In [1]:

    

from __future__ import print_function, division
import numpy as np
import matplotlib.pyplot as plt
from os.path import join as pjoin
import os.path as osp

%matplotlib inline


    

In [2]:

    

def plot_motion(data, figsize=(16,10)):
    #plots motion params for translations and rotations
    fig = plt.figure(figsize=figsize)
    #plot translations
    ax = fig.add_subplot(1,2,1)
    ax.plot(data[:,0:3])
    plt.title('translations')
    #plot rotations
    ax = fig.add_subplot(1,2,2)
    ax.plot(data[:,3:7])
    plt.title('rotations')
    plt.suptitle('motion parameters') #plot which trajectory
    plt.show()


    

In [3]:

    

REALIGN_PARAMS = osp.join(osp.abspath(osp.curdir), 'rp_afS46_sess2-0024-00004-000004-01.txt')
realign_params = np.loadtxt(REALIGN_PARAMS)
realign_params.shape


    

    
Out[3]:





(270, 6)

In [4]:

    

plot_motion(realign_params)


    

In [5]:

    

from scipy.interpolate import interp1d
x = np.arange(0, realign_params.shape[0])
y = realign_params[:,0]
f = interp1d(x, y, kind='cubic')

slice_nb = 30
xnew = np.linspace(0,269, 270*slice_nb)
print(xnew.shape, 270*slice_nb)
print(xnew.min(), xnew.max())

ynew = f(xnew)

print(ynew.shape)


    

    




(8100,) 8100
0.0 269.0
(8100,)

In [6]:

    

npts = 30
plt.plot(x[:npts], y[:npts], 'o', xnew[:npts*slice_nb], ynew[:npts*slice_nb], '-')


    

    
Out[6]:





[<matplotlib.lines.Line2D at 0x7f8f5c149dd0>,
 <matplotlib.lines.Line2D at 0x7f8f5c0d7090>]






    

In [7]:

    

def block_interp(data, block_size):
    """
    just duplicate data
    """
    
    return np.kron(data, np.ones((block_size, 1)))
    
def gradiant2degree(data):
    
    return data*180/np.pi


    

In [8]:

    

print(block_interp(gradiant2degree(realign_params[:3,:3]), 10))
print(gradiant2degree(realign_params[:3,:3]))


    

    




[[ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.          0.          0.        ]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [ 0.10527859  2.53858302 -0.05172263]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]
 [-1.27709812  1.52869168 -1.97458898]]
[[ 0.          0.          0.        ]
 [ 0.10527859  2.53858302 -0.05172263]
 [-1.27709812  1.52869168 -1.97458898]]

In [9]:

    

out_fname = osp.join(osp.abspath(osp.curdir), 'rp_afS46_sess2-0024-00004-000004-01' \
                                                                 + '-iblock' + '.txt')
new_data = np.hstack((realign_params[:,:3], gradiant2degree(realign_params[:,3:])))

new_data = block_interp(new_data, slice_nb)

plot_motion(new_data[:2000,:])

# np.savetxt(out_fname, new_data)


    

In [10]:

    

import scipy.signal as ssig

def convolve_2darr(arr, kernel, mode='same'):
    """
    """
    returned_arr = arr.T.copy()
    assert arr.ndim == 2
    
    for idx, sig in enumerate(arr.T):
        returned_arr[idx] = ssig.convolve(sig, kernel, mode=mode)
       
    return returned_arr.T


    

In [11]:

    

new_data.ndim


    

    
Out[11]:





2

In [12]:

    

kernel = ssig.gaussian(10, 1.5)/ssig.gaussian(10, 1.2).sum() 
data_convolved = convolve_2darr(new_data, kernel)


    

In [13]:

    

_  = plt.plot(data_convolved[:60,:])


    

In [14]:

    

plot_motion(data_convolved)
print(data_convolved.min(axis=0), data_convolved.max(axis=0))


    

    













    




[-0.33969208 -1.25546069 -0.61216052 -0.96756495 -0.54802919 -0.49631668] [ 0.59997055  0.6744129   0.32190506  0.27936705  0.19216374  0.24429156]

In [15]:

    

np.savetxt(out_fname, data_convolved)


    

In [17]:

    

plot_motion(np.loadtxt(out_fname))


    

In [16]:

    

plot_motion(np.loadtxt(out_fname)[:100,:])


    

In [18]:

    

20*30


    

    
Out[18]:





600

In [ ]: