def omni_procrust_svd(embed_adjs):
omni = OmnibusEmbed(n_components=None, check_lcc=False)
joint_embed = omni.fit_transform(embed_adjs)
cat_embed = np.concatenate(joint_embed, axis=-1)
# print(f"Omni concatenated embedding shape: {cat_embed.shape}")
for e in cat_embed:
e[left_inds] = e[left_inds] @ orthogonal_procrustes(
e[lp_inds], e[rp_inds])[0]
cat_embed = np.concatenate(cat_embed, axis=-1)
U, S, Vt = selectSVD(cat_embed, n_elbows=3)
return U
def ase_procrust_svd(embed_adjs):
ase = AdjacencySpectralEmbed(n_components=None)
all_embeds = []
for a in embed_adjs:
embed = ase.fit_transform(a)
embed = np.concatenate(embed, axis=1)
embed[left_inds] = (embed[left_inds] @ orthogonal_procrustes(
embed[lp_inds], embed[rp_inds])[0])
print(embed.shape)
all_embeds.append(embed)
cat_embed = np.concatenate(all_embeds, axis=1)
print(cat_embed.shape)
U, S, Vt = selectSVD(cat_embed, n_elbows=3)
return U
def _embed(self, adj=None):
if adj is None:
adj = self.adj
# TODO look into PTR at this level as well
# lp_inds, rp_inds = get_paired_inds(self.meta)
lp_inds = self.left_pair_inds
rp_inds = self.right_pair_inds
embed_adj = pass_to_ranks(adj)
if self.embed == "ase":
embedder = AdjacencySpectralEmbed(
n_components=self.n_components, n_elbows=self.n_elbows
)
embed = embedder.fit_transform(embed_adj)
elif self.embed == "lse":
embedder = LaplacianSpectralEmbed(
n_components=self.n_components,
n_elbows=self.n_elbows,
regularizer=self.regularizer,
)
embed = embedder.fit_transform(embed_adj)
elif self.embed == "unscaled_ase":
embed_adj = pass_to_ranks(adj)
embed_adj = augment_diagonal(embed_adj)
embed = selectSVD(
embed_adj, n_components=self.n_components, n_elbows=self.n_elbows
)
embed = (embed[0], embed[2].T)
X = np.concatenate(embed, axis=1)
fraction_paired = (len(lp_inds) + len(rp_inds)) / len(self.root_inds)
print(f"Learning transformation with {fraction_paired} neurons paired")
R, _ = orthogonal_procrustes(X[lp_inds], X[rp_inds])
X[self.left_inds] = X[self.left_inds] @ R
if self.normalize:
row_sums = np.sum(X, axis=1)
X /= row_sums[:, None]
return X
def _embed(self, adj=None):
if adj is None:
adj = self.adj
lp_inds = self.left_pair_inds
rp_inds = self.right_pair_inds
embed_adj = pass_to_ranks(adj) # TODO PTR here?
if self.plus_c:
embed_adj += 1 / adj.size
if self.embed == "ase":
embedder = AdjacencySpectralEmbed(n_components=self.n_components,
n_elbows=self.n_elbows)
embed = embedder.fit_transform(embed_adj)
elif self.embed == "lse":
embedder = LaplacianSpectralEmbed(
n_components=self.n_components,
n_elbows=self.n_elbows,
regularizer=self.regularizer,
)
embed = embedder.fit_transform(embed_adj)
elif self.embed == "unscaled_ase":
embed_adj = augment_diagonal(embed_adj)
embed = selectSVD(embed_adj,
n_components=self.n_components,
n_elbows=self.n_elbows)
embed = (embed[0], embed[2].T)
X = np.concatenate(embed, axis=1)
fraction_paired = (len(lp_inds) + len(rp_inds)) / len(self.root_inds)
print(f"Learning transformation with {fraction_paired} neurons paired")
X = self._procrustes(X)
if self.normalize:
row_norms = np.linalg.norm(X, axis=1)
X /= row_norms[:, None]
return X
def svd(X, n_components=n_svd_components):
return selectSVD(X, n_components=n_components, algorithm="full")[0]
mg = mg.make_lcc()
print(len(mg))
mg.calculate_degrees(inplace=True)
meta = mg.meta
meta["inds"] = range(len(meta))
adj = mg.adj.copy()
lp_inds, rp_inds = get_paired_inds(meta)
left_inds = meta[meta["left"]]["inds"]
embed = lateral_omni(adj, lp_inds, rp_inds)
labels = np.concatenate((meta["merge_class"].values[lp_inds],
meta["merge_class"].values[rp_inds]))
U, S, V = selectSVD(embed, n_components=6)
plot_pairs(
U,
labels,
left_pair_inds=np.arange(len(lp_inds)),
right_pair_inds=np.arange(len(lp_inds)) + len(lp_inds),
)
stashfig(f"pairs-low-threshold-quantile={quantile}")
print()
# %% [markdown]
# ## look at removing high degree nodes
quantiles = np.linspace(0, 0.5, 6)
master_mg = load_metagraph(graph_type, version="2020-04-01")
for quantile in quantiles:
verbose=10,
)
agmm.fit(all_hop_hist.T)
# %% [markdown]
# ##
from graspy.embed import select_dimension
select_dimension(all_hop_hist.T, n_elbows=5)
#%%
from graspy.embed import selectSVD
from graspy.plot import pairplot
n_elbows = 3
U, S, V = selectSVD(all_hop_hist.T, n_elbows=n_elbows)
plot_df = pd.DataFrame(data=U)
plot_df["label"] = meta["merge_class"].values
pg = sns.PairGrid(plot_df,
hue="label",
palette=CLASS_COLOR_DICT,
vars=np.arange(U.shape[1]),
height=4)
# pg._legend.remove()
# pg.map_diag(plt.hist)
pg.map_offdiag(sns.scatterplot, s=15, linewidth=0, alpha=0.7)
all_embeds.append(embed)
pdist = pairwise_distances(embed, metric="cosine")
all_pdists.append(pdist)
from mvlearn.embed import MVMDS
mvmds = MVMDS(n_components=6)
mv_embed = mvmds.fit_transform(all_pdists)
pairs(mv_embed)
# %% [markdown]
# ##
cat_embeds = np.concatenate(all_embeds, axis=-1)
U, S, Vt = selectSVD(cat_embeds, n_elbows=4)
pairs(U)
# %% [markdown]
# ##
results = crossval_cluster(U,
left_inds,
right_inds,
left_pair_inds=lp_inds,
right_pair_inds=rp_inds)
plot_metrics(results)
metric = "bic"
k = 6
# %% [markdown]
# ##
meta = mg.meta
lp_inds, rp_inds = get_paired_inds(meta)
left_inds = meta[meta["left"]]["inds"]
right_inds = meta[meta["right"]]["inds"]
cat_embed = np.concatenate(joint_embed, axis=-1)
for e in cat_embed:
e[left_inds] = e[left_inds] @ orthogonal_procrustes(
e[lp_inds], e[rp_inds])[0]
cat_embed = np.concatenate(cat_embed, axis=-1)
print(select_dimension(cat_embed, n_elbows=3))
U, S, Vt = selectSVD(cat_embed, n_elbows=3)
pg = pairplot(U,
labels=meta["merge_class"].values,
palette=CLASS_COLOR_DICT,
size=20,
alpha=0.4)
pg._legend.remove()
stashfig("omni-reduced-dim")
# %% [markdown]
# ##
results = crossval_cluster(U,
left_inds,
right_inds,
embed = bilateral_ase(a)
all_embeds.append(embed[0])
all_embeds.append(embed[1])
# U, _, _ = selectSVD(embed, n_elbows=2)
# plot_pairs(
# U,
# labels=meta["merge_class"].values,
# left_pair_inds=lp_inds,
# right_pair_inds=rp_inds,
# )
cat_all_embeds = np.concatenate(all_embeds, axis=1)
# %% [markdown]
# ##
# align_joint_embed = np.concatenate((align_ipsi_embed, align_contra_embed), axis=1)
# U, S, V = selectSVD(align_joint_embed)
U, S, V = selectSVD(cat_all_embeds, n_elbows=4)
print(U.shape)
plt.plot(S)
# %% [markdown]
# ##
plot_pairs(
U,
labels=meta["merge_class"].values,
left_pair_inds=lp_inds,
right_pair_inds=rp_inds,
)
# %% [markdown]
# ##
from graspy.utils import symmetrize
temp_adj = temp_mg.adj
adjs.append(temp_adj)
# embed_adjs = [pass_to_ranks(a) for a in adjs]
# %% [markdown]
# ## just omni on the 4 colors for the right right subgraph
right_embed_adjs = [pass_to_ranks(a[np.ix_(rp_inds, rp_inds)]) for a in adjs]
omni = OmnibusEmbed(check_lcc=False)
embeds = omni.fit_transform(right_embed_adjs)
embeds = np.concatenate(embeds, axis=-1)
embeds = np.concatenate(embeds, axis=-1)
print(embeds.shape)
U, S, V = selectSVD(embeds, n_components=8)
labels = meta["merge_class"].values[rp_inds]
plot_pairs(U, labels)
stashfig(f"simple-omni-right-reduced-4-color")
# %% [markdown]
# ## Look at what each edge color type looks like when regularized by g
# only the right right subgraph
right_full_adj = pass_to_ranks(adj[np.ix_(rp_inds, rp_inds)])
labels = meta["merge_class"].values[rp_inds]
all_reg_embeds = []
for a in right_embed_adjs:
omni = OmnibusEmbed(check_lcc=False)
embeds = omni.fit_transform([right_full_adj, a])