally-managed --compile --install-headers 'c:\user s\administrator\untitled2\venv\include\site\python3.7\pyaudio' cwd: C:\Users\ADMINI~1\AppData\Local\Temp\pip-install-ikm8s3da\pyaudio
Complete output (9 lines): running install running build running build_py creating build creating build\lib.win-amd64-3.7 copying src\pyaudio.py -> build\lib.win-amd64-3.7 running build_ext building '_portaudio' extension error: Microsoft Visual C++ 14.0 is required. Get it with "Build Tools for Visual Studio": https://visualstudio.microsoft.com/downloads/ ---------------------------------------- ERROR: Command errored out with exit status 1: 'c:\users\administrator\untitled2\venv\scripts\python.exe' -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '" '"'C:\Users\ADMINI~1\AppData\Local\Temp\pip-install-ikm8s3da\pyaudio\setup.py'"'"'; file='"'"'C:\Users\ADMINI~1\AppData\Local\Temp\pip-install-ik m8s3da\pyaudio\setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' install --record 'C:\Users\ADMINI~1\AppData\Local\Temp\pip-record-ds3k0l5c\install-record.txt' --single-version-externally-managed --c ompile --install-headers 'c:\users\administrator\untitled2\venv\include\site\python3.7\pyaudio' Check the logs for full command output.

opened by Saishankar2001 14
Error in similarity score?
I think there is an error in the calculation of the similarity score. I first noticed the problem because I was getting scores larger than 1 and even a score different from 1 for the similarity of an item with itself. I believe I have found the source of the error.

In recommender_base.py the similarity score is defined by the following function:

def _get_similarity_score(self, factor, factors, norms, N): scores = factors.dot(factor) / norms best = np.argpartition(scores, -N)[-N:] return sorted(zip(best, scores[best]), key=lambda x: -x[1]

The error here is that the dot product is being divided by only one of the norms. I think the following definition is correct:

def _get_similarity_score(self, factor, factors, norms, N): scores = factors.dot(factor) / (np.linalg.norm(factor)*norms) best = np.argpartition(scores, -N)[-N:] return sorted(zip(best, scores[best]), key=lambda x: -x[1]

I have tested it locally and it works.
opened by dwolfeu 13
UnboundLocalError: local variable 'vali_user_items' referenced before assignment

`factors=10 regularization=0.01 iterations=15

model = implicit.als.AlternatingLeastSquares(factors=factors,regularization=regularization,iterations=iterations) model.fit(items_users) ` Gives an error

/usr/local/lib/python3.6/dist-packages/implicit/als.py in fit(self, item_users, show_progress) 175 176 if self.use_gpu: --> 177 return self._fit_gpu(Ciu, Cui, vali_user_items, show_progress) 178 179 solver = self.solver

UnboundLocalError: local variable 'vali_user_items' referenced before assignment
bug

opened by filion 12
Can not install on google colaboratory

Hi, I am working on google colaboratory. After new updating, the package can not be installed. It seems it has a problem in bdist_wheel. Running setup.py bdist_wheel for implicit ... error Failed building wheel for implicit

opened by leilanorouzi 12
error: Microsoft Visual C++ 14.0 is required. Get it with "Microsoft Visual C++ Build Tools": http://landinghub.visualstudio.com/visual-cpp-build-tools
I am using "Python 3.6.5 (v3.6.5:f59c0932b4, Mar 28 2018, 17:00:18) [MSC v.1900 64 bit (AMD64)] on win32"

I tried to install "wordcloud" using pip install wordcloud

I faced that error!!!! Any one here to help me how to get rid of it. ////////////////////////////////////////\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

C:\Windows\system32>pip install wordcloud Collecting wordcloud Using cached wordcloud-1.4.1.tar.gz Requirement already satisfied: matplotlib in c:\program files\python36\lib\site-packages (from wordcloud) Requirement already satisfied: numpy>=1.6.1 in c:\program files\python36\lib\site-packages (from wordcloud) Requirement already satisfied: pillow in c:\program files\python36\lib\site-packages (from wordcloud) Requirement already satisfied: six>=1.10 in c:\program files\python36\lib\site-packages (from matplotlib->wordcloud) Requirement already satisfied: python-dateutil>=2.1 in c:\program files\python36\lib\site-packages (from matplotlib->wordcloud) Requirement already satisfied: pytz in c:\program files\python36\lib\site-packages (from matplotlib->wordcloud) Requirement already satisfied: cycler>=0.10 in c:\program files\python36\lib\site-packages (from matplotlib->wordcloud) Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in c:\program files\python36\lib\site-packages (from matplotlib->wordcloud) Requirement already satisfied: kiwisolver>=1.0.1 in c:\program files\python36\lib\site-packages (from matplotlib->wordcloud) Requirement already satisfied: setuptools in c:\program files\python36\lib\site-packages (from kiwisolver>=1.0.1->matplotlib->wordcloud) Installing collected packages: wordcloud Running setup.py install for wordcloud ... error Complete output from command "c:\program files\python36\python.exe" -u -c "import setuptools, tokenize;file='C:\Users\DILSHAD\AppData\Local\Temp\pip-build-s0v4o0v_\wordcloud\setup.py';f=getattr(tokenize, 'open', open)(file);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, file, 'exec'))" install --record C:\Users\DILSHAD\AppData\Local\Temp\pip-0ki_078y-record\install-record.txt --single-version-externally-managed --compile: running install running build running build_py creating build creating build\lib.win-amd64-3.6 creating build\lib.win-amd64-3.6\wordcloud copying wordcloud\color_from_image.py -> build\lib.win-amd64-3.6\wordcloud copying wordcloud\tokenization.py -> build\lib.win-amd64-3.6\wordcloud copying wordcloud\wordcloud.py -> build\lib.win-amd64-3.6\wordcloud copying wordcloud\wordcloud_cli.py -> build\lib.win-amd64-3.6\wordcloud copying wordcloud_init_.py -> build\lib.win-amd64-3.6\wordcloud copying wordcloud\stopwords -> build\lib.win-amd64-3.6\wordcloud copying wordcloud\DroidSansMono.ttf -> build\lib.win-amd64-3.6\wordcloud running build_ext building 'wordcloud.query_integral_image' extension error: Microsoft Visual C++ 14.0 is required. Get it with "Microsoft Visual C++ Build Tools": http://landinghub.visualstudio.com/visual-cpp-build-tools

----------------------------------------

Command ""c:\program files\python36\python.exe" -u -c "import setuptools, tokenize;file='C:\Users\DILSHAD\AppData\Local\Temp\pip-build-s0v4o0v_\wordcloud\setup.py';f=getattr(tokenize, 'open', open)(file);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, file, 'exec'))" install --record C:\Users\DILSHAD\AppData\Local\Temp\pip-0ki_078y-record\install-record.txt --single-version-externally-managed --compile" failed with error code 1 in C:\Users\DILSHAD\AppData\Local\Temp\pip-build-s0v4o0v_\wordcloud\
opened by Dilshad737 12
Install problem: Cannot compile on platform 'posix' with 'nvidia' compiler
I am installing implicit through pip install git+https://github.com/benfred/[email protected]

And I got this error during installation

error: don't know how to compile C/C++ code on platform 'posix' with 'nvidia' compiler ---------------------------------------- ERROR: Failed building wheel for implicit Failed to build implicit ERROR: Could not build wheels for implicit, which is required to install pyproject.toml-based projects

I am using tensorflow/tensorflow:2.3.0-gpu as my docker base image, image digest: 9a52554e85ae666767c28f9af70e56caa7e6d75325951ab021fbd4bb1447a340

This error never happened before, I am not sure what happened to the compiler or which compiler should be used?

Please help me, any help or advise are welcomed.
opened by andyfcx 11

Can't install implicit

I am trying to install the implicit package on windows 10 with Python 2.7.15 using

pip install implicit

There seems to be an error with implicit/bpr.cpp if I am not mistaken. I have no idea what to do.

implicit\bpr.cpp(1277) : fatal error C1003: error count exceeds 100; stopping compilation

This is the full log I am getting:

C:\WINDOWS\system32>pip install implicit
Collecting implicit
  Using cached https://files.pythonhosted.org/packages/d9/0c/7f9f065cceab3c27b7207bbbc3127ff698f1dbaabc2e5f5ef92cb3a39a43/implicit-0.3.6.tar.gz
Requirement already satisfied: numpy in c:\python27\lib\site-packages (from implicit) (1.11.2+mkl)
Requirement already satisfied: scipy>=0.16 in c:\python27\lib\site-packages (from implicit) (0.18.1)
Requirement already satisfied: h5py in c:\python27\lib\site-packages (from implicit) (2.8.0)
Requirement already satisfied: tqdm in c:\python27\lib\site-packages (from implicit) (4.24.0)
Requirement already satisfied: six in c:\python27\lib\site-packages (from h5py->implicit) (1.10.0)
Building wheels for collected packages: implicit
  Running setup.py bdist_wheel for implicit ... error
  Complete output from command c:\python27\python.exe -u -c "import setuptools, tokenize;__file__='c:\\users\\shiro\\appdata\\local\\temp\\pip-install-civoat\\implicit\\setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" bdist_wheel -d c:\users\shiro\appdata\local\temp\pip-wheel-iv5suw --python-tag cp27:
  WARNING:root:The nvcc binary could not be located in your $PATH. Either add it to your path, or set $CUDAHOME to enable CUDA extensions
  Failed to find CUDA toolkit. Building without GPU acceleration.
  running bdist_wheel
  running build
  running build_py
  creating build
  creating build\lib.win-amd64-2.7
  creating build\lib.win-amd64-2.7\implicit
  copying implicit\als.py -> build\lib.win-amd64-2.7\implicit
  copying implicit\approximate_als.py -> build\lib.win-amd64-2.7\implicit
  copying implicit\nearest_neighbours.py -> build\lib.win-amd64-2.7\implicit
  copying implicit\recommender_base.py -> build\lib.win-amd64-2.7\implicit
  copying implicit\utils.py -> build\lib.win-amd64-2.7\implicit
  copying implicit\__init__.py -> build\lib.win-amd64-2.7\implicit
  creating build\lib.win-amd64-2.7\implicit\cuda
  copying implicit\cuda\__init__.py -> build\lib.win-amd64-2.7\implicit\cuda
  creating build\lib.win-amd64-2.7\implicit\datasets
  copying implicit\datasets\lastfm.py -> build\lib.win-amd64-2.7\implicit\datasets
  copying implicit\datasets\million_song_dataset.py -> build\lib.win-amd64-2.7\implicit\datasets
  copying implicit\datasets\movielens.py -> build\lib.win-amd64-2.7\implicit\datasets
  copying implicit\datasets\sketchfab.py -> build\lib.win-amd64-2.7\implicit\datasets
  copying implicit\datasets\_download.py -> build\lib.win-amd64-2.7\implicit\datasets
  copying implicit\datasets\__init__.py -> build\lib.win-amd64-2.7\implicit\datasets
  running build_ext
  building 'implicit._als' extension
  creating build\temp.win-amd64-2.7
  creating build\temp.win-amd64-2.7\Release
  creating build\temp.win-amd64-2.7\Release\implicit
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\cl.exe /c /nologo /Ox /MD /W3 /GS- /DNDEBUG -Ic:\python27\include -Ic:\python27\PC /Tpimplicit\_als.cpp /Fobuild\temp.win-amd64-2.7\Release\implicit\_als.obj /O2 /openmp
  _als.cpp
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Include\xlocale(342) : warning C4530: C++ exception handler used, but unwind semantics are not enabled. Specify /EHsc
  implicit\_als.cpp(3637) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
  implicit\_als.cpp(3919) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
  implicit\_als.cpp(4404) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(4405) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(4850) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(5568) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(5569) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(6732) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(6733) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(7178) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(7896) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(7897) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(9060) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(9061) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(9506) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(10224) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(10225) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(11678) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
  implicit\_als.cpp(11960) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
  implicit\_als.cpp(12468) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(12469) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(12854) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(12999) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(13040) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(13526) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(13527) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(14584) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(14585) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(14970) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(15115) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(15156) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(15642) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(15643) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(16700) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(16701) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(17086) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(17231) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(17272) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(17758) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(17759) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(19070) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
  implicit\_als.cpp(19352) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
  implicit\_als.cpp(19842) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(19843) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(19844) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(20148) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(20652) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(20653) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(20654) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(21462) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(21463) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(21464) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(21768) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(22272) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(22273) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(22274) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(23082) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(23083) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(23084) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(23388) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
  implicit\_als.cpp(23892) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(23893) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  implicit\_als.cpp(23894) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\link.exe /DLL /nologo /INCREMENTAL:NO /LIBPATH:c:\python27\libs /LIBPATH:c:\python27\PCbuild\amd64 /LIBPATH:c:\python27\PC\VS9.0\amd64 /EXPORT:init_als build\temp.win-amd64-2.7\Release\implicit\_als.obj /OUT:build\lib.win-amd64-2.7\implicit\_als.pyd /IMPLIB:build\temp.win-amd64-2.7\Release\implicit\_als.lib /MANIFESTFILE:build\temp.win-amd64-2.7\Release\implicit\_als.pyd.manifest
  _als.obj : warning LNK4197: export 'init_als' specified multiple times; using first specification
	 Creating library build\temp.win-amd64-2.7\Release\implicit\_als.lib and object build\temp.win-amd64-2.7\Release\implicit\_als.exp
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\WinSDK\Bin\x64\mt.exe -nologo -manifest build\temp.win-amd64-2.7\Release\implicit\_als.pyd.manifest -outputresource:build\lib.win-amd64-2.7\implicit\_als.pyd;2
  building 'implicit._nearest_neighbours' extension
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\cl.exe /c /nologo /Ox /MD /W3 /GS- /DNDEBUG -Iimplicit -Ic:\python27\include -Ic:\python27\PC /Tpimplicit\_nearest_neighbours.cpp /Fobuild\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.obj /O2 /openmp
  _nearest_neighbours.cpp
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Include\xlocale(342) : warning C4530: C++ exception handler used, but unwind semantics are not enabled. Specify /EHsc
  implicit\_nearest_neighbours.cpp(2867) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
  implicit\_nearest_neighbours.cpp(3187) : warning C4800: 'int' : forcing value to bool 'true' or 'false' (performance warning)
  implicit\_nearest_neighbours.cpp(3604) : warning C4267: '=' : conversion from 'size_t' to 'int', possible loss of data
  implicit\_nearest_neighbours.cpp(4027) : warning C4800: 'int' : forcing value to bool 'true' or 'false' (performance warning)
  implicit\_nearest_neighbours.cpp(4444) : warning C4267: '=' : conversion from 'size_t' to 'int', possible loss of data
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\link.exe /DLL /nologo /INCREMENTAL:NO /LIBPATH:c:\python27\libs /LIBPATH:c:\python27\PCbuild\amd64 /LIBPATH:c:\python27\PC\VS9.0\amd64 /EXPORT:init_nearest_neighbours build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.obj /OUT:build\lib.win-amd64-2.7\implicit\_nearest_neighbours.pyd /IMPLIB:build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.lib /MANIFESTFILE:build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.pyd.manifest
  _nearest_neighbours.obj : warning LNK4197: export 'init_nearest_neighbours' specified multiple times; using first specification
	 Creating library build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.lib and object build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.exp
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\WinSDK\Bin\x64\mt.exe -nologo -manifest build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.pyd.manifest -outputresource:build\lib.win-amd64-2.7\implicit\_nearest_neighbours.pyd;2
  building 'implicit.bpr' extension
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\cl.exe /c /nologo /Ox /MD /W3 /GS- /DNDEBUG -Iimplicit -Ic:\python27\include -Ic:\python27\PC /Tpimplicit\bpr.cpp /Fobuild\temp.win-amd64-2.7\Release\implicit\bpr.obj /O2 /openmp
  bpr.cpp
  C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Include\xlocale(342) : warning C4530: C++ exception handler used, but unwind semantics are not enabled. Specify /EHsc
  implicit\bpr.cpp(954) : error C2039: 'mt19937' : is not a member of 'std'
  implicit\bpr.cpp(954) : error C2065: 'mt19937' : undeclared identifier
  implicit\bpr.cpp(955) : error C2039: 'uniform_int_distribution' : is not a member of 'std'
  implicit\bpr.cpp(955) : error C2065: 'uniform_int_distribution' : undeclared identifier
  implicit\bpr.cpp(955) : error C2062: type 'long' unexpected
  implicit\bpr.cpp(956) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(966) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(981) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(991) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(994) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(1004) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(1017) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(1027) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(1032) : error C2059: syntax error : ')'
  implicit\bpr.cpp(1033) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(1045) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(1047) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(1060) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(1062) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(1074) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(1075) : error C2059: syntax error : ')'
  implicit\bpr.cpp(1076) : error C2059: syntax error : ')'
  implicit\bpr.cpp(1077) : error C2059: syntax error : ','
  implicit\bpr.cpp(1077) : error C2059: syntax error : ')'
  implicit\bpr.cpp(1078) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1079) : error C2059: syntax error : ','
  implicit\bpr.cpp(1079) : error C2059: syntax error : ')'
  implicit\bpr.cpp(1080) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1081) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1082) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(1094) : error C2143: syntax error : missing ';' before '{'
  implicit\bpr.cpp(1096) : error C2143: syntax error : missing ';' before '}'
  implicit\bpr.cpp(1165) : error C2065: 'attr_name' : undeclared identifier
  implicit\bpr.cpp(1165) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1165) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1178) : error C2065: 'kw_name' : undeclared identifier
  implicit\bpr.cpp(1178) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1178) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1181) : error C2065: 'argnames' : undeclared identifier
  implicit\bpr.cpp(1181) : error C2059: syntax error : ']'
  implicit\bpr.cpp(1182) : error C2065: 'kwds2' : undeclared identifier
  implicit\bpr.cpp(1182) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1182) : error C2146: syntax error : missing ',' before identifier 'values'
  implicit\bpr.cpp(1182) : error C2065: 'values' : undeclared identifier
  implicit\bpr.cpp(1182) : error C2059: syntax error : ']'
  implicit\bpr.cpp(1183) : error C2146: syntax error : missing ',' before identifier 'function_name'
  implicit\bpr.cpp(1183) : error C2065: 'function_name' : undeclared identifier
  implicit\bpr.cpp(1183) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1190) : error C2065: 'arg' : undeclared identifier
  implicit\bpr.cpp(1190) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1190) : error C2146: syntax error : missing ',' before identifier 'kw'
  implicit\bpr.cpp(1190) : error C2065: 'kw' : undeclared identifier
  implicit\bpr.cpp(1190) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1213) : error C2065: 'type' : undeclared identifier
  implicit\bpr.cpp(1213) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1213) : error C2146: syntax error : missing ',' before identifier 'value'
  implicit\bpr.cpp(1213) : error C2065: 'value' : undeclared identifier
  implicit\bpr.cpp(1213) : error C2146: syntax error : missing ',' before identifier 'tb'
  implicit\bpr.cpp(1213) : error C2065: 'tb' : undeclared identifier
  implicit\bpr.cpp(1213) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1214) : error C2065: 'type' : undeclared identifier
  implicit\bpr.cpp(1214) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1214) : error C2146: syntax error : missing ',' before identifier 'value'
  implicit\bpr.cpp(1214) : error C2065: 'value' : undeclared identifier
  implicit\bpr.cpp(1214) : error C2146: syntax error : missing ',' before identifier 'tb'
  implicit\bpr.cpp(1214) : error C2065: 'tb' : undeclared identifier
  implicit\bpr.cpp(1214) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1232) : error C2065: 'value' : undeclared identifier
  implicit\bpr.cpp(1232) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1232) : error C2146: syntax error : missing ',' before identifier 'tb'
  implicit\bpr.cpp(1232) : error C2065: 'tb' : undeclared identifier
  implicit\bpr.cpp(1232) : error C2146: syntax error : missing ',' before identifier 'cause'
  implicit\bpr.cpp(1232) : error C2065: 'cause' : undeclared identifier
  implicit\bpr.cpp(1232) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1246) : error C2065: 'args' : undeclared identifier
  implicit\bpr.cpp(1246) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1246) : error C2146: syntax error : missing ',' before identifier 'nargs'
  implicit\bpr.cpp(1246) : error C2065: 'nargs' : undeclared identifier
  implicit\bpr.cpp(1246) : error C2146: syntax error : missing ',' before identifier 'kwargs'
  implicit\bpr.cpp(1246) : error C2065: 'kwargs' : undeclared identifier
  implicit\bpr.cpp(1246) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1254) : error C2065: 'arg' : undeclared identifier
  implicit\bpr.cpp(1254) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1254) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1258) : error C2065: 'arg' : undeclared identifier
  implicit\bpr.cpp(1258) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1258) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1269) : error C2065: 'op2' : undeclared identifier
  implicit\bpr.cpp(1269) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1269) : error C2146: syntax error : missing ',' before identifier 'intval'
  implicit\bpr.cpp(1269) : error C2065: 'intval' : undeclared identifier
  implicit\bpr.cpp(1269) : error C2146: syntax error : missing ',' before identifier 'inplace'
  implicit\bpr.cpp(1269) : error C2065: 'inplace' : undeclared identifier
  implicit\bpr.cpp(1269) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1277) : error C2065: 'op2' : undeclared identifier
  implicit\bpr.cpp(1277) : error C2275: 'PyObject' : illegal use of this type as an expression
		  c:\python27\include\object.h(108) : see declaration of 'PyObject'
  implicit\bpr.cpp(1277) : error C2146: syntax error : missing ',' before identifier 'intval'
  implicit\bpr.cpp(1277) : error C2065: 'intval' : undeclared identifier
  implicit\bpr.cpp(1277) : error C2146: syntax error : missing ',' before identifier 'inplace'
  implicit\bpr.cpp(1277) : error C2065: 'inplace' : undeclared identifier
  implicit\bpr.cpp(1277) : error C2143: syntax error : missing ',' before ')'
  implicit\bpr.cpp(1277) : fatal error C1003: error count exceeds 100; stopping compilation
  error: command 'C:\\Users\\shiro\\AppData\\Local\\Programs\\Common\\Microsoft\\Visual C++ for Python\\9.0\\VC\\Bin\\amd64\\cl.exe' failed with exit status 2

  ----------------------------------------
  Failed building wheel for implicit
  Running setup.py clean for implicit
Failed to build implicit
Installing collected packages: implicit
  Running setup.py install for implicit ... error
	Complete output from command c:\python27\python.exe -u -c "import setuptools, tokenize;__file__='c:\\users\\shiro\\appdata\\local\\temp\\pip-install-civoat\\implicit\\setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" install --record c:\users\shiro\appdata\local\temp\pip-record-bhxxuv\install-record.txt --single-version-externally-managed --compile:
	WARNING:root:The nvcc binary could not be located in your $PATH. Either add it to your path, or set $CUDAHOME to enable CUDA extensions
	Failed to find CUDA toolkit. Building without GPU acceleration.
	running install
	running build
	running build_py
	creating build
	creating build\lib.win-amd64-2.7
	creating build\lib.win-amd64-2.7\implicit
	copying implicit\als.py -> build\lib.win-amd64-2.7\implicit
	copying implicit\approximate_als.py -> build\lib.win-amd64-2.7\implicit
	copying implicit\nearest_neighbours.py -> build\lib.win-amd64-2.7\implicit
	copying implicit\recommender_base.py -> build\lib.win-amd64-2.7\implicit
	copying implicit\utils.py -> build\lib.win-amd64-2.7\implicit
	copying implicit\__init__.py -> build\lib.win-amd64-2.7\implicit
	creating build\lib.win-amd64-2.7\implicit\cuda
	copying implicit\cuda\__init__.py -> build\lib.win-amd64-2.7\implicit\cuda
	creating build\lib.win-amd64-2.7\implicit\datasets
	copying implicit\datasets\lastfm.py -> build\lib.win-amd64-2.7\implicit\datasets
	copying implicit\datasets\million_song_dataset.py -> build\lib.win-amd64-2.7\implicit\datasets
	copying implicit\datasets\movielens.py -> build\lib.win-amd64-2.7\implicit\datasets
	copying implicit\datasets\sketchfab.py -> build\lib.win-amd64-2.7\implicit\datasets
	copying implicit\datasets\_download.py -> build\lib.win-amd64-2.7\implicit\datasets
	copying implicit\datasets\__init__.py -> build\lib.win-amd64-2.7\implicit\datasets
	running build_ext
	building 'implicit._als' extension
	creating build\temp.win-amd64-2.7
	creating build\temp.win-amd64-2.7\Release
	creating build\temp.win-amd64-2.7\Release\implicit
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\cl.exe /c /nologo /Ox /MD /W3 /GS- /DNDEBUG -Ic:\python27\include -Ic:\python27\PC /Tpimplicit\_als.cpp /Fobuild\temp.win-amd64-2.7\Release\implicit\_als.obj /O2 /openmp
	_als.cpp
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Include\xlocale(342) : warning C4530: C++ exception handler used, but unwind semantics are not enabled. Specify /EHsc
	implicit\_als.cpp(3637) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
	implicit\_als.cpp(3919) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
	implicit\_als.cpp(4404) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(4405) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(4850) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(5568) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(5569) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(6732) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(6733) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(7178) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(7896) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(7897) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(9060) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(9061) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(9506) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(10224) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(10225) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(11678) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
	implicit\_als.cpp(11960) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
	implicit\_als.cpp(12468) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(12469) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(12854) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(12999) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(13040) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(13526) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(13527) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(14584) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(14585) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(14970) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(15115) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(15156) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(15642) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(15643) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(16700) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(16701) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(17086) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(17231) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(17272) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(17758) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(17759) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(19070) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
	implicit\_als.cpp(19352) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
	implicit\_als.cpp(19842) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(19843) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(19844) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(20148) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(20652) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(20653) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(20654) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(21462) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(21463) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(21464) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(21768) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(22272) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(22273) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(22274) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(23082) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(23083) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(23084) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(23388) : warning C4244: '=' : conversion from 'double' to 'float', possible loss of data
	implicit\_als.cpp(23892) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(23893) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	implicit\_als.cpp(23894) : warning C4244: '=' : conversion from 'Py_ssize_t' to 'int', possible loss of data
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\link.exe /DLL /nologo /INCREMENTAL:NO /LIBPATH:c:\python27\libs /LIBPATH:c:\python27\PCbuild\amd64 /LIBPATH:c:\python27\PC\VS9.0\amd64 /EXPORT:init_als build\temp.win-amd64-2.7\Release\implicit\_als.obj /OUT:build\lib.win-amd64-2.7\implicit\_als.pyd /IMPLIB:build\temp.win-amd64-2.7\Release\implicit\_als.lib /MANIFESTFILE:build\temp.win-amd64-2.7\Release\implicit\_als.pyd.manifest
	_als.obj : warning LNK4197: export 'init_als' specified multiple times; using first specification
	   Creating library build\temp.win-amd64-2.7\Release\implicit\_als.lib and object build\temp.win-amd64-2.7\Release\implicit\_als.exp
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\WinSDK\Bin\x64\mt.exe -nologo -manifest build\temp.win-amd64-2.7\Release\implicit\_als.pyd.manifest -outputresource:build\lib.win-amd64-2.7\implicit\_als.pyd;2
	building 'implicit._nearest_neighbours' extension
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\cl.exe /c /nologo /Ox /MD /W3 /GS- /DNDEBUG -Iimplicit -Ic:\python27\include -Ic:\python27\PC /Tpimplicit\_nearest_neighbours.cpp /Fobuild\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.obj /O2 /openmp
	_nearest_neighbours.cpp
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Include\xlocale(342) : warning C4530: C++ exception handler used, but unwind semantics are not enabled. Specify /EHsc
	implicit\_nearest_neighbours.cpp(2867) : warning C4244: '=' : conversion from 'long' to 'char', possible loss of data
	implicit\_nearest_neighbours.cpp(3187) : warning C4800: 'int' : forcing value to bool 'true' or 'false' (performance warning)
	implicit\_nearest_neighbours.cpp(3604) : warning C4267: '=' : conversion from 'size_t' to 'int', possible loss of data
	implicit\_nearest_neighbours.cpp(4027) : warning C4800: 'int' : forcing value to bool 'true' or 'false' (performance warning)
	implicit\_nearest_neighbours.cpp(4444) : warning C4267: '=' : conversion from 'size_t' to 'int', possible loss of data
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\link.exe /DLL /nologo /INCREMENTAL:NO /LIBPATH:c:\python27\libs /LIBPATH:c:\python27\PCbuild\amd64 /LIBPATH:c:\python27\PC\VS9.0\amd64 /EXPORT:init_nearest_neighbours build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.obj /OUT:build\lib.win-amd64-2.7\implicit\_nearest_neighbours.pyd /IMPLIB:build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.lib /MANIFESTFILE:build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.pyd.manifest
	_nearest_neighbours.obj : warning LNK4197: export 'init_nearest_neighbours' specified multiple times; using first specification
	   Creating library build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.lib and object build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.exp
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\WinSDK\Bin\x64\mt.exe -nologo -manifest build\temp.win-amd64-2.7\Release\implicit\_nearest_neighbours.pyd.manifest -outputresource:build\lib.win-amd64-2.7\implicit\_nearest_neighbours.pyd;2
	building 'implicit.bpr' extension
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Bin\amd64\cl.exe /c /nologo /Ox /MD /W3 /GS- /DNDEBUG -Iimplicit -Ic:\python27\include -Ic:\python27\PC /Tpimplicit\bpr.cpp /Fobuild\temp.win-amd64-2.7\Release\implicit\bpr.obj /O2 /openmp
	bpr.cpp
	C:\Users\shiro\AppData\Local\Programs\Common\Microsoft\Visual C++ for Python\9.0\VC\Include\xlocale(342) : warning C4530: C++ exception handler used, but unwind semantics are not enabled. Specify /EHsc
	implicit\bpr.cpp(954) : error C2039: 'mt19937' : is not a member of 'std'
	implicit\bpr.cpp(954) : error C2065: 'mt19937' : undeclared identifier
	implicit\bpr.cpp(955) : error C2039: 'uniform_int_distribution' : is not a member of 'std'
	implicit\bpr.cpp(955) : error C2065: 'uniform_int_distribution' : undeclared identifier
	implicit\bpr.cpp(955) : error C2062: type 'long' unexpected
	implicit\bpr.cpp(956) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(966) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(981) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(991) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(994) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(1004) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(1017) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(1027) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(1032) : error C2059: syntax error : ')'
	implicit\bpr.cpp(1033) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(1045) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(1047) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(1060) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(1062) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(1074) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(1075) : error C2059: syntax error : ')'
	implicit\bpr.cpp(1076) : error C2059: syntax error : ')'
	implicit\bpr.cpp(1077) : error C2059: syntax error : ','
	implicit\bpr.cpp(1077) : error C2059: syntax error : ')'
	implicit\bpr.cpp(1078) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1079) : error C2059: syntax error : ','
	implicit\bpr.cpp(1079) : error C2059: syntax error : ')'
	implicit\bpr.cpp(1080) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1081) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1082) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(1094) : error C2143: syntax error : missing ';' before '{'
	implicit\bpr.cpp(1096) : error C2143: syntax error : missing ';' before '}'
	implicit\bpr.cpp(1165) : error C2065: 'attr_name' : undeclared identifier
	implicit\bpr.cpp(1165) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1165) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1178) : error C2065: 'kw_name' : undeclared identifier
	implicit\bpr.cpp(1178) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1178) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1181) : error C2065: 'argnames' : undeclared identifier
	implicit\bpr.cpp(1181) : error C2059: syntax error : ']'
	implicit\bpr.cpp(1182) : error C2065: 'kwds2' : undeclared identifier
	implicit\bpr.cpp(1182) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1182) : error C2146: syntax error : missing ',' before identifier 'values'
	implicit\bpr.cpp(1182) : error C2065: 'values' : undeclared identifier
	implicit\bpr.cpp(1182) : error C2059: syntax error : ']'
	implicit\bpr.cpp(1183) : error C2146: syntax error : missing ',' before identifier 'function_name'
	implicit\bpr.cpp(1183) : error C2065: 'function_name' : undeclared identifier
	implicit\bpr.cpp(1183) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1190) : error C2065: 'arg' : undeclared identifier
	implicit\bpr.cpp(1190) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1190) : error C2146: syntax error : missing ',' before identifier 'kw'
	implicit\bpr.cpp(1190) : error C2065: 'kw' : undeclared identifier
	implicit\bpr.cpp(1190) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1213) : error C2065: 'type' : undeclared identifier
	implicit\bpr.cpp(1213) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1213) : error C2146: syntax error : missing ',' before identifier 'value'
	implicit\bpr.cpp(1213) : error C2065: 'value' : undeclared identifier
	implicit\bpr.cpp(1213) : error C2146: syntax error : missing ',' before identifier 'tb'
	implicit\bpr.cpp(1213) : error C2065: 'tb' : undeclared identifier
	implicit\bpr.cpp(1213) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1214) : error C2065: 'type' : undeclared identifier
	implicit\bpr.cpp(1214) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1214) : error C2146: syntax error : missing ',' before identifier 'value'
	implicit\bpr.cpp(1214) : error C2065: 'value' : undeclared identifier
	implicit\bpr.cpp(1214) : error C2146: syntax error : missing ',' before identifier 'tb'
	implicit\bpr.cpp(1214) : error C2065: 'tb' : undeclared identifier
	implicit\bpr.cpp(1214) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1232) : error C2065: 'value' : undeclared identifier
	implicit\bpr.cpp(1232) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1232) : error C2146: syntax error : missing ',' before identifier 'tb'
	implicit\bpr.cpp(1232) : error C2065: 'tb' : undeclared identifier
	implicit\bpr.cpp(1232) : error C2146: syntax error : missing ',' before identifier 'cause'
	implicit\bpr.cpp(1232) : error C2065: 'cause' : undeclared identifier
	implicit\bpr.cpp(1232) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1246) : error C2065: 'args' : undeclared identifier
	implicit\bpr.cpp(1246) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1246) : error C2146: syntax error : missing ',' before identifier 'nargs'
	implicit\bpr.cpp(1246) : error C2065: 'nargs' : undeclared identifier
	implicit\bpr.cpp(1246) : error C2146: syntax error : missing ',' before identifier 'kwargs'
	implicit\bpr.cpp(1246) : error C2065: 'kwargs' : undeclared identifier
	implicit\bpr.cpp(1246) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1254) : error C2065: 'arg' : undeclared identifier
	implicit\bpr.cpp(1254) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1254) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1258) : error C2065: 'arg' : undeclared identifier
	implicit\bpr.cpp(1258) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1258) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1269) : error C2065: 'op2' : undeclared identifier
	implicit\bpr.cpp(1269) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1269) : error C2146: syntax error : missing ',' before identifier 'intval'
	implicit\bpr.cpp(1269) : error C2065: 'intval' : undeclared identifier
	implicit\bpr.cpp(1269) : error C2146: syntax error : missing ',' before identifier 'inplace'
	implicit\bpr.cpp(1269) : error C2065: 'inplace' : undeclared identifier
	implicit\bpr.cpp(1269) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1277) : error C2065: 'op2' : undeclared identifier
	implicit\bpr.cpp(1277) : error C2275: 'PyObject' : illegal use of this type as an expression
			c:\python27\include\object.h(108) : see declaration of 'PyObject'
	implicit\bpr.cpp(1277) : error C2146: syntax error : missing ',' before identifier 'intval'
	implicit\bpr.cpp(1277) : error C2065: 'intval' : undeclared identifier
	implicit\bpr.cpp(1277) : error C2146: syntax error : missing ',' before identifier 'inplace'
	implicit\bpr.cpp(1277) : error C2065: 'inplace' : undeclared identifier
	implicit\bpr.cpp(1277) : error C2143: syntax error : missing ',' before ')'
	implicit\bpr.cpp(1277) : fatal error C1003: error count exceeds 100; stopping compilation
	error: command 'C:\\Users\\shiro\\AppData\\Local\\Programs\\Common\\Microsoft\\Visual C++ for Python\\9.0\\VC\\Bin\\amd64\\cl.exe' failed with exit status 2

	----------------------------------------
Command "c:\python27\python.exe -u -c "import setuptools, tokenize;__file__='c:\\users\\shiro\\appdata\\local\\temp\\pip-install-civoat\\implicit\\setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" install --record c:\users\shiro\appdata\local\temp\pip-record-bhxxuv\install-record.txt --single-version-externally-managed --compile" failed with error code 1 in c:\users\shiro\appdata\local\temp\pip-install-civoat\implicit\

opened by dimitris93 11

Fix incorrect test set values in leave_k_out splits with sparse user rows
Closes #639

This PR fixes a bug in the evaluation of the leave_k_out_split in which the produced test matrix would contain values that were many multiples of their original value. Tests are also added on static (non-random) matrices that otherwise fail in the un-corrected implementation.

This bug resulted from a calculation that required an input array with sequential values - the fact that non-sequential values were provided led to an error in processing.

Specifically, the arr argument in _take_tails

---------- arr: ndarray The input array. This should be an array of integers in the range 0->n, where the ordered unique set of integers in said array should produce an array of consecutive integers. Concretely, the array [1, 0, 1, 1, 0, 3] would be invalid, but the array [1, 0, 1, 1, 0, 2] would not be.

was being provided as candidate_users, from which user indices falling below the threshold were removed, resulting in a list in which the ordered set of unique integers was not consecutive and therefore the provided array was invalid.
opened by chrisjkuch 0

leave_k_out_split produces incorrect values in test set

I noticed odd behavior in leave_k_out_split - specifically, under certain circumstances (many rows with one value?), the number returned for the withheld test set value is different from the actual value. This causes train + test to fail to reconstruct the original matrix.

Script to reproduce:

import implicit

implicit.__version__
#> '0.6.2'

from implicit.evaluation import leave_k_out_split
from scipy import sparse

ratings = sparse.csr_matrix(
    [[3, 2, 1, 0], [1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0], [0, 1, 1, 1]]
)

train, test = leave_k_out_split(ratings, K=1, random_state=42)

diff = (train + test) - ratings

diff.nnz
#> 1

train.todense()
#> matrix([[3, 2, 0, 0],
#>         [1, 0, 0, 0],
#>         [0, 1, 0, 0],
#>         [0, 0, 1, 0],
#>         [0, 1, 0, 1]])

test.todense()
#> matrix([[0, 0, 4, 0],
#>         [0, 0, 0, 0],
#>         [0, 0, 0, 0],
#>         [0, 0, 0, 0],
#>         [0, 0, 1, 0]])

diff.todense()
#> matrix([[0, 0, 3, 0],
#>         [0, 0, 0, 0],
#>         [0, 0, 0, 0],
#>         [0, 0, 0, 0],
#>         [0, 0, 0, 0]])

^{Created at 2022-12-23 13:59:28 CST by reprexlite v0.5.0}

I believe the issue is in _take_tails - the returned test_idx array has multiple copies of the first user index returned, so we end up with a test set value that is copied multiple times.

(As an aside, the call to _take_tails when shuffled=True does not pass on the rng, so the random state cannot be maintained.)

opened by chrisjkuch 1

Add ialspp optimizer in cpu.ALS
This PR implements

iALS++: Speeding up Matrix Factorization with Subspace Optimization

Revisiting the Performance of iALS on Item Recommendation Benchmarks , which is an approximate updating method for ALS.

correctness verification

I get ml-1m data and preprocessed as below

counts = get_movielens(variant='1m') user_items = counts.T.tocsr() user_items.data[user_items.data <= 4.0] = 0 user_items.eliminate_zeros() user_items.data[:] = 1.0

and run ALS for both CG and ialspp approach using following hyperparameter set

model = AlternatingLeastSquares( factors=512, regularization=2.0, alpha=5.0, use_native=True, use_cg=False, use_gpu=False, use_ialspp=True, iterations=10, calculate_training_loss=False, num_threads=12, )

and ranking metrics give almost same results

ranking_metrics_at_k(model, tr, te, K=50) > {'precision': 0.1126659490491568, 'map': 0.028802994722126836, 'ndcg': 0.07890071767410052, 'auc': 0.5602418716459558} > {'precision': 0.11149982059562254, 'map': 0.029011087049113284, 'ndcg': 0.07887135532015757, 'auc': 0.5603164551005837}

Training Speed-Up

i found that this approach's performance gain vary across implementation.

In Buffalo, I observed performance gain from 256 dim, but in Implicit, training becomes fast from num factors of 512.

Speedup along dimensions

| 512 | 1024 | 2048 | 4096 -- | -- | -- | -- | -- ials++ | 62.49 | 64.57 | 78.8 | 87.42 cg | 70.62 | 107.11 | 124.4 | 610.32

Average training time per iteration becomes 1.6-1.8 times faster in moderate dimensions [512, 2048], but extremely huge more than 4096 dimensions. However, I think it is not very realistic to use dimensions larger than 2048, though.
opened by ita9naiwa 0
Method 'recommend' is broken
Hi,

I see a problem with current implementation of "recommend" function.

==== Example of wrong behaviour ====

For Last fm: current behaviour with default parameters:

most probably expected behaviour (however still not sure if output is correct):

==== Problem ====

There is some problem in recommend function logic for class MatrixFactorizationBase. I might be wrong, but you can't expect user_items.shape[0] != user_count and at the same time have current version of class method user = self._user_factor(....). I think the correct logic should be user_items.shape[0] > user_count where user_count = max(user_id)?

Consider that by default user most probably will do on of this two calls:

model.recommend(11, user_plays)- this won't work in current version, but worked in version 0.4.0 model.recommend(11, user_plays[11:12])- this will work in 0.6.1, but then it will get strange behaviour after calling: self._user_factor(userid, user_items, recalculate_user).

==== Code snippet with current implementation: ====

def recommend( self, userid, user_items, N=10, filter_already_liked_items=True, filter_items=None, recalculate_user=False, items=None, ): if filter_already_liked_items or recalculate_user: if not isinstance(user_items, csr_matrix): raise ValueError("user_items needs to be a CSR sparse matrix") user_count = 1 if np.isscalar(userid) else len(userid) if user_items.shape[0] != user_count: raise ValueError("user_items must contain 1 row for every user in userids") user = self._user_factor(userid, user_items, recalculate_user)
opened by IvanPetrovMck 3
NDCG computation
Hello @benfred. I think there might be an issue with the computation of NDCG in ranking_metrics_at_k. I believe NDCG should be calculated by first sorting the recommended ids by the recommendation scores, so that the logarithmic discount is applied in the correct ranking of the recommendations. Right now, Implicit's recommend methods do not return the ids sorted by their descending scores, so the way NDCG is being calculated does not take into account the order of the ranking.

I would appreciate your opinion, maybe I'm missing something and I'm wrong. Thanks in advance.

If changes turn out to be necessary, I would suggest adding some lines on this part of ranking_metrics_at_k

while start_idx < len(to_generate): batch = to_generate[start_idx: start_idx + batch_size] ids, scores = model.recommend(batch, train_user_items[batch], N=K) start_idx += batch_size # added sorted_scores = np.flip(np.argsort(scores, axis=1), axis=1) ids = ids[np.arange(ids.shape[0])[:, None], sorted_scores]
opened by malonsocortes 2
.recommend return different output when different 'N' parameter is specified for bm25 recommender

I trained a bm25 nearest neighbor recommender. When I specify different 'N' parameter in the model.recommend() method, I get different scores and order for the recommendations. With ALS model, it returns the same recommendations for different N. Why would it return different rank and scores for different values of N for the bm25 nearest neighbor?

opened by saranggupta 2

Releases(v0.6.2)

v0.6.2(Dec 11, 2022)
Changes

🚀 Features

Build arm64 wheels for linux @benfred (#630)

Build python 3.11 wheels @benfred (#634)

🐛 Bug Fixes

Fix more issues with pickling GPU models @benfred (#636)

Fix pickling GPU models @benfred (#632)

Fix ordering in .recommend for KNN models @benfred (#629)

📄 Documentation

Update README / github actions versions @benfred (#635)

Update evaluation.pyx @mrticker (#617)

Fix CosineRecommender documentation @benfred (#619)

🧰 Maintenance

fix for newer packaging versions @benfred (#633)

Remove unnecessary argument @Soonmok (#621)

Source code(tar.gz)
Source code(zip)
implicit-0.6.2-cp310-cp310-macosx_10_9_universal2.whl(1.28 MB)
implicit-0.6.2-cp310-cp310-macosx_10_9_x86_64.whl(706.54 KB)
implicit-0.6.2-cp310-cp310-macosx_11_0_arm64.whl(653.20 KB)
implicit-0.6.2-cp310-cp310-manylinux2014_aarch64.whl(848.46 KB)
implicit-0.6.2-cp310-cp310-manylinux2014_x86_64.whl(17.72 MB)
implicit-0.6.2-cp310-cp310-win_amd64.whl(619.60 KB)
implicit-0.6.2-cp311-cp311-macosx_10_9_universal2.whl(1.24 MB)
implicit-0.6.2-cp311-cp311-macosx_10_9_x86_64.whl(690.59 KB)
implicit-0.6.2-cp311-cp311-macosx_11_0_arm64.whl(634.03 KB)
implicit-0.6.2-cp311-cp311-manylinux2014_aarch64.whl(852.49 KB)
implicit-0.6.2-cp311-cp311-manylinux2014_x86_64.whl(17.72 MB)
implicit-0.6.2-cp311-cp311-win_amd64.whl(618.27 KB)
implicit-0.6.2-cp36-cp36m-macosx_10_9_x86_64.whl(686.32 KB)
implicit-0.6.2-cp36-cp36m-manylinux2014_aarch64.whl(851.52 KB)
implicit-0.6.2-cp36-cp36m-manylinux2014_x86_64.whl(17.71 MB)
implicit-0.6.2-cp36-cp36m-win_amd64.whl(618.69 KB)
implicit-0.6.2-cp37-cp37m-macosx_10_9_x86_64.whl(687.25 KB)
implicit-0.6.2-cp37-cp37m-manylinux2014_aarch64.whl(846.92 KB)
implicit-0.6.2-cp37-cp37m-manylinux2014_x86_64.whl(17.71 MB)
implicit-0.6.2-cp37-cp37m-win_amd64.whl(618.65 KB)
implicit-0.6.2-cp38-cp38-macosx_10_9_universal2.whl(1.24 MB)
implicit-0.6.2-cp38-cp38-macosx_10_9_x86_64.whl(688.11 KB)
implicit-0.6.2-cp38-cp38-macosx_11_0_arm64.whl(634.23 KB)
implicit-0.6.2-cp38-cp38-manylinux2014_aarch64.whl(861.84 KB)
implicit-0.6.2-cp38-cp38-manylinux2014_x86_64.whl(17.73 MB)
implicit-0.6.2-cp38-cp38-win_amd64.whl(631.84 KB)
implicit-0.6.2-cp39-cp39-macosx_10_9_universal2.whl(1.27 MB)
implicit-0.6.2-cp39-cp39-macosx_10_9_x86_64.whl(705.17 KB)
implicit-0.6.2-cp39-cp39-macosx_11_0_arm64.whl(649.04 KB)
implicit-0.6.2-cp39-cp39-manylinux2014_aarch64.whl(860.29 KB)
implicit-0.6.2-cp39-cp39-manylinux2014_x86_64.whl(17.73 MB)
implicit-0.6.2-cp39-cp39-win_amd64.whl(632.04 KB)
v0.6.1(Aug 31, 2022)
Changes

🚀 Features

Tweak callback API @benfred (#612)

Allow GPU models to train on sparse matrices that exceed the size of available GPU memory @benfred (#605)

Fit callback in BPR model @apat1n (#590)

🐛 Bug Fixes

Handle more than 2**31 parameters on GPU models @benfred (#599)

Fix recalculate_item / recalculate_user on the GPU @benfred (#598)

📄 Documentation

Minor documentation fixes @benfred (#609)

docs: Fix a few typos @timgates42 (#596)

Source code(tar.gz)
Source code(zip)
implicit-0.6.1-cp310-cp310-macosx_10_9_universal2.whl(1.27 MB)
implicit-0.6.1-cp310-cp310-macosx_10_9_x86_64.whl(1014.01 KB)
implicit-0.6.1-cp310-cp310-macosx_11_0_arm64.whl(646.80 KB)
implicit-0.6.1-cp310-cp310-manylinux2014_x86_64.whl(17.72 MB)
implicit-0.6.1-cp310-cp310-win_amd64.whl(618.31 KB)
implicit-0.6.1-cp36-cp36m-macosx_10_9_x86_64.whl(987.62 KB)
implicit-0.6.1-cp36-cp36m-manylinux2014_x86_64.whl(17.72 MB)
implicit-0.6.1-cp36-cp36m-win_amd64.whl(617.32 KB)
implicit-0.6.1-cp37-cp37m-macosx_10_9_x86_64.whl(988.30 KB)
implicit-0.6.1-cp37-cp37m-manylinux2014_x86_64.whl(17.70 MB)
implicit-0.6.1-cp37-cp37m-win_amd64.whl(617.23 KB)
implicit-0.6.1-cp38-cp38-macosx_10_9_universal2.whl(1.24 MB)
implicit-0.6.1-cp38-cp38-macosx_10_9_x86_64.whl(994.83 KB)
implicit-0.6.1-cp38-cp38-macosx_11_0_arm64.whl(631.51 KB)
implicit-0.6.1-cp38-cp38-manylinux2014_x86_64.whl(17.73 MB)
implicit-0.6.1-cp38-cp38-win_amd64.whl(630.91 KB)
implicit-0.6.1-cp39-cp39-macosx_10_9_universal2.whl(1.27 MB)
implicit-0.6.1-cp39-cp39-macosx_10_9_x86_64.whl(1013.76 KB)
implicit-0.6.1-cp39-cp39-macosx_11_0_arm64.whl(646.35 KB)
implicit-0.6.1-cp39-cp39-manylinux2014_x86_64.whl(17.73 MB)
implicit-0.6.1-cp39-cp39-win_amd64.whl(630.76 KB)
v0.6.0(Jul 11, 2022)
Changes

🚀 Features

Add an alpha parameter for the ALS models @benfred (#581)

Use num_threads for batch recommendations in MF models @benfred (#562)

Add save/load functions for models @benfred (#552)

Publish binary wheels to PyPi using cibuildwheel @benfred (#546)

🐛 Bug Fixes

Fix default argument typo in benchmark_als.py @atakanfilgoz (#585)

Fix save/load for models that haven't been fit @benfred (#577)

Warn on non-csr input passed to model.fit @benfred (#578)

Fix partial_fit_items bug with only growing by a single item @benfred (#563)

🏎💨 Performance Improvements

Only call sort_indices when verifying negative samples @benfred (#583)

📄 Documentation

Update link to dataset in lastfm example script @benfred (#573)

🧰 Maintenance

Add release-drafter workflow for generating release notes @benfred (#579)

Disable building GPU extension if IMPLICIT_DISABLE_CUDA env var is set @benfred (#561)

Build binary wheels for apple silicon @benfred (#553)

Use scikit-build>=0.13.1 @benfred (#549)

Disable compute_86 compilation for cuda 11.0 @benfred (#536)

Require cmake v3.17+ @benfred (#534)

Source code(tar.gz)
Source code(zip)
implicit-0.6.0-cp310-cp310-macosx_10_9_universal2.whl(1.27 MB)
implicit-0.6.0-cp310-cp310-macosx_10_9_x86_64.whl(1011.36 KB)
implicit-0.6.0-cp310-cp310-macosx_11_0_arm64.whl(644.83 KB)
implicit-0.6.0-cp310-cp310-manylinux2014_x86_64.whl(17.71 MB)
implicit-0.6.0-cp310-cp310-win_amd64.whl(615.31 KB)
implicit-0.6.0-cp36-cp36m-macosx_10_9_x86_64.whl(984.68 KB)
implicit-0.6.0-cp36-cp36m-manylinux2014_x86_64.whl(17.70 MB)
implicit-0.6.0-cp36-cp36m-win_amd64.whl(615.22 KB)
implicit-0.6.0-cp37-cp37m-macosx_10_9_x86_64.whl(985.51 KB)
implicit-0.6.0-cp37-cp37m-manylinux2014_x86_64.whl(17.70 MB)
implicit-0.6.0-cp37-cp37m-win_amd64.whl(615.37 KB)
implicit-0.6.0-cp38-cp38-macosx_10_9_universal2.whl(1.24 MB)
implicit-0.6.0-cp38-cp38-macosx_10_9_x86_64.whl(991.96 KB)
implicit-0.6.0-cp38-cp38-macosx_11_0_arm64.whl(629.08 KB)
implicit-0.6.0-cp38-cp38-manylinux2014_x86_64.whl(17.72 MB)
implicit-0.6.0-cp38-cp38-win_amd64.whl(627.78 KB)
implicit-0.6.0-cp39-cp39-macosx_10_9_universal2.whl(1.27 MB)
implicit-0.6.0-cp39-cp39-macosx_10_9_x86_64.whl(1011.96 KB)
implicit-0.6.0-cp39-cp39-macosx_11_0_arm64.whl(643.60 KB)
implicit-0.6.0-cp39-cp39-manylinux2014_x86_64.whl(17.72 MB)
implicit-0.6.0-cp39-cp39-win_amd64.whl(627.83 KB)
v0.5.0(Jan 28, 2022)
v0.5.0

Breaking API Changes

The API for implicit has substantially changed in v0.5.0 - and any code written for the previous API will need to be rewritten:

Change model.fit to take a user_items sparse matrix #484

Return numpy arrays from recommend methods #482

Don't require empty rows in user_items and item_users parameters #526

Unify API for rank_items/recommend/recommend_all #489

Performance Improvements

Speedup evaluation by using batch recommend functions #520

Use FAISS for GPU inference #506

Multithreaded speedups for CPU models #517

Use thrust::binary_search to verify negative samples on GPU #524

Release GIL on GPU calls #528

New Features

Add incremental retraining support for ALS models #527

Add filtering for similar_items and similar_users #488

Add support for recalculate_users/items on the GPU #515

Add methods for converting MF models to/from gpu #521

Add a tutorial notebook for the lastfm example #529

Approximate nearest neighbour for BPR/LMF and GPU models #487

Dynamically detect CUDA availability #174

Source code(tar.gz)
Source code(zip)

Fast Python Collaborative Filtering for Implicit Feedback Datasets

Related tags

Overview

Implicit

Installation

Basic Usage

Articles about Implicit

Requirements

Benchmarks

Optimal Configuration

Comments

Literature:

This PR implements

correctness verification

Training Speed-Up

i found that this approach's performance gain vary across implementation.

Speedup along dimensions

Releases(v0.6.2)

v0.6.2(Dec 11, 2022)

Changes

🚀 Features

🐛 Bug Fixes

📄 Documentation

🧰 Maintenance

v0.6.1(Aug 31, 2022)

Changes

🚀 Features

🐛 Bug Fixes

📄 Documentation

v0.6.0(Jul 11, 2022)

Changes

🚀 Features

🐛 Bug Fixes

🏎💨 Performance Improvements

📄 Documentation

🧰 Maintenance

v0.5.0(Jan 28, 2022)

v0.5.0

Breaking API Changes

Performance Improvements

New Features

Owner

Ben Frederickson

Temporal Meta-path Guided Explainable Recommendation (WSDM2021)

The implementation of the submitted paper "Deep Multi-Behaviors Graph Network for Voucher Redemption Rate Prediction" in SIGKDD 2021 Applied Data Science Track.

An Efficient and Effective Framework for Session-based Social Recommendation

Bert4rec for news Recommendation

Spark-movie-lens - An on-line movie recommender using Spark, Python Flask, and the MovieLens dataset

Recommender systems are the systems that are designed to recommend things to the user based on many different factors

Incorporating User Micro-behaviors and Item Knowledge 59 60 3 into Multi-task Learning for Session-based Recommendation

Jointly Learning Explainable Rules for Recommendation with Knowledge Graph

This library intends to be a reference for recommendation engines in Python

Attentive Social Recommendation: Towards User And Item Diversities

Group-Buying Recommendation for Social E-Commerce

Persine is an automated tool to study and reverse-engineer algorithmic recommendation systems.

Mutual Fund Recommender System. Tailor for fund transactions.

Deep recommender models using PyTorch.

Price-aware Recommendation with Graph Convolutional Networks,

Knowledge-aware Coupled Graph Neural Network for Social Recommendation

Beyond Clicks: Modeling Multi-Relational Item Graph for Session-Based Target Behavior Prediction

Collaborative variational bandwidth auto-encoder (VBAE) for recommender systems.

A PyTorch implementation of "Say No to the Discrimination: Learning Fair Graph Neural Networks with Limited Sensitive Attribute Information" (WSDM 2021)