Curve Fit: docs draft

Author: VesnaT

doc/data-mining-library/source/reference/regression.rst

@@ -148,3 +148,12 @@ Gradient Boosted Trees
 
 .. autoclass:: XGBRFRegressor
    :members:
+
+
+Curve Fit
+----------------------
+
+.. automodule:: Orange.regression.curvefit
+
+.. autoclass:: CurveFitLearner
+   :members:

doc/visual-programming/source/index.rst

@@ -109,6 +109,7 @@ Model
    widgets/model/logisticregression
    widgets/model/naivebayes
    widgets/model/adaboost
+   widgets/model/curvefit
    widgets/model/neuralnetwork
    widgets/model/stochasticgradient
    widgets/model/stacking

doc/visual-programming/source/widgets/model/curvefit.md

@@ -0,0 +1,18 @@
+Curve Fit
+=========
+
+Fit a function to data.
+
+**Inputs**
+
+- Data: input dataset
+- Preprocessor: preprocessing method(s)
+
+**Outputs**
+
+- Learner: curve fit learning algorithm
+- Model: trained model
+- Coefficients: fitted coefficients
+
+The **Curve Fit** widget fits an arbitrary function to the input data. It only works for regression tasks.
+The widget uses [scipy.curve_fit](https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.curve_fit.html) to find the optimal values of the parameters.

doc/widgets.json

@@ -631,6 +631,15 @@
      "boost"
     ]
    },
+   {
+    "text": "Curve Fit",
+    "doc": "visual-programming/source/widgets/model/curvefit.md",
+    "icon": "../Orange/widgets/model/icons/CurveFit.svg",
+    "background": "#FAC1D9",
+    "keywords": [
+     "function"
+    ]
+   },
    {
     "text": "Neural Network",
     "doc": "visual-programming/source/widgets/model/neuralnetwork.md",