|
|
import gradio as gr |
|
|
import pandas as pd |
|
|
import matplotlib.pyplot as plt |
|
|
import io |
|
|
import json |
|
|
from PIL import Image, ImageDraw |
|
|
import google.generativeai as genai |
|
|
import traceback |
|
|
|
|
|
def process_file(file, instructions, api_key): |
|
|
try: |
|
|
|
|
|
genai.configure(api_key=api_key) |
|
|
model = genai.GenerativeModel('gemini-2.5-pro-preview-03-25') |
|
|
|
|
|
|
|
|
file_path = file.name |
|
|
df = pd.read_csv(file_path) if file_path.endswith('.csv') else pd.read_excel(file_path) |
|
|
|
|
|
|
|
|
prompt = f""" |
|
|
Analyze the following dataset and instructions: |
|
|
|
|
|
Data columns: {list(df.columns)} |
|
|
Instructions: {instructions} |
|
|
|
|
|
Based on this, create 3 appropriate visualizations. For each visualization, provide: |
|
|
1. A title |
|
|
2. The most suitable plot type (choose from: bar, line, scatter, hist) |
|
|
3. The column to use for the x-axis |
|
|
4. The column to use for the y-axis (use None for histograms) |
|
|
5. A brief explanation of why this visualization is appropriate |
|
|
|
|
|
Return your response as a JSON string in this format: |
|
|
[ |
|
|
{{"title": "...", "plot_type": "...", "x": "...", "y": "...", "explanation": "..."}}, |
|
|
{{"title": "...", "plot_type": "...", "x": "...", "y": "...", "explanation": "..."}}, |
|
|
{{"title": "...", "plot_type": "...", "x": "...", "y": "...", "explanation": "..."}} |
|
|
] |
|
|
""" |
|
|
|
|
|
response = model.generate_content(prompt) |
|
|
plots = json.loads(response.text) |
|
|
|
|
|
|
|
|
images = [] |
|
|
for plot in plots: |
|
|
fig, ax = plt.subplots(figsize=(10, 6)) |
|
|
title, plot_type, x, y = plot['title'], plot['plot_type'], plot['x'], plot['y'] |
|
|
|
|
|
if plot_type == 'bar': |
|
|
df.plot(kind='bar', x=x, y=y, ax=ax) |
|
|
elif plot_type == 'line': |
|
|
df.plot(kind='line', x=x, y=y, ax=ax) |
|
|
elif plot_type == 'scatter': |
|
|
df.plot(kind='scatter', x=x, y=y, ax=ax) |
|
|
elif plot_type == 'hist': |
|
|
df[x].hist(ax=ax) |
|
|
|
|
|
ax.set_title(title) |
|
|
ax.set_xlabel(x) |
|
|
ax.set_ylabel(y if y else 'Frequency') |
|
|
plt.tight_layout() |
|
|
|
|
|
buf = io.BytesIO() |
|
|
plt.savefig(buf, format='png') |
|
|
buf.seek(0) |
|
|
img = Image.open(buf) |
|
|
images.append((img, plot['explanation'])) |
|
|
plt.close(fig) |
|
|
|
|
|
return images if len(images) == 3 else images + [(Image.new('RGB', (800, 600), (255,255,255)), "")]*(3-len(images)) |
|
|
|
|
|
except Exception as e: |
|
|
error_message = f"Error: {str(e)}\n\nTraceback:\n{traceback.format_exc()}" |
|
|
print(error_message) |
|
|
error_image = Image.new('RGB', (800, 400), (255, 255, 255)) |
|
|
draw = ImageDraw.Draw(error_image) |
|
|
draw.text((10, 10), error_message, fill=(255, 0, 0)) |
|
|
return [(error_image, "An error occurred")] * 3 |
|
|
|
|
|
with gr.Blocks(theme=gr.themes.Default()) as demo: |
|
|
gr.Markdown("# Data Analysis Dashboard") |
|
|
|
|
|
with gr.Row(): |
|
|
file = gr.File(label="Upload Dataset", file_types=[".csv", ".xlsx"]) |
|
|
instructions = gr.Textbox(label="Analysis Instructions", placeholder="Describe the analysis you want...") |
|
|
|
|
|
api_key = gr.Textbox(label="Gemini API Key", type="password") |
|
|
submit = gr.Button("Generate Insights", variant="primary") |
|
|
|
|
|
output_images = [gr.Image(label=f"Visualization {i+1}") for i in range(3)] |
|
|
output_texts = [gr.Textbox(label=f"Explanation {i+1}") for i in range(3)] |
|
|
|
|
|
submit.click( |
|
|
process_file, |
|
|
inputs=[file, instructions, api_key], |
|
|
outputs=output_images + output_texts |
|
|
) |
|
|
|
|
|
if __name__ == "__main__": |
|
|
demo.launch() |
