def test_flow_forward_apply(flow_spec):
z0 = pm.tt_rng().normal(size=(10, 20))
flow = flow_spec(dim=20, z0=z0)
with change_flags(compute_test_value='off'):
dist = flow.forward
shape_dist = dist.shape.eval()
assert tuple(shape_dist) == (10, 20)
def test_flow_det_shape(flow_spec):
with change_flags(compute_test_value='off'):
z0 = pm.tt_rng().normal(size=(10, 20))
flow = flow_spec(dim=20, z0=z0)
det = flow.logdet
det_dist = det.shape.eval()
assert tuple(det_dist) == (10,)
def test_flow_det_shape(flow_spec):
with change_flags(compute_test_value='off'):
z0 = pm.tt_rng().normal(size=(10, 20))
flow = flow_spec(dim=20, z0=z0)
det = flow.logdet
det_dist = det.shape.eval()
assert tuple(det_dist) == (10, )
def test_flow_forward_apply(flow_spec):
z0 = pm.tt_rng().normal(size=(10, 20))
flow = flow_spec(dim=20, z0=z0)
with change_flags(compute_test_value='off'):
dist = flow.forward
shape_dist = dist.shape.eval()
assert tuple(shape_dist) == (10, 20)
def test_gen_cloning_with_shape_change(self, datagen):
gen = generator(datagen)
gen_r = tt_rng().normal(size=gen.shape).T
X = gen.dot(gen_r)
res, _ = theano.scan(lambda x: x.sum(), X, n_steps=X.shape[0])
assert res.eval().shape == (50,)
shared = theano.shared(datagen.data.astype(gen.dtype))
res2 = theano.clone(res, {gen: shared**2})
assert res2.eval().shape == (1000,)
def test_gen_cloning_with_shape_change(self, datagen):
gen = generator(datagen)
gen_r = tt_rng().normal(size=gen.shape).T
X = gen.dot(gen_r)
res, _ = theano.scan(lambda x: x.sum(), X, n_steps=X.shape[0])
assert res.eval().shape == (50,)
shared = theano.shared(datagen.data.astype(gen.dtype))
res2 = theano.clone(res, {gen: shared**2})
assert res2.eval().shape == (1000,)
def rslice(total, size, seed):
if size is None:
return slice(None)
elif isinstance(size, int):
return (pm.tt_rng(seed).uniform(size=(size, ),
low=0.0,
high=pm.floatX(total) -
1e-16).astype('int64'))
else:
raise TypeError('Unrecognized size type, %r' % size)
def rslice(self, total, size, seed):
if size is None:
return slice(None)
elif isinstance(size, int):
rng = pm.tt_rng(seed)
Minibatch.RNG[id(self)].append(rng)
return rng.uniform(size=(size, ),
low=0.0,
high=pm.floatX(total) - 1e-16).astype("int64")
else:
raise TypeError("Unrecognized size type, %r" % size)
def rslice(self, total, size, seed):
if size is None:
return slice(None)
elif isinstance(size, int):
rng = pm.tt_rng(seed)
Minibatch.RNG[id(self)].append(rng)
return (rng
.uniform(size=(size, ), low=0.0, high=pm.floatX(total) - 1e-16)
.astype('int64'))
else:
raise TypeError('Unrecognized size type, %r' % size)
def run_model(prior, cov, lik, R, K, M, N, tau, h2, data):
nmp.random.seed(42)
pmc.tt_rng(42)
res = cln.infer(data,
model_args={
'K': 20,
'prior': prior,
'cov': cov,
'lik': lik,
'threshold': 0.0
},
pymc3_args={
'niters': 40000,
'method': 'advi',
'flow': 'scale-loc',
'learning_rate': 1e-2,
'random_seed': 42
})
z_true = data[['MUTID', 'CLUSTERID']].drop_duplicates().CLUSTERID.values
z_pred = res['data'][['MUTID',
'CLUSTERID']].drop_duplicates().CLUSTERID.values
return pnd.DataFrame(
{
'REP': R,
'NCLUSTERS': K,
'NSAMPLES': M,
'NMUTS': N,
'TAU': tau,
'H2': h2,
'PRIOR': prior,
'COV': cov,
'LIK': lik,
'ARI': mtr.adjusted_rand_score(z_true, z_pred),
'AMI': mtr.adjusted_mutual_info_score(z_true, z_pred),
'FMI': mtr.fowlkes_mallows_score(z_true, z_pred),
},
index=[0]).reset_index(drop=True)
y_data = hist.flatten()
#%% pymc3 minibatch setup
# Not suitable for 2D mapping problem, overestimated lengthscale
batchsize = 10
Xbatch = pm.Minibatch(x_data, batchsize**2)
Ybatch = pm.Minibatch(y_data, batchsize**2)
#%% set up minibatch
data = hist
batchsize = 10
z1, z2 = batchsize, batchsize
s1, s2 = np.shape(data)
yshared = theano.shared(data)
x1shared = theano.shared(ycenters[:, np.newaxis].repeat(64, axis=1))
x2shared = theano.shared(xcenters[:, np.newaxis].T.repeat(64, axis=0))
ixs1 = pm.tt_rng().uniform(size=(1, ), low=0,
high=s1 - z1 - 1e-10).astype('int64')
ixs2 = pm.tt_rng().uniform(size=(1, ), low=0,
high=s2 - z2 - 1e-10).astype('int64')
range1 = tt.arange(ixs1.squeeze(), (ixs1 + z1).squeeze())
range2 = tt.arange(ixs2.squeeze(), (ixs2 + z2).squeeze())
Ybatch = yshared[range1][:, range2].flatten()
Xbatch1 = x1shared[range1][:, range2].flatten()
Xbatch2 = x2shared[range1][:, range2].flatten()
Xbatch = tt.stack((Xbatch1, Xbatch2)).T
import theano
theano.config.compute_test_value = 'off'
#%%
with pm.Model() as model:
#hyper-parameter priors
def test_flow_init_loop(flow_spec):
flow = pm.tt_rng().normal(size=(10, 2))
for i in range(10):
flow = flow_spec(z0=flow, dim=2)
flow.forward.eval()
def test_flow_init_loop(flow_spec):
flow = pm.tt_rng().normal(size=(10, 2))
for i in range(10):
flow = flow_spec(z0=flow, dim=2)
flow.forward.eval()
