def aggregateFormulaFeatureMaps(fMap, onlyTheorems, ignoreNumbers):
documentFeatureMap = { }
for fid, features in fMap.items():
if ignoreNumbers:
features = dict(filter(lambda kv : not(kv[0][:3] == "num"), features.items()))
if not onlyTheorems or re.search(r"thm", fid.lower()):
addToDict(documentFeatureMap, features)
return documentFeatureMap
def aggregateFormulaFeatureMaps(fMap, onlyTheorems, ignoreNumbers):
documentFeatureMap = {}
for fid, features in fMap.items():
if ignoreNumbers:
features = dict(
filter(lambda kv: not (kv[0][:3] == "num"), features.items()))
if not onlyTheorems or re.search(r"thm", fid.lower()):
addToDict(documentFeatureMap, features)
return documentFeatureMap
def formulasToFeatureCounts(formulaIds):
formulaFile = "derived_data/formula_features/" + filter(lambda c : c in (digits + ascii_letters), documentId) + ".json"
formulas = json.load(open(formulaFile))
m = { }
formulaFeatureList = map(lambda fid : formulas[fid], formulaIds)
for formulaFeatureMap in formulaFeatureList:
addToDict(m, formulaFeatureMap)
return m
def formulasToFeatureCounts(formulaIds):
formulaFile = "derived_data/formula_features/" + filter(
lambda c: c in (digits + ascii_letters), documentId) + ".json"
formulas = json.load(open(formulaFile))
m = {}
formulaFeatureList = map(lambda fid: formulas[fid], formulaIds)
for formulaFeatureMap in formulaFeatureList:
addToDict(m, formulaFeatureMap)
return m
