def filt(fname_in,fname_out,two_way,f_hi=None,f_lo=None):
print "Input file: %s"%fname_in
print "Output file: %s"%fname_out
n_ch,sample_rate = get_dat_pars(fname_in)
n_ch,_ = get_dat_pars(fname_in)
in_memmap = np.memmap(fname_in,mode="r",dtype=DT_IN).reshape(-1,n_ch)
n_s = in_memmap.shape[0]
out_memmap = np.memmap(fname_out,mode="w+",shape=(n_s,n_ch),dtype=DT_OUT)
if f_lo is None: b,a = butter(BUTTER_ORD,f_hi/(sample_rate/2),"high")
elif f_hi is None: b,a = butter(BUTTER_ORD,f_hi/(sample_rate/2),"low")
else: b,a = butter(BUTTER_ORD,(f_lo/(sample_rate/2),f_hi/(sample_rate/2)),"pass")
if two_way: filter_2way(b,a,in_memmap, out_memmap)
else: filter_1way(b,a,in_memmap,out_memmap)
def downsample(fname_in,fname_out,two_way, ratio = 16):
print "Input file: %s"%fname_in
print "Output file: %s"%fname_out
n_ch,_ = get_dat_pars(fname_in)
in_memmap = np.memmap(fname_in,mode="r",dtype=DT_IN).reshape(-1,n_ch)
n_s = in_memmap.shape[0]
out_memmap = np.memmap(fname_out,mode="w+",shape=(n_s//ratio,n_ch),dtype=DT_OUT)
if two_way: downsample_2way(in_memmap, out_memmap,ratio=16)
else: downsample_1way(in_memmap,out_memmap,ratio=16)
def check_crosstalk(DatFileName,ProbeFileName=None,output_dir = None):
"Makes a colormap that shows conditional correlations between channels. Output it as a png file."
N_SAMPLES = 80000
if not os.path.exists(DatFileName):
raise Exception("Dat file %s does not exist"%DatFileName)
DatFileName = os.path.abspath(DatFileName)
DatDir = os.path.abspath(os.path.dirname(DatFileName))
n_ch_dat,sample_rate = get_dat_pars(DatFileName)
ProbeFileName = ProbeFileName or find_file_with_ext(DatDir,'probe',ex_if_not_found=False)
if ProbeFileName is not None:
probe_stuff.load_probe(ProbeFileName)
Channels_dat = [site.dat for site in probe_stuff.PROBE_SITES]
ChNames = [site.name for site in probe_stuff.PROBE_SITES]
ChannelGraph = probe_stuff.PROBE_GRAPH
else:
Channels_dat = range(n_ch_dat)
ChNames = [str(i) for i in range(n_ch_dat)]
b,a = butter(3,(600./(sample_rate/2),.95),'pass')
X_nc = np.fromfile(DatFileName,dtype=np.int16,count=N_SAMPLES*n_ch_dat).reshape(-1,n_ch_dat)[:,Channels_dat]
X_nc = lfilter(b,a,X_nc,axis=0)
Cov_cc = np.cov(X_nc.T)
try: Con_cc = np.linalg.inv(Cov_cc)
except np.linalg.LinAlgError:
vals,vec_mat = np.linalg.eigh(Cov_cc)
print "Singular covariance matrix. Small eigvals/eigvecs:"
np.set_printoptions(precision=2,suppress=True)
for _val,vec in zip(filter(lambda x: x < TINY, vals),vec_mat.T):
print vec
exit()
n_ch = len(Channels_dat)
plt.pcolor(np.ma.array(np.abs(Con_cc),mask=np.eye(Con_cc.shape[0]))
,cmap=plt.cm.cool)
#plt.colorbar()
if ProbeFileName is not None and len(edges(ChannelGraph))>0:
ptsx1,ptsy1 = np.array(edges(ChannelGraph)).T
plt.plot(ptsx1+.5,ptsy1+.5,'kx',ms=10,mew=3)
ax = plt.gca()
ax.set_xticks(np.arange(n_ch)+.5)
ax.set_yticks(np.arange(n_ch)+.5)
ax.set_xticklabels(ChNames,rotation='vertical')
ax.set_yticklabels(ChNames)
basename = basename_noext(ProbeFileName) if ProbeFileName else basename_noext(DatFileName)
outpath = os.path.join(output_dir or ProbeFileName and os.path.dirname(ProbeFileName) or DatDir,"%s_crosstalk.png"%basename)
print("saving figure in %s"%outpath)
plt.savefig(outpath)
def downsample(fname_in, fname_out, two_way, ratio=16):
print "Input file: %s" % fname_in
print "Output file: %s" % fname_out
n_ch, _ = get_dat_pars(fname_in)
in_memmap = np.memmap(fname_in, mode="r", dtype=DT_IN).reshape(-1, n_ch)
n_s = in_memmap.shape[0]
out_memmap = np.memmap(fname_out,
mode="w+",
shape=(n_s // ratio, n_ch),
dtype=DT_OUT)
if two_way: downsample_2way(in_memmap, out_memmap, ratio=16)
else: downsample_1way(in_memmap, out_memmap, ratio=16)
def filt(fname_in, fname_out, two_way, f_hi=None, f_lo=None):
print "Input file: %s" % fname_in
print "Output file: %s" % fname_out
n_ch, sample_rate = get_dat_pars(fname_in)
n_ch, _ = get_dat_pars(fname_in)
in_memmap = np.memmap(fname_in, mode="r", dtype=DT_IN).reshape(-1, n_ch)
n_s = in_memmap.shape[0]
out_memmap = np.memmap(fname_out,
mode="w+",
shape=(n_s, n_ch),
dtype=DT_OUT)
if f_lo is None:
b, a = butter(BUTTER_ORD, f_hi / (sample_rate / 2), "high")
elif f_hi is None:
b, a = butter(BUTTER_ORD, f_hi / (sample_rate / 2), "low")
else:
b, a = butter(BUTTER_ORD,
(f_lo / (sample_rate / 2), f_hi / (sample_rate / 2)),
"pass")
if two_way: filter_2way(b, a, in_memmap, out_memmap)
else: filter_1way(b, a, in_memmap, out_memmap)