def get_results(dataset_name, coldstart, cs_type='none', n_entries=0):
# Get dataset
dataset = get_dataset(dataset_name)
models = []
# Hybrid Model
from hybrid_model.hybrid import HybridModel
from hybrid_model.config import hybrid_config
model_type = HybridModel
config = hybrid_config
models.append(EvalModel(model_type.__name__, model_type, config))
# Bias Baseline
from hybrid_model.models import BiasEstimator
model_type = BiasEstimator
config = {}
models.append(EvalModel(model_type.__name__, model_type, config))
# SVD
from hybrid_model.models import SVD
model_type = SVD
config = {}
models.append(EvalModel(model_type.__name__, model_type, config))
results = evaluate_models_xval(dataset,
models,
coldstart=coldstart,
cs_type=cs_type,
n_entries=n_entries)
return results
models = []
# Hybrid Model
from hybrid_model.hybrid import HybridModel
from hybrid_model.config import hybrid_config
model_type = HybridModel
config = hybrid_config
models.append(EvalModel(model_type.__name__, model_type, config))
evaluater = evaluation.Evaluation(evaluation.metrics_rmse_prec,
evaluation.get_parting_all(10))
results = evaluate_models_xval(dataset,
models,
coldstart=False,
evaluater=evaluater,
n_fold=10,
repeat=3)
print_results(results)
"""
------- HybridModel
=== Part full
rmse: 0.8915 ± 0.0062 prec@5: 0.8741 ± 0.0049
=== Part user_1
rmse: 0.8698 ± 0.0094 prec@5: 0.6756 ± 0.0214
=== Part user_2
rmse: 0.8683 ± 0.0107 prec@5: 0.7185 ± 0.0260
=== Part user_3
rmse: 0.8716 ± 0.0170 prec@5: 0.7824 ± 0.0129
=== Part user_4
# config = {}
# models.append(EvalModel(model_type.__name__, model_type, config))
# # SVDpp
# from hybrid_model.models import SVDpp
# model_type = SVDpp
# config = {}
# models.append(EvalModel(model_type.__name__, model_type, config))
#
# # SigmoidSVDpp
# from hybrid_model.models import SigmoidSVDpp
# model_type = SigmoidSVDpp
# config = {}
# models.append(EvalModel(model_type.__name__, model_type, config))
# SigmoidUserAsymFactoring
from hybrid_model.models import SigmoidUserAsymFactoring
model_type = SigmoidUserAsymFactoring
config = {}
models.append(EvalModel(model_type.__name__, model_type, config))
# SigmoidItemAsymFactoring
from hybrid_model.models import SigmoidItemAsymFactoring
model_type = SigmoidItemAsymFactoring
config = {}
models.append(EvalModel(model_type.__name__, model_type, config))
results = evaluate_models_xval(dataset, models, coldstart=False, cs_type='user', n_entries=0)
print_results(results)
model_type = HybridModel
config = hybrid_config
models.append(EvalModel(model_type.__name__, model_type, config))
params_entries = [0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60]
results = {}
rmses_before_cf = []
rmses_before_md = []
rmses_after_cf = []
rmses_after_md = []
for entries_plus in params_entries:
result = evaluate_models_xval(dataset,
models,
coldstart=True,
cs_type='item',
n_entries=entries_plus,
repeat=1)
results[entries_plus] = result
rmses_before_cf.append(result[0][1][0].mean_rmse_cf())
rmses_before_md.append(result[0][1][0].mean_rmse_md())
rmses_after_cf.append(result[0][1][1].mean_rmse_cf())
rmses_after_md.append(result[0][1][1].mean_rmse_md())
print('rmses_before_cf =', rmses_before_cf)
print('rmses_before_md =', rmses_before_md)
print('rmses_after_cf =', rmses_after_cf)
print('rmses_after_md =', rmses_after_md)
"""
from hybrid_model.dataset import get_dataset
dataset = get_dataset('ml100k')
models = []
# Hybrid Model
from hybrid_model.hybrid import HybridModel
from hybrid_model.config import hybrid_config
model_type = HybridModel
config = hybrid_config
models.append(EvalModel(model_type.__name__, model_type, config))
start = time.time()
results = evaluate_models_xval(dataset, models)
end = time.time()
elapsed = end - start
print('Elapsed time: {}s'.format(elapsed))
print_results(results)
"""
Elapsed time: 317.58788084983826s
------- HybridModel
=== Part full
rmse: 0.8959 ± 0.0025 prec@5: 0.7820 ± 0.0032
------- HybridModel_SVDpp
=== Part full
rmse: 0.9002 ± 0.0028 prec@5: 0.7774 ± 0.0021
# Get dataset
dataset = get_dataset('ml100k')
# dataset = get_dataset('ml1m')
models = []
# Hybrid Model
from hybrid_model.hybrid import HybridModel
from hybrid_model.config import hybrid_config
model_type = HybridModel
config = hybrid_config
models.append(EvalModel(model_type.__name__, model_type, config))
results = evaluate_models_xval(dataset, models, coldstart=False, n_fold=10)
print('Normal 10-fold')
print_results(results)
"""
Normal 10-fold
------- HybridModel
=== Part full
rmse: 0.8908 ± 0.0044 prec@5: 0.8756 ± 0.0035
------- HybridModel_SVDpp
=== Part full
rmse: 0.8953 ± 0.0048 prec@5: 0.8747 ± 0.0031
------- HybridModel_AttributeBiasExperimental
=== Part full
rmse: 0.9235 ± 0.0041 prec@5: 0.8633 ± 0.0043
model_type = HybridModel
config = hybrid_config
models.append(EvalModel(model_type.__name__, model_type, config))
params = [0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60]
results = {}
rmses_before_cf = []
rmses_before_md = []
rmses_after_cf = []
rmses_after_md = []
for entries_plus in params:
result = evaluate_models_xval(dataset,
models,
coldstart=True,
n_entries=entries_plus,
repeat=3)
results[entries_plus] = result
rmses_before_cf.append(result[0][1][0].mean_rmse_cf())
rmses_before_md.append(result[0][1][0].mean_rmse_md())
rmses_after_cf.append(result[0][1][1].mean_rmse_cf())
rmses_after_md.append(result[0][1][1].mean_rmse_md())
print('rmses_before_cf =', rmses_before_cf)
print('rmses_before_md =', rmses_before_md)
print('rmses_after_cf =', rmses_after_cf)
print('rmses_after_md =', rmses_after_md)
"""