def test_build_averaged_flowgram(self):
f1 = [0.3, 1.1, 4.0 , 0.01, 0.8, 0.0, 0.0, 2.0]
f2 = [0.6, 0.9, 4.05, 0.1, 1.2, 0.1, 0.4]
f3 = [0.4, 1.2, 4.05, 0.2, 1.3, 0.2]
f4 = [0.7, 1.0, 4.0 , 0.02, 1.5]
flowgrams = [f1,f2,f3,f4]
self.assertFloatEqual(build_averaged_flowgram(flowgrams),
[0.5, 1.05, 4.03, 0.08, 1.2, 0.1, 0.2, 2.0])
self.assertFloatEqual(build_averaged_flowgram([f1,f1,f1,f1,f1,f1]),
[0.3, 1.1, 4.0 , 0.01, 0.8, 0.0, 0.0, 2.0])
def test_build_averaged_flowgram(self):
f1 = [0.3, 1.1, 4.0, 0.01, 0.8, 0.0, 0.0, 2.0]
f2 = [0.6, 0.9, 4.05, 0.1, 1.2, 0.1, 0.4]
f3 = [0.4, 1.2, 4.05, 0.2, 1.3, 0.2]
f4 = [0.7, 1.0, 4.0, 0.02, 1.5]
flowgrams = [f1, f2, f3, f4]
self.assertFloatEqual(build_averaged_flowgram(flowgrams),
[0.5, 1.05, 4.03, 0.08, 1.2, 0.1, 0.2, 2.0])
self.assertFloatEqual(
build_averaged_flowgram([f1, f1, f1, f1, f1, f1]),
[0.3, 1.1, 4.0, 0.01, 0.8, 0.0, 0.0, 2.0])
def _average_flowgrams(mapping, flowgrams, sample_keys):
"""average flowgrams according to cluster mapping.
mapping: a dictionary of lists as cluster mapping
flowgrams: an iterable flowgram source, all flowgram ids from this source must be in the mapping
sample_keys: the keys that should be averaged over for each cluster.
"""
# accumulates flowgram for each key until sample for this key is empty
flows = defaultdict(list)
invert_map = invert_mapping(mapping)
for f in flowgrams:
key = invert_map[f.Name]
samples = sample_keys[key]
if (f.Name in samples):
flows[key].append(f.flowgram)
samples.remove(f.Name)
if (len(samples) == 0):
# we gathered all sampled flowgrams for this cluster,
# now average
ave_flowgram = build_averaged_flowgram(flows[key])
ave_f = Flowgram(ave_flowgram, Name=key)
del(flows[key])
yield ave_f, key
def _average_flowgrams(mapping, flowgrams, sample_keys):
"""average flowgrams according to cluster mapping.
mapping: a dictionary of lists as cluster mapping
flowgrams: an iterable flowgram source, all flowgram ids from this source must be in the mapping
sample_keys: the keys that should be averaged over for each cluster.
"""
flows = defaultdict(list) # accumulates flowgram for each key until sample for this key is empty
invert_map = invert_mapping(mapping)
for f in flowgrams:
key = invert_map[f.Name]
samples = sample_keys[key]
if (f.Name in samples):
flows[key].append(f.flowgram)
samples.remove(f.Name)
if (len(samples)==0):
#we gathered all sampled flowgrams for this cluster,
#now average
ave_flowgram = build_averaged_flowgram(flows[key])
ave_f = Flowgram(ave_flowgram, Name=key)
del(flows[key])
yield ave_f, key
