def generate_world_formatted_forecast(forecast_date,
target_metric='death',
target_aggr='inc'):
world_forecast = pd.DataFrame()
forecast_date = pd.to_datetime(forecast_date).date()
last_epiweek_enddate = get_epiweek_enddate(forecast_date +
epiweeks.timedelta(-7))
country_list = mu.get_data(scope='global', type='deaths').Country.unique()
top_country_list = [
'US', 'India', 'Brazil', 'Russia', 'France', 'United Kingdom',
'Turkey', 'Italy', 'Spain', 'Germany', 'Colombia', 'Argentina',
'Mexico', 'Poland', 'Iran', 'Iraq', 'Ukraine', 'South Africa', 'Peru',
'Netherlands', 'Belgium', 'Chile', 'Romania', 'Canada', 'Ecuador',
'Czechia', 'Pakistan', 'Hungary', 'Philippines', 'Switzerland'
]
for country in top_country_list:
try:
print(country)
country_forecast = generate_formatted_forecast('World', country, forecast_date)\
.query('target!="9 wk ahead inc death"')
latest_cum_country = mu.get_data_by_country(
country).loc[last_epiweek_enddate][0]
country_forecast = add_cum_forecast(country_forecast,
latest_cum_country)
world_forecast = pd.concat([world_forecast, country_forecast])
except (ValueError, IndexError):
pass
world_forecast.to_csv(
'data_processed/World-{}-AIpert-pwllnod.csv'.format(forecast_date),
index=False)
def generate_US_formatted_forecast(forecast_date,
target_metric='death',
target_aggr='inc'):
forecast_date = pd.to_datetime(forecast_date).date()
last_epiweek_enddate = get_epiweek_enddate(forecast_date +
epiweeks.timedelta(-7))
US_forecast = pd.DataFrame()
US_state_list = mu.get_data(scope='US', type='deaths').State.unique()
for state in US_state_list:
try:
print(state)
state_forecast = generate_formatted_forecast('US', state, forecast_date)\
.query('target!="9 wk ahead inc death"')
latest_cum_state = mu.get_data_by_state(
state).loc[last_epiweek_enddate][0]
state_forecast = add_cum_forecast(state_forecast, latest_cum_state)
US_forecast = pd.concat([US_forecast, state_forecast])
except (ValueError, IndexError):
pass
# Aggregate all states for US forecast
US_forecast_new = US_forecast.groupby(
['forecast_date', 'target', 'target_end_date', 'quantile',
'type']).sum().reset_index()
US_forecast_new['location'] = "US"
US_forecast_new = pd.concat([US_forecast_new, US_forecast])
US_forecast_new.to_csv(
'data_processed/{}-AIpert-pwllnod.csv'.format(forecast_date),
index=False)
def main():
X_train_rus, y_train_rus, X_test, y_test = get_data()
lr_clf = LogisticRegression(random_state=0)
lr_clf.fit(X_train_rus, y_train_rus)
y_score = lr_clf.predict_proba(X_test)[:, 1]
y_hat = lr_clf.predict(X_test)
lr_scores = pd.DataFrame({'y_true': y_test, 'y_hat': y_hat, 'y_score': y_score})
lr_scores.to_csv('../../Data/Modeling/Output/scores/logistic_regression_scores.csv')
def main():
X_train_rus, y_train_rus, X_test, y_test = get_data()
clf_xgb = xgb.XGBClassifier()
'''
parameters = {'eta': [0.3],
'max_depth': [8],
'min_child_weight': [5],
'gamma': [0.2],
'colsample_bytree': [1],
'objective': ['binary:logistic']}
'''
parameters = {
'eta': [0.1, 0.15, 0.2, 0.25, 0.3, 0.35],
'max_depth': [3, 4, 5, 6, 7, 8, 10, 12],
'min_child_weight': [1, 3, 5, 7],
'gamma': [0, 0.1, 0.2, 0.3, 0.4],
'colsample_bytree': [0.5, 0.7, 1],
'objective': ['binary:logistic']
}
grid = GridSearchCV(clf_xgb,
parameters,
n_jobs=4,
scoring='average_precision',
cv=2,
verbose=5)
best_params = grid.fit(X_train_rus, y_train_rus).best_params_
D_train = xgb.DMatrix(X_train_rus, label=y_train_rus)
D_test = xgb.DMatrix(X_test, label=y_test)
clf_xgb = xgb.train(best_params, D_train, 20)
y_score = clf_xgb.predict(D_test)
#y_hat = np.asarray([np.argmax(line) for line in y_score])
y_hat = y_score.copy()
y_hat[y_hat < 0.5] = 0
y_hat[y_hat >= 0.5] = 1
xgb_scores = pd.DataFrame({
'y_true': y_test,
'y_hat': y_hat,
'y_score': y_score
})
xgb_scores.to_csv('../../Data/Modeling/Output/scores/xgb_scores.csv')
from pathlib import Path
from fastai.vision import ImageImageList, ImageList
from model_utils import get_data, create_gen_learner
# Loading Paths for Model Load
path = Path('') # Path to data folder to load your model
path_lr = path / '' # Path to model weights
# Loading Paths to Inference
path_t = Path('') # Path to undamaged files
dmgpath = Path('') # Path to damage templates
inf_path = Path('') # Directory for files to be inferenced
# Creating gen and Loading saved Weights
src = ImageImageList.from_folder(path_lr).split_by_rand_pct(0.1, seed=42)
data_gen = get_data(1, 500, src, path_lr)
learn_gen = create_gen_learner(data_gen).load('') # LOAD MODEL HERE
test_list = ImageList.from_folder(inf_path)
# Starting Streamlit App
st.markdown('# **ML for Photo Repair**')
st.markdown('### Choose an Image and Damage Template:')
st.markdown('Click **Generate** to Create a damaged photo')
names = []
for filename in (os.listdir(path_t)):
if '.png' in filename:
names.append(filename)
elif '.jpg' in filename:
names.append(filename)
dmgnames = []
for filename in (os.listdir(dmgpath)):
st.write('Daily metrics', daily)
st.markdown(
mu.get_table_download_link(cumulative,
filename='cumulative_' + local + '_' +
str(dt.date.today()) + '.csv'),
unsafe_allow_html=True)
st.write('Cumulative metrics', cumulative)
mu.append_row_2_logs([dt.datetime.today(), scope, local, model_beta],
'logs/fitted_models.csv')
scope = st.sidebar.selectbox('Country or US State', ['Country', 'State'],
index=0)
if scope == 'Country':
#data_load_state = st.text('Loading data...')
death_data = mu.get_data(scope='global', type='deaths')
#data_load_state.text('Loading data... done!')
local = st.sidebar.selectbox('Which country do you like to see prognosis',
death_data.Country.unique(),
index=156)
lockdown_date = st.sidebar.date_input(
'When did full lockdown happen? Very IMPORTANT to get accurate prediction',
mu.get_lockdown_date_by_country(local))
forecast_fun = mu.get_metrics_by_country
debug_fun = mu.get_log_daily_predicted_death_by_country
else:
#data_load_state = st.text('Loading data...')
death_data = mu.get_data(scope='US', type='deaths')
#data_load_state.text('Loading data... done!')
local = st.sidebar.selectbox('Which US state do you like to see prognosis',
death_data.State.unique(),
import fastai
from fastai.vision import ImageList, ImageImageList
from pathlib import Path
from model_utils import get_data
from model_utils import create_gen_learner
path = Path('../data/')
path_hr = path / 'preprocessed'
path_lr = path / 'processed'
path_test = path / 'test_imgs'
# Gather and select Data / Output size
bs, size = 1, 500
src = ImageImageList.from_folder(path_lr).split_by_rand_pct(0.1, seed=42)
data_gen = get_data(bs, size, src, path_hr)
# Load model to inference from
learn_gen = create_gen_learner(data_gen).load('') # Input model to load
# Open file to be inferenced
test_list = ImageList.from_folder(path_test)
test_list.open(test_list.items[0])
# Inference and display output
test_list[0].show(figsize=(7, 7),y=learn_gen.predict(test_list[0])[0])
# Save File if desired
#y = learn_gen.predict(test_list[8])[0]
#y.save(path_test/'inf1.png')