def test_sample_ppc_w(self):
data = np.random.normal(0, 1, size=200)
with pm.Model() as model_0:
mu = pm.Normal('mu', mu=0, sd=1)
y = pm.Normal('y', mu=mu, sd=1, observed=data)
trace_0 = pm.sample()
with pm.Model() as model_1:
mu = pm.Normal('mu', mu=0, sd=1, shape=len(data))
y = pm.Normal('y', mu=mu, sd=1, observed=data)
trace_1 = pm.sample()
traces = [trace_0, trace_0]
models = [model_0, model_0]
ppc = pm.sample_ppc_w(traces, 1000, models)
assert ppc['y'].shape == (1000,)
_, pval = stats.kstest(ppc['y'], stats.norm().cdf)
assert pval > 0.001
traces = [trace_0, trace_1]
models = [model_0, model_1]
ppc = pm.sample_ppc_w(traces, 50, models)
assert ppc['y'].shape == (50, 200)
def test_sample_ppc_w(self):
data0 = np.random.normal(0, 1, size=500)
with pm.Model() as model_0:
mu = pm.Normal('mu', mu=0, sd=1)
y = pm.Normal('y', mu=mu, sd=1, observed=data0, shape=500)
trace_0 = pm.sample()
with pm.Model() as model_1:
mu = pm.Normal('mu', mu=0, sd=1, shape=len(data0))
y = pm.Normal('y', mu=mu, sd=1, observed=data0, shape=500)
trace_1 = pm.sample()
traces = [trace_0, trace_0]
models = [model_0, model_0]
ppc = pm.sample_ppc_w(traces, 100, models)
assert ppc['y'].shape == (100, 500)
traces = [trace_0, trace_1]
models = [model_0, model_1]
ppc = pm.sample_ppc_w(traces, 100, models)
assert ppc['y'].shape == (100, 500)
np.exp(0.61)
# %%
logistic(4)
# %%
logistic(4 + 0.61)
# %%
d_pred = pd.DataFrame({"prosoc_left": [0, 1, 0, 1], "condition": [0, 0, 1, 1]})
traces = [trace_10_1, trace_10_2, trace_10_3]
models = [model_10_1, model_10_2, model_10_3]
chimp_ensemble = pm.sample_ppc_w(
traces=traces,
models=models,
samples=1000,
weights=comp_df.weight.sort_index(ascending=True),
)
# %%
rt = chimp_ensemble["pulled_left"]
pred_mean = np.zeros((1000, 4))
cond = d.condition.unique()
prosoc_l = d.prosoc_left.unique()
for i in range(len(rt)):
tmp = []
if rt[i].size < 2:
continue
for cp in cond:
for pl in prosoc_l:
