ML model visualizer

Polynomial Regression

Fit curved patterns with polynomial features and gradient descent.

TypeRegressionFeaturesPolynomialOptimizerGradient DescentOutputContinuous
Example PointsFormat: (x, y), (x, y). Use 3-30 points.
0.0010.2
1800

Visualization

Epoch 0 / 160
0.02.03.95.97.89.80.01.83.65.47.29.0XY
Data pointsPolynomial curveError

Ready to optimize polynomial coefficients with gradient descent.

Current polynomialy = 0.000 + 0.000z + 0.000z^2where z = (x - 4.5) / 4.5
Prediction for x = 60.000
StatusReady

Metrics

MSE25.636
R2 Score-6.635
Leading coeff0.000
c00.000c10.000c20.000

Loss Curve

25.61.870.1370.010EpochMSE0 / 160

Training Steps

  1. Choose polynomial degree d
  2. Create features [1, z, z^2, ..., z^d]
  3. Predict y_hat = c0 + c1 z + ... + cd z^d
  4. Compute MSE
  5. Update coefficients using gradients
  6. Return trained polynomial