0.5.1

.Rbuildignore

@@ -13,3 +13,4 @@ docs/
 ^Makefile$
 extradata/
 revdep/
+^CRAN-SUBMISSION$

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: rsparse
 Type: Package
 Title: Statistical Learning on Sparse Matrices
-Version: 0.5.0
+Version: 0.5.1
 Authors@R: c(
     person("Dmitriy", "Selivanov", role=c("aut", "cre", "cph"), email="[email protected]", 
     comment = c(ORCID = "0000-0001-5413-1506")),
@@ -57,5 +57,5 @@ Suggests:
 StagedInstall: TRUE
 URL: https://github.com/rexyai/rsparse
 BugReports: https://github.com/rexyai/rsparse/issues
-RoxygenNote: 7.1.1
+RoxygenNote: 7.2.1
 NeedsCompilation: yes

NEWS.md

@@ -1,21 +1,30 @@
-- 2021-10-17 - `v0.5.0`
-    - reworked non-negative matrix factorization with brand-new Coordinate Descent solver for OLS
-    - WRMF can model user, item and global biases
-    - various performance improvements
-- 2020-04-01 - `v0.4.0`
-    - updated docs with roxygen2 7.1
-    - added `ScaleNormalize` transformer
-    - added sparse*float S4 methods
-- 2019-07-17 - `v0.3.3.2`
-    - faster `find_top_product()` - avoid BLAS and openmp thread contention
-    - correctly identify openmp on OSX
-    - fixed issue with CRAN 'rcnst' check
-    - use `install_name_tool` hook in the `.onLoad()` - changes location of the `float.so` for CRAN binary installation - see #25
-- 2019-04-14 - `v0.3.3.1`
-    - fixed out of bound memory access as reported by CRAN UBSAN
-    - added ability to init GloVe embeddings with user provided values
-- 2019-03-16 - - `v0.3.3`
-    - added methods to natively slice CSR matrices without converting them to triplet/CSC
-- 2018-10-25
-    - add GloVe matrix factorization (adapted from `text2vec`)
-    - link to `float` package - credits to @snoweye and @wrathematics
+# rsparse 0.5.1 (2022-09-11)
+- update `configure` script, thanks to @david-cortes, see #73
+- minor fixes in WRMF
+- update docs with new roxygen2 to pass CRAN checks
+- update NEWS.md ro follow CRAN format
+
+# rsparse 0.5.0 (2021-10-17)
+- reworked non-negative matrix factorization with brand-new Coordinate Descent solver for OLS
+- WRMF can model user, item and global biases
+- various performance improvements
+
+# rsparse 0.4.0 (2020-04-01)
+- updated docs with roxygen2 7.1
+- added `ScaleNormalize` transformer
+- added sparse*float S4 methods
+
+# rsparse 0.3.3.2 (2019-07-17)
+- faster `find_top_product()` - avoid BLAS and openmp thread contention
+- correctly identify openmp on OSX
+- fixed issue with CRAN 'rcnst' check
+- use `install_name_tool` hook in the `.onLoad()` - changes location of the `float.so` for CRAN binary installation - see #25
+
+# rsparse 0.3.3.1 (2019-04-14)
+- fixed out of bound memory access as reported by CRAN UBSAN
+- added ability to init GloVe embeddings with user provided values
+
+# rsparse 0.3.3 (2019-03-16)
+- added methods to natively slice CSR matrices without converting them to triplet/CSC
+- add GloVe matrix factorization (adapted from `text2vec`)
+- link to `float` package - credits to @snoweye and @wrathematics

R/model_FTRL.R

@@ -147,7 +147,7 @@ FTRL = R6::R6Class(
     },
     #-----------------------------------------------------------------
     load = function(x) {
-      if (class(x) != "ftrl_model_dump")
+      if (!inherits(x, "ftrl_model_dump"))
         stop("input should be class of 'ftrl_model_dump' -  list of model parameters")
       private$init_model_param(learning_rate = x$learning_rate, learning_rate_decay = x$learning_rate_decay,
                                lambda = x$lambda, l1_ratio = x$l1_ratio,

README.md

@@ -22,7 +22,7 @@ Please reach us if you need **commercial support** - [[email protected]](mailto:hell
 
 ### Classification/Regression
 
-1. [Follow the proximally-regularized leader](https://www.jmlr.org/proceedings/papers/v15/mcmahan11b/mcmahan11b.pdf) which allows to solve **very large linear/logistic regression** problems with elastic-net penalty. Solver uses stochastic gradient descent with adaptive learning rates (so can be used for online learning - not necessary to load all data to RAM). See [Ad Click Prediction: a View from the Trenches](https://www.eecs.tufts.edu/~dsculley/papers/ad-click-prediction.pdf) for more examples.
+1. [Follow the proximally-regularized leader](http://proceedings.mlr.press/v15/mcmahan11b/mcmahan11b.pdf) which allows to solve **very large linear/logistic regression** problems with elastic-net penalty. Solver uses stochastic gradient descent with adaptive learning rates (so can be used for online learning - not necessary to load all data to RAM). See [Ad Click Prediction: a View from the Trenches](https://www.eecs.tufts.edu/~dsculley/papers/ad-click-prediction.pdf) for more examples.
     - Only logistic regerssion implemented at the moment
     - Native format for matrices is CSR - `Matrix::RsparseMatrix`. However common R `Matrix::CsparseMatrix` (`dgCMatrix`) will be converted automatically.
 1. [Factorization Machines](https://www.csie.ntu.edu.tw/~b97053/paper/Rendle2010FM.pdf) supervised learning algorithm which learns second order polynomial interactions in a factorized way. We provide highly optimized SIMD accelerated implementation.  
@@ -31,7 +31,7 @@ Please reach us if you need **commercial support** - [[email protected]](mailto:hell
 
 1. Vanilla **Maximum Margin Matrix Factorization** - classic approch for "rating" prediction. See `WRMF` class and constructor option `feedback = "explicit"`. Original paper which indroduced MMMF could be found [here](https://ttic.uchicago.edu/~nati/Publications/MMMFnips04.pdf).
     * <img src="https://raw.githubusercontent.com/rexyai/rsparse/master/docs/img/MMMF.png" width="400">
-1. **Weighted Regularized Matrix Factorization (WRMF)** from [Collaborative Filtering for Implicit Feedback Datasets](https://www.researchgate.net/profile/Yifan-Hu-25/publication/220765111_Collaborative_Filtering_for_Implicit_Feedback_Datasets/links/0912f509c579ddd954000000/Collaborative-Filtering-for-Implicit-Feedback-Datasets.pdf). See `WRMF` class and constructor option `feedback = "implicit"`. 
+1. **Weighted Regularized Matrix Factorization (WRMF)** from [Collaborative Filtering for Implicit Feedback Datasets](http://yifanhu.net/PUB/cf.pdf). See `WRMF` class and constructor option `feedback = "implicit"`. 
 We provide 2 solvers:
     1. Exact based on Cholesky Factorization
     1. Approximated based on fixed number of steps of **Conjugate Gradient**.

cran-comments.md

@@ -1,6 +1,8 @@
-# Resubmission
+# New submission
 
-- fixed pkg imports to not break dependencies
+- fixed HTML validation problems discovered by CRAN checks
+- fixed NEWS.md to follow CRAN format
+- failing tests on "r-oldrel-windows-ix86+x86_64" are due to the lack of `MatrixExtra` library. I'm not sure why this happened on CRAN server
 
 # Test environments
 
@@ -9,4 +11,4 @@
 
 # R CMD check results
 
-0 errors | 0 warning | 0 notes
+One misc note which shows 403 error when accessing [http://yifanhu.net/PUB/cf.pdf](http://yifanhu.net/PUB/cf.pdf). This seems CRAN server specific.