def save_from_origin(self, labeled_comments: LabeledComments, origin_name):
path = fileutils.join(self._folderpath, labeled_comments.name)
path_origin = fileutils.join(self._folderpath, origin_name)
ids = []
with open(path_origin, "r") as f:
reader = csv.DictReader(f)
for row in reader:
ids.append(row["id"]),
with open(path, "w") as f:
writer = csv.DictWriter(f, fieldnames=["id", "text", "polarity"])
for id, comment, label in zip(ids, labeled_comments.comments,
labeled_comments.labels):
row = {"id": id, "text": comment, "polarity": label}
writer.writerow(row)
def save(self, labeled_comments: LabeledComments):
path = fileutils.join(self._folderpath, labeled_comments.name)
with open(path, "w") as f:
writer = csv.DictWriter(f)
for comment, label in zip(labeled_comments.comments,
labeled_comments.labels):
row = {"text": comment, "polarity": label}
writer.writerow(row)
def get(self, name: str) -> LabeledComments:
path = fileutils.join(self._folderpath, name)
comments = []
labels = []
with open(path, "r") as f:
reader = csv.DictReader(f)
for row in reader:
comments.append(self._cleaner.clean(row["text"])),
labels.append(row["polarity"])
return LabeledComments(name, comments, labels)
def save_sklearn(self, name:str, model:SVC):
name = f"{name}.joblib"
fullpath = fileutils.join(self._folderpath, name)
joblib.dump(model, fullpath)
def get_sklearn(self, name:str) -> SVC:
name = f"{name}.joblib"
fullpath = fileutils.join(self._folderpath, name)
model = joblib.load(fullpath)
return model
def get_keras_weights(self, name:str, model:Model):
name = f"{name}.hdf5"
fullpath = fileutils.join(self._folderpath, name)
model.load_weights(fullpath)
return model
def save_keras_weights(self, name:str, model:Model):
name = f"{name}.hdf5"
fullpath = fileutils.join(self._folderpath, name)
model.save_weights(fullpath, overwrite=True)
def get_keras(self, name:str) -> Model:
name = f"{name}.h5"
fullpath = fileutils.join(self._folderpath, name)
model = load_model(fullpath, compile=True)
return model
def __init__(self, folderpath, filename, cleaner:SentenceCleaner):
self._name = filename
self._xml = et.parse(fileutils.join(folderpath, filename))
self._cleaner = cleaner
def save(self, folderpath, filename, xml):
path = fileutils.join(folderpath, filename)
xml.write(path)
