'item_id'] = df_times_in_impressions.index.astype(int)
df_times_in_impressions = df_times_in_impressions.reindex(
columns=['item_id', 'num_times_item_impressed'])
df_times_in_impressions = df_times_in_impressions.sort_values(
by=['item_id']).reset_index(drop=True)
feature = pd.merge(clk_expanded,
df_times_in_impressions,
how='left',
on=['item_id']).fillna(0)
feature.num_times_item_impressed = feature.num_times_item_impressed.astype(
int)
return feature[[
'user_id', 'session_id', 'item_id', 'num_times_item_impressed'
]]
if __name__ == '__main__':
import utils.menu as menu
mode = menu.mode_selection()
cluster = menu.cluster_selection()
c = NumTimesItemImpressed(mode, cluster)
print('Creating {} for {} {}'.format(c.name, c.mode, c.cluster))
c.save_feature()
print(c.read_feature())
actions_involving_impression_session_no_action))
count += 1
return r
train = data.train_df(mode=self.mode, cluster=self.cluster)
test = data.test_df(mode=self.mode, cluster=self.cluster)
df = pd.concat([train, test])
df = df.drop(['timestamp', 'step', 'platform', 'city', 'device', 'current_filters', 'prices'], axis=1)
s = df.groupby(
['user_id', 'session_id']).progress_apply(func)
s = s.apply(pd.Series).reset_index().melt(id_vars=['user_id', 'session_id'], value_name='tuple').sort_values(
by=['user_id', 'session_id']).dropna()
s[['item_id', 'actions_involving_impression_session_clickout_item',
'actions_involving_impression_session_interaction_item_deals',
'actions_involving_impression_session_interaction_item_image',
'actions_involving_impression_session_interaction_item_info',
'actions_involving_impression_session_interaction_item_rating',
'actions_involving_impression_session_search_for_item',
'actions_involving_impression_session_no_action']] = pd.DataFrame(s['tuple'].tolist(), index=s.index)
s = s.drop(['variable', 'tuple'], axis=1)
s = s.reset_index(drop=True)
return s
if __name__ == '__main__':
from utils.menu import mode_selection, cluster_selection
cluster = cluster_selection()
mode = mode_selection()
c = ActionsInvolvingImpressionSession(mode=mode, cluster=cluster)
c.save_feature()
cluster='no_cluster',
dataset_name=flags_dict['dataset_name'],
pred_name=name)
model.name = f'tf_ranking_{name}'
model.run()
HERA.send_message('EXPORTED... mode:{}, name:{}'.format(
flags_dict['mode'], name))
if __name__ == '__main__':
global flags_dict
flags_dict = {}
# let the user insert the mode cluster and dataset name
_MODE = menu.mode_selection()
_CLUSTER = menu.cluster_selection()
_DATASET_NAME = input('insert dataset name: \n')
_BASE_PATH = f'dataset/preprocessed/tf_ranking/{_CLUSTER}/{_MODE}/{_DATASET_NAME}'
_TRAIN_PATH = f'{_BASE_PATH}/train.hdf'
_TEST_PATH = f'{_BASE_PATH}/test.hdf'
_VALI_PATH = f'{_BASE_PATH}/vali.hdf'
min_mrr = float(input('insert the min_MRR from which export the sub: \n'))
# lets load the context features id
context_features_id = list(
np.load(f'{_BASE_PATH}/context_features_id.npy'))
print(f'context features id are: {context_features_id}')
flags_dict['context_features_id'] = context_features_id
