Merge pull request #72 from kaybenleroll/nlp-workshop

nlp workshop

ws_nlp_202301/.gitignore

@@ -3,3 +3,6 @@
 output.log
 
 temp*.R
+
+*_cache/*
+*_files/*

ws_nlp_202301/Dockerfile

@@ -2,6 +2,20 @@ FROM rocker/verse:4.2.1
 
 ENV RETICULATE_MINICONDA_ENABLED=FALSE
 
+
+WORKDIR /tmp
+
+COPY build/test_report.qmd /tmp
+RUN quarto render test_report.qmd --to pdf \
+  && rm -fv /tmp/test_report.pdf
+
+
+RUN git clone https://github.com/lindenb/makefile2graph.git \
+  && cd makefile2graph \
+  && make \
+  && make install
+
+
 RUN apt-get update \
   && apt-get upgrade -y \
   && apt-get install -y --no-install-recommends \
@@ -46,12 +60,20 @@ RUN apt-get update \
     broom \
     cleanNLP \
     conflicted \
+    corpora \
+    corporaexplorer \
     corpus \
     cowplot \
     DataExplorer \
+    DT \
     directlabels \
     fs \
     furrr \
+    ggnetwork \
+    ggraph \
+    ggwordcloud \
+    hunspell \
+    ineq \
     kernlab \
     lda \
     lobstr \
@@ -66,13 +88,19 @@ RUN apt-get update \
     pryr \
     qdap \
     quanteda \
+    quanteda.textmodels \
+    quanteda.textplots \
+    quanteda.textstats \
     quarto \
+    readtext \
     rvest \
     sessioninfo \
+    sna \
     snakecase \
     spacyr \
     tesseract \
     text2vec \
+    textdata \
     textfeatures \
     textplot \
     tidygraph \
@@ -82,29 +110,17 @@ RUN apt-get update \
     topicmodels \
     udpipe \
     wordcloud \
+    word2vec \
     zipfR
 
 
-WORKDIR /tmp
-
-COPY build/test_report.qmd /tmp
-RUN quarto render test_report.qmd --to pdf \
-  && rm -fv /tmp/test_report.pdf
-
-
-RUN git clone https://github.com/lindenb/makefile2graph.git \
-  && cd makefile2graph \
-  && make \
-  && make install
-
 RUN cp -r $HOME/.R /home/rstudio \
   && chown -R rstudio:rstudio /home/rstudio/.R
 
 
 COPY build/docker_install_sys_rpkgs.R /tmp
 RUN Rscript /tmp/docker_install_sys_rpkgs.R
 
-
 WORKDIR /home/rstudio
 USER rstudio
 

ws_nlp_202301/Makefile

@@ -42,17 +42,22 @@ full_deps.dot:
 
 depgraph: full_deps.png
 
+initial_nlp_work.html: retrieve_extract_parse_film_scripts.html
+text_annotation_politeness.html: initial_nlp_work.html
 
 
 
-mrproper: clean-data clean-scraped clean-scripts
+
+
+mrproper: clean-html clean-data clean-scraped clean-scripts
 	rm -fv *.dot
 	rm -fv output.log
 
-
+clean-html:
+	rm -fv ${HTML_FILES}
 
 clean-data:
-	rm -rfv data/*.rds
+	rm -fv data/*.rds
 
 clean-parsed: clean-data
 	rm -fv data/parsed_scripts/*.rds

ws_nlp_202301/build/docker_install_sys_rpkgs.R

@@ -4,6 +4,18 @@ remotes::install_github(
   upgrade = "never"
 )
 
+remotes::install_github(
+  "quanteda/quanteda.corpora",
+  ref     = "ec4b76d841afc9a734cc0351b1fa87236c83b456",
+  upgrade = "never"
+)
+
+remotes::install_github(
+  "kbenoit/quanteda.dictionaries",
+  ref     = "b3c91606afad56603915fd622ef0aba4cc95135f",
+  upgrade = "never"
+)
+
 install.packages("StanfordCoreNLP",
   dependencies = TRUE,
   repos        = "http://datacube.wu.ac.at/"

ws_nlp_202301/build/docker_install_user_config.R

@@ -11,3 +11,4 @@ cnlp_download_corenlp(lang = "en")
 cnlp_init_corenlp(lang = "en")
 
 cnlp_init_udpipe(model_name = "english")
+

ws_nlp_202301/build/test_report.qmd

@@ -1,63 +1,12 @@
 ---
 title: "Temporary Quarto Document"
-author: "Mick Cooney <mcooney@describedata.com>"
+author: "Mick Cooney <mickcooney@gmail.com>"
 date: "`r Sys.Date()`"
 ---
 
 
 ```{r knit_opts, include = FALSE}
-library(conflicted)
 library(tidyverse)
-library(magrittr)
-library(rlang)
-library(scales)
-library(cowplot)
-
-resolve_conflicts <- function(pkg_priority) {
-  get_index <- function(pkg_name) {
-    idx <- str_which(pkg_priority, pkg_name)
-
-    if(length(idx) == 0) {
-      idx <- 0L
-    }
-
-    return(idx)
-  }
-
-  conflict_lst <- conflict_scout()
-
-  for(func_name in names(conflict_lst)) {
-    pkg_index <- map_int(conflict_lst[[func_name]], get_index)
-
-    pkg_index <- pkg_index[pkg_index > 0]
-
-    if(length(pkg_index) == 0) {
-      pkg_use <- conflict_lst[[func_name]][1]
-    } else {
-      pkg_use <- pkg_index %>%
-        min() %>%
-        pkg_priority[.]
-    }
-
-    conflict_prefer(func_name, pkg_use)
-  }
-
-  return(conflict_lst)
-}
-
-conflict_lst <- resolve_conflicts(
-  c("xml2", "magrittr", "rlang", "dplyr", "readr", "purrr", "ggplot2")
-  )
-
-
-knitr::opts_chunk$set(
-  tidy       = FALSE,
-  cache      = FALSE,
-  warning    = FALSE,
-  message    = FALSE,
-  fig.height =     8,
-  fig.width  =    11
-  )
 
 options(
   width = 80L,
@@ -66,6 +15,4 @@ options(
   )
 
 set.seed(42)
-
-theme_set(theme_cowplot())
 ```