Update app.R

app.R

@@ -1,58 +1,161 @@
+# app.R: Shiny App for Generating Organizational Embeddings using LinkOrgs
+
+# Required libraries
 library(shiny)
+library(LinkOrgs)
+library(DT)
 library(bslib)
-library(dplyr)
-library(ggplot2)
+library(shinyWidgets)
 
-df <- readr::read_csv("penguins.csv")
-# Find subset of columns that are suitable for scatter plot
-df_num <- df |> select(where(is.numeric), -Year)
+# Startup code
+options(shiny.maxRequestSize = 30 * 1024^2)
 
+# UI definition
 ui <- page_sidebar(
-  theme = bs_theme(bootswatch = "minty"),
-  title = "Penguins explorer",
+  theme = bs_theme(bootswatch = "flatly"),
+  title = "OrgEmbed: Generate Organizational Embeddings",
   sidebar = sidebar(
-    varSelectInput("xvar", "X variable", df_num, selected = "Bill Length (mm)"),
-    varSelectInput("yvar", "Y variable", df_num, selected = "Bill Depth (mm)"),
-    checkboxGroupInput("species", "Filter by species",
-      choices = unique(df$Species), selected = unique(df$Species)
+    radioButtons("input_mode", "Input Method:", choices = c("CSV Upload" = "csv", "Paste Text" = "text")),
+    conditionalPanel(
+      condition = "input.input_mode == 'csv'",
+      fileInput("file", "Upload CSV", accept = ".csv"),
+      selectInput("col_select", "Names Column:", choices = NULL)
+    ),
+    conditionalPanel(
+      condition = "input.input_mode == 'text'",
+      textAreaInput("text_names", "Paste Names (one per line)", rows = 5)
     ),
-    hr(), # Add a horizontal rule
-    checkboxInput("by_species", "Show species", TRUE),
-    checkboxInput("show_margins", "Show marginal plots", TRUE),
-    checkboxInput("smooth", "Add smoother"),
+    numericInput("max_rows", "Max Rows to Process", value = 5000, min = 1),
+    selectInput("ml_version", "ML Version:", choices = c("v1", "v2", "v3", "v4"), selected = "v4"),
+    checkboxInput("include_original", "Include Original Names in Output", value = TRUE),
+    actionButton("process", "Generate Embeddings", class = "btn-primary"),
+    bs_tooltip(id = "text_names", title = "Names should be one per row, e.g., 'Apple Inc.'"),
+    actionButton("help_btn", "Help")
   ),
-  plotOutput("scatter")
+  mainPanel(
+    DTOutput("preview"),
+    uiOutput("summary_card"),
+    downloadButton("download", "Download Embeddings CSV", class = "btn-success"),
+    uiOutput("progress_ui")
+  )
 )
 
+# Server logic
 server <- function(input, output, session) {
-  subsetted <- reactive({
-    req(input$species)
-    df |> filter(Species %in% input$species)
+  
+  # On startup, initialize backend
+  observe({
+    tryCatch({
+      BuildBackend(conda = "auto")
+      showNotification("Backend initialized successfully.")
+    }, error = function(e) {
+      showNotification("Backend setup failed—check internet or conda path.", type = "error")
+    })
   })
-
-  output$scatter <- renderPlot(
-    {
-      p <- ggplot(subsetted(), aes(!!input$xvar, !!input$yvar)) +
-        theme_light() +
-        list(
-          theme(legend.position = "bottom"),
-          if (input$by_species) aes(color = Species),
-          geom_point(),
-          if (input$smooth) geom_smooth()
-        )
-
-      if (input$show_margins) {
-        margin_type <- if (input$by_species) "density" else "histogram"
-        p <- p |> ggExtra::ggMarginal(
-          type = margin_type, margins = "both",
-          size = 8, groupColour = input$by_species, groupFill = input$by_species
-        )
+  
+  # Reactive: Parse user input
+  input_df <- reactive({
+    req(input$process)
+    if (input$input_mode == "csv") {
+      req(input$file)
+      df <- read.csv(input$file$datapath, stringsAsFactors = FALSE)
+      updateSelectInput(session, "col_select", choices = colnames(df))
+      req(input$col_select)
+      if (!input$col_select %in% colnames(df) || nrow(df) == 0) {
+        showNotification("Invalid names column—please select again.", type = "error")
+        return(NULL)
       }
-
-      p
-    },
-    res = 100
+      data.frame(names = df[[input$col_select]])
+    } else {
+      req(input$text_names != "")
+      names_list <- trimws(strsplit(input$text_names, "\n")[[1]])
+      names_list <- names_list[names_list != ""]
+      data.frame(names = names_list)
+    }
+  })
+  
+  # Preview table
+  output$preview <- renderDT({
+    req(input_df())
+    datatable(head(input_df(), 10), options = list(pageLength = 10))
+  })
+  
+  # Reactive: Embeddings result
+  embeddings <- reactiveVal(NULL)
+  
+  # Trigger embedding generation
+  observeEvent(input$process, {
+    df <- input_df()
+    if (is.null(df) || nrow(df) == 0) {
+      showNotification("Please provide at least one valid name.", type = "error")
+      return()
+    }
+    if (nrow(df) > input$max_rows) {
+      showModal(modalDialog("Large dataset detected—processing may take time. Proceed?", 
+                            footer = tagList(modalButton("Cancel"), actionButton("proceed_large", "Proceed"))))
+      return()
+    }
+    process_embeddings()
+  })
+  
+  # Handle large input confirmation
+  observeEvent(input$proceed_large, {
+    removeModal()
+    process_embeddings()
+  })
+  
+  # Function to process embeddings with progress
+  process_embeddings <- function() {
+    withProgress(message = "Generating Embeddings", value = 0, {
+      incProgress(0.1, detail = "Parsing input...")
+      df <- head(input_df(), input$max_rows)
+      showNotification(sprintf("Processing %d names...", nrow(df)))
+      
+      incProgress(0.2, detail = "Initializing model...")
+      tryCatch({
+        rep_x <- LinkOrgs(
+          x = df, y = NULL, by.x = "names",
+          algorithm = "ml", ml_version = input$ml_version,
+          ExportEmbeddingsOnly = TRUE
+        )
+        embed_df <- rep_x$embedx
+        if (!input$include_original) embed_df <- embed_df[, -1]
+        embeddings(embed_df)
+        incProgress(0.7, detail = "Generating embeddings...")
+      }, error = function(e) {
+        showNotification(sprintf("Error: %s", e$message), type = "error")
+        embeddings(NULL)
+      })
+      incProgress(1, detail = "Complete!")
+    })
+  }
+  
+  # Summary card
+  output$summary_card <- renderUI({
+    req(embeddings())
+    card(
+      sprintf("Embeddings generated for %d names. Dimensions: %d.", 
+              nrow(embeddings()), ncol(embeddings()))
+    )
+  })
+  
+  # Download handler
+  output$download <- downloadHandler(
+    filename = "org_embeddings.csv",
+    content = function(file) {
+      req(embeddings())
+      write.csv(embeddings(), file, row.names = FALSE)
+    }
   )
+  
+  # Help modal
+  observeEvent(input$help_btn, {
+    showModal(modalDialog(
+      title = "Usage Guide",
+      "1. Upload/paste names.\n2. Click Process.\n3. Download embeddings for use in linkage models."
+    ))
+  })
 }
 
-shinyApp(ui, server)
+# Run the app
+shinyApp(ui = ui, server = server)