def get_all_topics_table(self):
""" Transform all topics from gensim model to table. """
all_words = self._topics_words(self.n_words)
all_weights = self._topics_weights(self.n_words)
sorted_words = sorted(all_words[0])
n_topics = len(all_words)
X = []
for words, weights in zip(all_words, all_weights):
weights = [we for wo, we in sorted(zip(words, weights))]
X.append(weights)
X = np.array(X)
# take only first n_topics; e.g. when user requested 10, but gensim
# returns only 9 — when the rank is lower than num_topics requested
names = np.array(self.topic_names[:n_topics], dtype=object)[:, None]
attrs = [ContinuousVariable(w) for w in sorted_words]
metas = [
StringVariable('Topics'),
ContinuousVariable('Marginal Topic Probability')
]
topic_proba = np.array(self._marginal_probability(
self.tokens, self.doc_topic),
dtype=object)
t = Table.from_numpy(Domain(attrs, metas=metas),
X=X,
metas=np.hstack((names, topic_proba)))
t.name = 'All topics'
return t
def apply_domain_edit(self):
if self.data is None:
table = None
else:
domain, cols = self.domain_editor.get_domain(
self.data.domain, self.data)
printData = self.data
printDomain = self.data.domain
if not (domain.variables or domain.metas):
table = None
else:
X, y, m = cols
#X是输入，domain.attributes;y是输出class_var;m是元特征
## 下面解决将self.data的数据付给了table。
# 1data's name; 2数据编号ids；3数据属性attributes
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
index = self.data.ids
table.ids = np.array(self.data.ids)
# print('ids',table.ids)
data = self.data
table.attributes = getattr(self.data, 'attributes', {})
## 将table的属性定义为{}
''' 对Ourputs的data赋值为table'''
# print('table is :',table)
# print('table domain',table.domain)
# print('table name',table.name)
# print('table class_var name',table.domain.class_vars[0].name)
self.Outputs.data.send(table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
self.Warning.performance_warning.clear()
self.Warning.renamed_vars.clear()
if self.data is None:
table = None
else:
domain, cols, renamed = \
self.domain_editor.get_domain(self.data.domain, self.data,
deduplicate=True)
if not (domain.variables or domain.metas):
table = None
elif domain is self.data.domain:
table = self.data
else:
X, y, m = cols
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
table.ids = np.array(self.data.ids)
table.attributes = getattr(self.data, 'attributes', {})
self._inspect_discrete_variables(domain)
if renamed:
self.Warning.renamed_vars(f"Renamed: {', '.join(renamed)}")
summary = len(table) if table else self.info.NoOutput
details = format_summary_details(table) if table else ""
self.info.set_output_summary(summary, details)
self.Outputs.data.send(table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
self.Warning.performance_warning.clear()
self.Warning.renamed_vars.clear()
if self.data is None:
table = None
else:
domain, cols, renamed = \
self.domain_editor.get_domain(self.data.domain, self.data,
deduplicate=True)
if not (domain.variables or domain.metas):
table = None
elif domain is self.data.domain:
table = self.data
else:
X, y, m = cols
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
table.ids = np.array(self.data.ids)
table.attributes = getattr(self.data, 'attributes', {})
self._inspect_discrete_variables(domain)
if renamed:
self.Warning.renamed_vars(f"Renamed: {', '.join(renamed)}")
self.Warning.multiple_targets(
shown=table is not None and len(table.domain.class_vars) > 1)
self.Outputs.data.send(table)
self.apply_button.setEnabled(False)
def commit(self):
if self._embedder != '' and self._smiles_attr != '':
smiles = self.data[:, self._smiles_attr].metas.flatten()
embedded, valid = self.to_fingerprints(smiles, self._embedder)
invalid = list(set(range(len(smiles))) - set(valid))
if not valid == []:
domain = [ContinuousVariable.make("C_{}".format(x)) for x in
range(embedded.shape[1])]
if self._embedder == MACCS:
domain = [ContinuousVariable.make(name)
for name, _ in Chem.MACCSkeys.smartsPatts.values()]
domain.append(ContinuousVariable.make('?'))
embedded_table = Table.from_numpy(
Domain(domain,
self.data.domain.class_vars,
self.data.domain.metas
),
embedded,
self.data.Y[valid],
self.data.metas[valid],
self.data.W[valid]
)
self.Outputs.embedded_smiles.send(embedded_table)
else:
self.Outputs.embedded_smiles.send(None)
if not invalid == []:
invalid_table = Table.from_numpy(
self.data.domain,
self.data.X[invalid],
self.data.Y[invalid],
self.data.metas[invalid],
self.data.W[invalid]
)
self.Outputs.skipped_smiles.send(invalid_table)
else:
self.Outputs.skipped_smiles.send(None)
def apply_domain_edit(self):
attributes = []
class_vars = []
metas = []
places = [attributes, class_vars, metas]
X, y, m = [], [], []
cols = [X, y, m] # Xcols, Ycols, Mcols
def is_missing(x):
return str(x) in ("nan", "")
for column, (name, tpe, place, vals, is_con), (orig_var, orig_plc) in \
zip(count(), self.editor_model.variables,
chain([(at, 0) for at in self.data.domain.attributes],
[(cl, 1) for cl in self.data.domain.class_vars],
[(mt, 2) for mt in self.data.domain.metas])):
if place == 3:
continue
if orig_plc == 2:
col_data = list(chain(*self.data[:, orig_var].metas))
else:
col_data = list(chain(*self.data[:, orig_var]))
if name == orig_var.name and tpe == type(orig_var):
var = orig_var
elif tpe == DiscreteVariable:
values = list(
str(i) for i in set(col_data) if not is_missing(i))
var = tpe(name, values)
col_data = [
np.nan if is_missing(x) else values.index(str(x))
for x in col_data
]
elif tpe == StringVariable and type(orig_var) == DiscreteVariable:
var = tpe(name)
col_data = [
orig_var.repr_val(x) if not np.isnan(x) else ""
for x in col_data
]
else:
var = tpe(name)
places[place].append(var)
cols[place].append(col_data)
domain = Domain(attributes, class_vars, metas)
X = np.array(X).T if len(X) else np.empty((len(self.data), 0))
y = np.array(y).T if len(y) else None
dtpe = object if any(
isinstance(m, StringVariable) for m in domain.metas) else float
m = np.array(m, dtype=dtpe).T if len(m) else None
table = Table.from_numpy(domain, X, y, m, self.data.W)
self.send("Data", table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
if self.data is not None:
domain, cols = self.domain_editor.get_domain(
self.data.domain, self.data)
X, y, m = cols
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
table.ids = np.array(self.data.ids)
table.attributes = getattr(self.data, 'attributes', {})
else:
table = self.data
self.send("Data", table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
if self.data is None:
table = None
else:
domain, cols = self.domain_editor.get_domain(self.data.domain, self.data)
if not (domain.variables or domain.metas):
table = None
else:
X, y, m = cols
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
table.ids = np.array(self.data.ids)
table.attributes = getattr(self.data, 'attributes', {})
self.Outputs.data.send(table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
if self.data is None:
table = None
else:
domain, cols = self.domain_editor.get_domain(self.data.domain, self.data)
if not (domain.variables or domain.metas):
table = None
else:
X, y, m = cols
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
table.ids = np.array(self.data.ids)
table.attributes = getattr(self.data, 'attributes', {})
self.Outputs.data.send(table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
attributes = []
class_vars = []
metas = []
places = [attributes, class_vars, metas]
X, y, m = [], [], []
cols = [X, y, m] # Xcols, Ycols, Mcols
def is_missing(x):
return str(x) in ("nan", "")
for column, (name, tpe, place, vals, is_con), (orig_var, orig_plc) in \
zip(count(), self.editor_model.variables,
chain([(at, 0) for at in self.data.domain.attributes],
[(cl, 1) for cl in self.data.domain.class_vars],
[(mt, 2) for mt in self.data.domain.metas])):
if place == 3:
continue
if orig_plc == 2:
col_data = list(chain(*self.data[:, orig_var].metas))
else:
col_data = list(chain(*self.data[:, orig_var]))
if name == orig_var.name and tpe == type(orig_var):
var = orig_var
elif tpe == DiscreteVariable:
values = list(str(i) for i in set(col_data) if not is_missing(i))
var = tpe(name, values)
col_data = [np.nan if is_missing(x) else values.index(str(x))
for x in col_data]
elif tpe == StringVariable and type(orig_var) == DiscreteVariable:
var = tpe(name)
col_data = [orig_var.repr_val(x) if not np.isnan(x) else ""
for x in col_data]
else:
var = tpe(name)
places[place].append(var)
cols[place].append(col_data)
domain = Domain(attributes, class_vars, metas)
X = np.array(X).T if len(X) else np.empty((len(self.data), 0))
y = np.array(y).T if len(y) else None
dtpe = object if any(isinstance(m, StringVariable)
for m in domain.metas) else float
m = np.array(m, dtype=dtpe).T if len(m) else None
table = Table.from_numpy(domain, X, y, m, self.data.W)
self.send("Data", table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
if self.data is not None:
domain, cols = self.domain_editor.get_domain(
self.data.domain, self.data)
X, y, m = cols
X = np.array(X).T if len(X) else np.empty((len(self.data), 0))
y = np.array(y).T if len(y) else None
dtpe = object if any(
isinstance(m, StringVariable) for m in domain.metas) else float
m = np.array(m, dtype=dtpe).T if len(m) else None
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
table.ids = np.array(self.data.ids)
table.attributes = getattr(self.data, 'attributes', {})
else:
table = self.data
self.send("Data", table)
self.apply_button.setEnabled(False)
def apply_domain_edit(self):
self.Warning.performance_warning.clear()
if self.data is None:
table = None
else:
domain, cols = self.domain_editor.get_domain(
self.data.domain, self.data)
if not (domain.variables or domain.metas):
table = None
else:
X, y, m = cols
table = Table.from_numpy(domain, X, y, m, self.data.W)
table.name = self.data.name
table.ids = np.array(self.data.ids)
table.attributes = getattr(self.data, 'attributes', {})
self._inspect_discrete_variables(domain)
self.Outputs.data.send(table)
self.apply_button.setEnabled(False)
def insert_topics_into_corpus(self, corp_in):
"""
Insert topical representation into corpus.
:param corp_in: Corpus into whic we want to insert topical representations
:return: `Orange.data.table.Table`
"""
matrix = matutils.corpus2dense(self.corpus,
num_terms=self.num_topics).T
# Generate the new table.
attr = [ContinuousVariable(n) for n in self.topic_names]
domain = Domain(attr,
corp_in.domain.class_vars,
metas=corp_in.domain.metas)
return Table.from_numpy(domain,
matrix,
Y=corp_in._Y,
metas=corp_in.metas)
def insert_topics_into_corpus(self, corp_in):
"""
Insert topical representation into corpus.
:param corp_in: Corpus into whic we want to insert topical representations
:return: `Orange.data.table.Table`
"""
matrix = matutils.corpus2dense(self.corpus,
num_terms=self.num_topics).T
# Generate the new table.
attr = [ContinuousVariable(n) for n in self.topic_names]
domain = Domain(attr,
corp_in.domain.class_vars,
metas=corp_in.domain.metas)
return Table.from_numpy(domain,
matrix,
Y=corp_in._Y,
metas=corp_in.metas)
def get_all_topics_table(self):
""" Transform all topics from gensim model to table. """
all_words = self._topics_words(self.n_words)
all_weights = self._topics_weights(self.n_words)
sorted_words = sorted(all_words[0])
n_topics = len(all_words)
X = []
for words, weights in zip(all_words, all_weights):
weights = [we for wo, we in sorted(zip(words, weights))]
X.append(weights)
X = np.array(X).T
# take only first n_topics; e.g. when user requested 10, but gensim
# returns only 9 — when the rank is lower than num_topics requested
attrs = [ContinuousVariable(n)
for n in self.topic_names[:n_topics]]
t = Table.from_numpy(Domain(attrs, metas=[StringVariable('Word')]),
X=X, metas=np.array(sorted_words)[:, None])
t.name = 'All topics'
return t
def get_all_topics_table(self):
""" Transform all topics from gensim model to table. """
all_words = self._topics_words(self.n_words)
all_weights = self._topics_weights(self.n_words)
sorted_words = sorted(all_words[0])
n_topics = len(all_words)
X = []
for words, weights in zip(all_words, all_weights):
weights = [we for wo, we in sorted(zip(words, weights))]
X.append(weights)
X = np.array(X).T
# take only first n_topics; e.g. when user requested 10, but gensim
# returns only 9 — when the rank is lower than num_topics requested
attrs = [ContinuousVariable(n) for n in self.topic_names[:n_topics]]
t = Table.from_numpy(Domain(attrs, metas=[StringVariable('Word')]),
X=X,
metas=np.array(sorted_words)[:, None])
t.name = 'All topics'
return t
