def __init__(self, baseplate, lock, *args, **kwargs):
super().__init__(baseplate, lock, acl=kwargs.get('acl'))
self.insult_array = []
insults = h.load_file(os.path.join(LOCAL_DIR, 'lists', 'quotes.txt'),
raw=True)
if insults:
self.insult_array = insults.splitlines()
def __init__(self, baseplate, lock, *args, **kwargs):
super().__init__(baseplate, lock, acl=kwargs.get('acl'))
self.role_modifiers_list = []
self.roles_list = []
self.companies_list = []
role_modifiers = h.load_file(os.path.join(LOCAL_DIR, 'lists',
'role_modifiers.txt'),
raw=True)
roles = h.load_file(os.path.join(LOCAL_DIR, 'lists', 'roles.txt'),
raw=True)
companies = h.load_file(os.path.join(LOCAL_DIR, 'lists',
'companies.txt'),
raw=True)
if role_modifiers and roles and companies:
self.role_modifiers_list = role_modifiers.splitlines()
self.roles_list = roles.splitlines()
self.companies_list = companies.splitlines()
def search(collection: str, query: str, populate: bool) -> list:
query = remove_accents(query)
model = load_file(f"{collection}.json", model_selector(collection))
def find_match(city: Model):
return re.search(query, remove_accents(city.name),
re.IGNORECASE | re.MULTILINE) != None
res = list(filter(find_match, model))
if (populate):
res = populate_collection(res, collection)
return res
def __init__(self, baseplate, lock, *args, **kwargs):
super().__init__(baseplate, lock, acl=kwargs.get('acl'))
self._set_bot_thread()
self._set_defaults(kwargs)
# set tx locations and load/initialize db
tables = {}
table_dir = os.path.join(LOCAL_DIR, 'tables')
for _file in os.listdir(table_dir):
path = os.path.join(table_dir, _file)
if os.path.isfile(path):
tables.update(h.load_file(path))
self._init_tx_db(tables, kwargs.get('seeds', {}))
def process_seeds(self, seeds):
insert = None
if seeds:
for table_name, info in seeds.items():
if table_name in self.tables:
table = getattr(self, table_name)
if table.count() == 0:
if isinstance(info, list):
insert = table.bulk_insert(info)
elif isinstance(info, dict):
if 'file' in info:
load_kwargs = info.get('load_kwargs', {})
data = load_file(info['file'], **load_kwargs)
else:
data = info.get('data', [])
transform = info.get('transform')
if transform:
data = jsearch(transform, data)
insert = table.bulk_insert(data)
return insert
print('Getting model...')
result = None
with open(os.path.join(model_path, 'train.pickle'), 'rb') as f:
result = pickle.load(f)
label_probs = result[0]
probs_per_label = result[1]
words = result[2]
labels = result[3]
predictor = partial(helpers.predict,
label_probs, probs_per_label, words, labels)
if os.path.isdir(input_path):
print('Loading dataset...')
test_target, test_data = helpers.load_dataset(input_path)
print('Testing dataset...')
print("============= RESULT ===============")
accuracy = helpers.get_accuracy(test_data, test_target, predictor)
print('Accuracy: {0:.2f}%'.format(accuracy * 100))
print('=====================================')
else:
print('Testing...')
label, tokens = helpers.load_file(input_path)
predict_label = predictor(tokens)
print('================= RESULT ================')
print('Expected label for the text: {}'.format(label))
print('Predicted label for the text: {}'.format(predict_label))
print('==========================================')
def _load_results(example_name,
wl_method,
evaluation_metric,
fold,
study,
alpha=None,
inferred_predicates_file='inferred-predicates.txt'):
# path to this file relative to caller
dirname = os.path.dirname(__file__)
# read inferred predicates
# predicted dataframe
if study == "sampling_study":
tuffy_experiment_directory = "{}/../../results/weightlearning/tuffy/sampling_study/{}/{}/{}/{}/{}".format(
dirname, example_name, wl_method, evaluation_metric, alpha, fold)
else:
tuffy_experiment_directory = "{}/../../results/weightlearning/tuffy/{}/{}/{}/{}/{}".format(
dirname, study, example_name, wl_method, evaluation_metric, fold)
results_path = os.path.join(tuffy_experiment_directory,
inferred_predicates_file)
print(results_path)
results_tmp = load_file(results_path)
results = []
targets_path = os.path.join(tuffy_experiment_directory, 'query.db')
for result in results_tmp:
if len(result) == 1:
# then we did not run in marginal mode, i.e. outputs in this file are all "true" or 1
predicate = result[0][result[0].find("(") +
1:result[0].find(")")].replace(' ',
'').split(',')
predicate.append(1.0)
results.append(predicate)
else:
# we ran this experiment in marginal mode, i.e., the marginal probability precedes the ground atom
predicate = result[1][result[1].find("(") +
1:result[1].find(")")].replace(' ',
'').split(',')
predicate.append(float(result[0]))
results.append(predicate)
# close the predictions if we ran in discrete mode, i.e. if the target was not in the results then we predicted 0
targets_tmp = load_file(targets_path)
targets = []
for target in targets_tmp:
predicate = target[0][target[0].find("(") +
1:target[0].find(")")].replace(' ',
'').split(',')
predicate.append(0.0)
targets.append(predicate)
# append the targets that were not in the inferred predicates
results_dict = {(result[0], result[1]): result[2] for result in results}
targets_dict = {(target[0], target[1]): target[2] for target in targets}
diff = set(targets_dict.keys()) - set(results_dict.keys())
for target in diff:
results.append(
[target[0], target[1], targets_dict[(target[0], target[1])]])
return results
def _load_moined(self):
path = os.path.join(self.tx_dir, 'moined.json')
if os.path.isfile(path):
return h.load_file(path)
return []
def main():
st.set_option('deprecation.showfileUploaderEncoding', False)
def _max_width_():
max_width_str = f"max-width: 1000px;"
st.markdown(
f"""
<style>
.reportview-container .main .block-container{{
{max_width_str}
}}
</style>
""",
unsafe_allow_html=True,
)
# Hide the Streamlit header and footer
def hide_header_footer():
hide_streamlit_style = """
<style>
footer {visibility: hidden;}
</style>
"""
st.markdown(hide_streamlit_style, unsafe_allow_html=True)
# increases the width of the text and tables/figures
_max_width_()
# hide the footer and optionally the streamlit menu in the topright corner which is unrelated to our app
hide_header_footer()
# show the intro page
text_markdown.intro_page()
# load the data, currently allows for csv and excel imports
st.sidebar.title(":floppy_disk: Upload Your File")
filename = st.sidebar.file_uploader("Choose a file", type=["xlsx", "csv"])
delim = st.sidebar.selectbox("In case of a CSV file, pick the delimiter.",
[",", ";", "|"])
if filename:
df = helpers.load_file(filename, delim)
else:
df = pd.read_excel("titanic.xlsx")
# space between sections
helpers.betweensection_space()
helpers.sidebar_space()
# Ensures navigation between pages
df = eda.first_inspection(df)
eda.visuals(df)
helpers.betweensection_space()
helpers.sidebar_space()
preprocessing.preprocess(df)
# bottom line and github logo
st.markdown("---")
