Upload required 4 files

Dockerfile

@@ -0,0 +1,34 @@
+# Use a base image that includes both Python and R. rocker/r-ver is a popular choice for R, with an R base and Python installed.
+FROM rocker/r-ver:4.1.0
+
+# Install Python, pip, and system dependencies for your Python packages, e.g., for matplotlib.
+# Dependencies being installed:
+# - libfreetype6-dev, libxft-dev for matplotlib
+# - libpcre2-dev for -lpcre2-8, liblzma-dev for -llzma, libbz2-dev for -lbz2, libicu-dev for -licuuc and -licui18n for rpy2 compilation requirements
+RUN apt-get update && apt-get install -y \
+    python3-pip \
+    python3-dev \
+    libfreetype6-dev \
+    libxft-dev \
+    libpcre2-dev \
+    liblzma-dev \
+    libbz2-dev \
+    libicu-dev \
+ && rm -rf /var/lib/apt/lists/*
+
+# Install R packages
+RUN R -e "install.packages(c('relaimpo', 'readxl'), repos='http://cran.rstudio.com/')"
+
+# Copy the requirements.txt file, your app script, and the R script into the container.
+COPY requirements.txt /requirements.txt
+COPY app.py /app.py
+COPY process_data.R /process_data.R  
+
+# Install Python dependencies from requirements.txt.
+RUN pip3 install --no-cache-dir -r /requirements.txt
+
+# Expose the port Gradio runs on.
+EXPOSE 7860
+
+# Command to run your app.
+CMD ["python3", "/app.py"]

app.py

@@ -0,0 +1,107 @@
+# Import libraries
+import subprocess
+import pandas as pd
+import matplotlib.pyplot as plt
+import seaborn as sns
+import gradio as gr
+import tempfile
+import logging
+from PIL import Image
+import io
+
+logging.basicConfig(level=logging.INFO)
+
+
+def analyze_excel(file_path):
+    # Temporary file for the textual output
+    with tempfile.NamedTemporaryFile(
+        delete=True, suffix=".txt", mode="r+"
+    ) as text_output_temp:
+        text_output_path = text_output_temp.name
+
+    # Temporary file for the CSV data
+    with tempfile.NamedTemporaryFile(
+        delete=True, suffix=".csv", mode="r+"
+    ) as csv_output_temp:
+        csv_output_path = csv_output_temp.name
+
+    # Call the R script, passing paths for both outputs
+    command = [
+        "Rscript",
+        "process_data.R",
+        file_path,
+        text_output_path,
+        csv_output_path,
+    ]
+
+    try:
+        subprocess.run(command, check=True)
+    except subprocess.CalledProcessError as e:
+        logging.error("R script failed with error: %s", e)
+        return None, "Error executing R script. Please check the input file format."
+
+    # Read the detailed textual output
+    with open(text_output_path, "r") as file:
+        calc_relaimpo_text = file.read()
+
+    # Read the CSV data into a DataFrame for plotting
+    results_df = pd.read_csv(csv_output_path)
+
+    # Create a new column for plotting with values Importance * 100
+    results_df["Importance_percent"] = results_df["Importance"] * 100
+
+    # Plotting
+    fig, ax = plt.subplots(figsize=(12, 10))
+    sns.barplot(
+        x="Importance_percent", y="Predictor", data=results_df, color="skyblue", ax=ax
+    )
+
+    # Add percentage annotations to the bars
+    for index, value in enumerate(results_df["Importance_percent"]):
+        ax.text(value + 1, index, f"{value:.2f}%", va="center", fontsize=8)
+
+    plt.xlabel("Relative Importance (%)")
+    plt.ylabel("Predictors")
+    plt.title("Relative Importance Ranking of Predictors (LMG)")
+
+    # Adjust the x-axis limit to create space for annotations
+    plt.xlim(0, (results_df["Importance"].max() * 100) + 10)
+
+    plt.tight_layout()
+
+    # Convert plot to image data
+    img_data = io.BytesIO()
+    plt.savefig(img_data, format="png")
+    img_data.seek(0)
+    img = Image.open(img_data)
+
+    plt.close(fig)  # Close the figure to free memory
+
+    # Return both the image and the detailed textual output
+    return img, calc_relaimpo_text
+
+
+# Define instruction text and the Gradio app interface as before
+instruction_text = """
+## Instructions
+
+Please upload an Excel file with your data. The file should contain columns for Stability, Development, Relationship, Benefit, Vision, and Competence, along with Trust as the target variable.
+
+The app will analyze the data, calculate the relative importance of predictors using Shapley values, and display the results.
+
+**Note:** The analysis may take a few seconds. Please wait for the results to be displayed.
+"""
+
+iface = gr.Interface(
+    fn=analyze_excel,
+    inputs=["file"],
+    outputs=[
+        gr.Image(type="pil", label="Shapley Regression Plot"),
+        gr.Textbox(label="Calculation Output"),
+    ],
+    title="TrustLogic: Trust Driver Analysis",
+    description=instruction_text,
+    allow_flagging="never",
+)
+
+iface.launch(server_name="0.0.0.0")

process_data.R

@@ -0,0 +1,26 @@
+library(readxl)
+library(relaimpo)
+
+args <- commandArgs(trailingOnly = TRUE)
+input_file <- args[1]
+output_text_file <- args[2]
+output_csv_file <- args[3]  
+
+# Read the data
+data <- read_excel(input_file)
+
+# Fit the model
+model <- lm(Trust ~ Stability + Development + Relationship + Benefit + Vision + Competence, data = data)
+
+# Calculate relative importance
+calc_relaimpo <- calc.relimp(model, type = "lmg", rela = TRUE)
+
+# Write the detailed output to a text file
+full_output <- capture.output(print(calc_relaimpo))
+writeLines(full_output, output_text_file)
+
+# Extract relevant data from calc_relaimpo for plotting
+results <- data.frame(Predictor = names(calc_relaimpo$lmg), Importance = calc_relaimpo$lmg)
+
+# Save the plotting data to a CSV file
+write.csv(results, file = output_csv_file, row.names = FALSE)

requirements.txt

@@ -0,0 +1,6 @@
+pandas
+matplotlib
+seaborn
+gradio
+Pillow  
+openpyxl