From 05238ae30e361e0ed391b467a28c2c467050c4f0 Mon Sep 17 00:00:00 2001
From: Ricardo Wurmus <rekado@elephly.net>
Date: Tue, 29 Jun 2021 12:27:54 +0200
Subject: [PATCH] gnu: Add r-text2vec.

* gnu/packages/cran.scm (r-text2vec): New variable.
---
 gnu/packages/cran.scm | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/gnu/packages/cran.scm b/gnu/packages/cran.scm
index fb62558b57..0ad05cf0df 100644
--- a/gnu/packages/cran.scm
+++ b/gnu/packages/cran.scm
@@ -17896,6 +17896,42 @@ (define-public r-rsparse
 Row} (CSR) format.")
     (license license:gpl2+)))
 
+(define-public r-text2vec
+  (package
+    (name "r-text2vec")
+    (version "0.6")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "text2vec" version))
+       (sha256
+        (base32
+         "0r75cv77x2zm1z66s95hic71dpbqmybz39n48q6mz7gfd3m7312y"))))
+    (properties `((upstream-name . "text2vec")))
+    (build-system r-build-system)
+    (propagated-inputs
+     `(("r-data-table" ,r-data-table)
+       ("r-digest" ,r-digest)
+       ("r-lgr" ,r-lgr)
+       ("r-matrix" ,r-matrix)
+       ("r-mlapi" ,r-mlapi)
+       ("r-r6" ,r-r6)
+       ("r-rcpp" ,r-rcpp)
+       ("r-rsparse" ,r-rsparse)
+       ("r-stringi" ,r-stringi)))
+    (native-inputs
+     `(("r-knitr" ,r-knitr)))
+    (home-page "http://text2vec.org")
+    (synopsis "Text mining framework for R")
+    (description
+     "This package provides fast and memory-friendly tools for text
+vectorization, topic modeling (LDA, LSA), word embeddings (GloVe),
+similarities.  It provides a source-agnostic streaming API, which allows
+researchers to perform analysis of collections of documents which are larger
+than available RAM.  All core functions are parallelized to benefit from
+multicore machines.")
+    (license license:gpl2+)))
+
 (define-public r-xmlparsedata
   (package
     (name "r-xmlparsedata")