Period detection is slightly off

```
x = np.random.uniform(0,100,10000)#linspace(0,100,1000)
y = np.sin(x/2)
yerr = np.ones_like(x) * 0.01
star = PIPS.photdata([x,y,yerr])
period,spectrum = star.amplitude_spectrum(p_min=0.1,p_max=20,N=1, Nterms=1,multiprocessing=False,repr_mode='chisq')

period[np.argmax(spectrum)]/ (np.pi*4)
```

returns 0.99463