def _cached_model_path(cls):
env_path = os.environ.get("FORMASAURUS_MODEL")
if env_path:
return os.path.expanduser(env_path)
path = "formasaurus-%s.joblib" % dependencies_string()
return at_root(path)
def _cached_model_path(cls):
env_path = os.environ.get("FORMASAURUS_MODEL")
if env_path:
return os.path.expanduser(env_path)
path = "formasaurus-%s.joblib" % dependencies_string()
return at_root(path)
# -*- coding: utf-8 -*-
from __future__ import absolute_import
import os
import six
import joblib
from formasaurus import formtype_model, fieldtype_model
from formasaurus.html import get_forms, get_fields_to_annotate, load_html
from formasaurus.storage import Storage
from formasaurus.utils import dependencies_string, at_root, thresholded
DEFAULT_DATA_PATH = at_root('data')
def extract_forms(tree_or_html, proba=False, threshold=0.05, fields=True):
"""
Given a lxml tree or HTML source code, return a list of
``(form_elem, form_info)`` tuples.
``form_info`` dicts contain results of :meth:`classify` or
:meth:`classify_proba`` calls, depending on ``proba`` parameter.
When ``fields`` is False, field type information is not computed.
"""
return get_instance().extract_forms(
tree_or_html=tree_or_html,
proba=proba,
threshold=threshold,
fields=fields,
)
# -*- coding: utf-8 -*-
from __future__ import absolute_import
import os
import six
from sklearn.externals import joblib
from formasaurus import formtype_model, fieldtype_model
from formasaurus.html import get_forms, get_fields_to_annotate, load_html
from formasaurus.storage import Storage
from formasaurus.utils import dependencies_string, at_root, thresholded
DEFAULT_DATA_PATH = at_root('data')
def extract_forms(tree_or_html, proba=False, threshold=0.05, fields=True):
"""
Given a lxml tree or HTML source code, return a list of
``(form_elem, form_info)`` tuples.
``form_info`` dicts contain results of :meth:`classify` or
:meth:`classify_proba`` calls, depending on ``proba`` parameter.
When ``fields`` is False, field type information is not computed.
"""
return get_instance().extract_forms(
tree_or_html=tree_or_html,
proba=proba,
threshold=threshold,
fields=fields,
)
