from openrec import ImplicitModelTrainer
from openrec.utils import ImplicitDataset
from openrec.recommenders import VisualCML
from openrec.utils.evaluators import AUC, Recall
from openrec.utils.samplers import PairwiseSampler
from config import sess_config
import dataloader
raw_data = dataloader.load_amazon_book()
batch_size = 1000
test_batch_size = 100
item_serving_size = 1000
display_itr = 10000
train_dataset = ImplicitDataset(raw_data['train_data'],
raw_data['max_user'],
raw_data['max_item'],
name='Train')
val_dataset = ImplicitDataset(raw_data['val_data'],
raw_data['max_user'],
raw_data['max_item'],
name='Val')
test_dataset = ImplicitDataset(raw_data['test_data'],
raw_data['max_user'],
raw_data['max_item'],
name='Test')
model = VisualCML(batch_size=batch_size,
max_user=raw_data['max_user'],
max_item=raw_data['max_item'],
l2_reg=0.001,
l2_reg_mlp=0.001,
import pandas as pd
if __name__ == "__main__":
# We'll need to do a lot more wrangling/cleaning up - but this seems to AT LEAST WORK WITH DIFF DATA.
max_users = 10000
max_items = 200000
csv = np.genfromtxt('Movies_ratings_small_merged_larger2.csv',
delimiter=",",
dtype=(int, int, float, int, int, float, int, int,
float),
names=True,
encoding=None)
print(csv)
train_dataset = ImplicitDataset(raw_data=csv,
max_user=max_users,
max_item=max_items,
name='Train')
val_dataset = ImplicitDataset(raw_data=csv,
max_user=max_users,
max_item=max_items,
name='Val')
test_dataset = ImplicitDataset(raw_data=csv,
max_user=max_users,
max_item=max_items,
name='Test')
bpr_model = BPR(batch_size=1000,
max_user=train_dataset.max_user(),
max_item=train_dataset.max_item(),
dim_embed=20,
opt='Adam')
import os
import sys
sys.path.append(os.getcwd())
from openrec import ImplicitModelTrainer
from openrec.utils import ImplicitDataset
from openrec.recommenders import WCML
from openrec.utils.evaluators import AUC
from openrec.utils.samplers import NPairwiseSampler
import dataloader
raw_data = dataloader.load_citeulike()
batch_size = 2000
test_batch_size = 100
display_itr = 500
train_dataset = ImplicitDataset(raw_data['train_data'], raw_data['max_user'], raw_data['max_item'], name='Train')
val_dataset = ImplicitDataset(raw_data['val_data'], raw_data['max_user'], raw_data['max_item'], name='Val')
test_dataset = ImplicitDataset(raw_data['test_data'], raw_data['max_user'], raw_data['max_item'], name='Test')
model = WCML(batch_size=batch_size, max_user=train_dataset.max_user(), max_item=train_dataset.max_item(),
dim_embed=20, neg_num=5, l2_reg=None, opt='Adam', sess_config=None)
sampler = NPairwiseSampler(batch_size=batch_size, dataset=train_dataset, negativenum=5, num_process=5)
model_trainer = ImplicitModelTrainer(batch_size=batch_size, test_batch_size=test_batch_size,
train_dataset=train_dataset, model=model, sampler=sampler)
auc_evaluator = AUC()
model_trainer.train(num_itr=int(1e5), display_itr=display_itr, eval_datasets=[val_dataset],
evaluators=[auc_evaluator], num_negatives=200)
"""
def get_focused(dataset, focus_bound, interactions_per_item):
focused_indices = []
for i, interaction in enumerate(dataset):
user_id, item_id = interaction
if interactions_per_item[item_id] < focus_bound[
1] and interactions_per_item[item_id] >= focus_bound[0]:
focused_indices.append(i)
return np.take(dataset, focused_indices)
raw_data = get_raw_data(sys.argv[1])
val_dataset = ImplicitDataset(raw_data['val_data'],
raw_data['max_user'],
raw_data['max_item'],
name='Val')
test_dataset = ImplicitDataset(raw_data['test_data'],
raw_data['max_user'],
raw_data['max_item'],
name='Test')
#Getting Focused sets
full_dataset = np.concatenate(
(raw_data['train_data'], raw_data['val_data'], raw_data['test_data']),
axis=0)
interactions_per_item = get_item_interaction_dict(full_dataset)
#focus_bound, percent_focused = get_focus_bound(full_dataset, interactions_per_item)
focused_val_set = get_focused(raw_data['val_data'], focus_bound,
interactions_per_item)
focused_test_set = get_focused(raw_data['test_data'], focus_bound,
