ApproxAtan (glimpse-full-assembly 4.0.0 API)

java.lang.Object
- com.metsci.glimpse.util.math.approx.ApproxAtan

```
public class ApproxAtan
extends Object
```
Similar to the classes in com.metsci.glimpse.util.math.fast, but uses linear interpolation between samples instead of nearest-neighbor.
Anecdotally, speed is about 7.5x faster than Math.atan(double). With 100k samples, max error is around 1e-10. Max error decreases as the number of samples increases.
Original domain is unbounded:
```
     a = y / x
     a ∊ [ -inf, +inf ]
 
```
Transform to bounded domain:
```
     b = a / ( 1.0 + abs( a ) )
     b ∊ [ -1, +1 ]
 
```
Invert the transform to compute initial samples:
```
     a = b / ( 1.0 - abs( b ) )
 
```

Constructor Summary

Constructors
Constructor Description

ApproxAtan(int numSamples)

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method	Description
`double`	`atan(double a)`
`double`	`atan2(double y, double x)`
`static int`	`rawSignBit(double d)`

Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail
- ApproxAtan
```
public ApproxAtan(int numSamples)
```

Method Detail

atan
```
public double atan(double a)
```

atan2

public double atan2(double y,
                    double x)

rawSignBit

public static int rawSignBit(double d)