Muffin's Data Visualizer

Upload a CSV or Excel file (or paste data), pick your columns, and get a chart instantly. Great for ML training curves, experiment logs, and quick data exploration.

// data input

📊

Upload CSV or Excel

or use paste below

// axes

X Axis column

Y Axis columns toggle all

— load data first —

// chart type

📈 Line

📊 Bar

⊹ Scatter

▦ Histogram

// options

Chart title

Highlight best value

Show markers on line

Chart Preview

◈

Load your data and click Plot Chart

or try a sample dataset below ↓

// sample datasets — click to load & plot

📈 ML Training Curve (loss + accuracy)

📊 Monthly Revenue & Expenses

⊹ Scatter: Height vs Weight

▦ Score Distribution Histogram

🌡 Temperature Over Time

⚡ Model Benchmark Comparison

🤖 5 Models F1 Score (multi-line)

What is Muffin's Data Visualizer?

Muffin's Data Visualizer is a free, browser-based tool that turns your CSV or Excel data into charts instantly — no code, no Python, no Matplotlib, no installations. Upload a file or paste your data, select your X and Y columns, choose a chart type, and click Plot. The chart renders immediately in your browser. Your data never leaves your device.

What data formats does it accept?

The visualizer accepts any flat CSV or Excel (.xlsx) file where the first row is a header row and each column has a name. Each column becomes one data series you can plot. The simplest format looks like this:

epoch, train_loss, val_loss, accuracy
1, 1.42, 1.51, 0.61
2, 1.18, 1.29, 0.68
3, 0.97, 1.10, 0.73
...

Note: multi-row headers (merged cells in Excel where one header spans multiple columns) are not supported. Flatten them to a single header row first before uploading.

Supported chart types

📈 Line Chart

Best for trends over time — training loss curves, accuracy over epochs, temperature over days, stock prices. Supports multiple lines for comparing models.

📊 Bar Chart

Best for comparing values across categories — model accuracy comparison, monthly revenue by category, benchmark results across different systems.

⊹ Scatter Plot

Best for correlations between two variables — height vs weight, inference speed vs accuracy, training data size vs model performance.

▦ Histogram

Best for understanding data distribution — score distributions, prediction confidence distribution, error magnitude distribution across a dataset.

Plot multiple models on the same chart

The Y axis uses checkboxes — tick as many columns as you want and they all appear as separate colored lines on the same chart, each with a different neon color and a legend entry. This is useful for comparing training curves of multiple models, plotting both loss and accuracy together, or visualizing several metrics side by side.

epoch, BERT_f1, RoBERTa_f1, DistilBERT_f1, XLNet_f1
1, 0.59, 0.61, 0.56, 0.58
2, 0.65, 0.68, 0.62, 0.66
...

Set X = epoch, check all 4 model columns — you get 4 colored lines with a legend, each model clearly distinguishable. Click the toggle all link next to the Y axis label to select or deselect all columns at once.

How to plot your ML training curve

If you log your training metrics to a CSV file — loss, accuracy, F1, BLEU, or any other metric per epoch — you can visualize them here in seconds. Upload your CSV, set X = epoch, check the metrics you want to see, select Line chart, and click Plot. Enable Highlight best value to automatically mark the epoch where your model peaked. Download the chart as a 1200×700px PNG ready to drop into a paper or report.

100% private — your data never leaves your browser

Research data, experiment logs, and model benchmarks are often confidential before publication. Muffin processes your CSV and Excel files entirely in your browser using PapaParse and SheetJS for parsing and Plotly.js for rendering. Nothing is sent to any server. You can disconnect from the internet after the page loads and the visualizer will still work.