def test_als_batch_accuracy():
from lenskit.algorithms import basic
import lenskit.crossfold as xf
import lenskit.metrics.predict as pm
ratings = lktu.ml100k.ratings
lu_algo = als.BiasedMF(25, iterations=20, damping=5, method='lu')
cd_algo = als.BiasedMF(25, iterations=25, damping=5, method='cd')
# algo = basic.Fallback(svd_algo, basic.Bias(damping=5))
def eval(train, test):
_log.info('training LU')
lu_algo.fit(train)
_log.info('training CD')
cd_algo.fit(train)
_log.info('testing %d users', test.user.nunique())
return test.assign(lu_pred=lu_algo.predict(test), cd_pred=cd_algo.predict(test))
folds = xf.partition_users(ratings, 5, xf.SampleFrac(0.2))
preds = pd.concat(eval(train, test) for (train, test) in folds)
preds['abs_diff'] = np.abs(preds.lu_pred - preds.cd_pred)
_log.info('predictions:\n%s', preds.sort_values('abs_diff', ascending=False))
_log.info('diff summary:\n%s', preds.abs_diff.describe())
lu_mae = pm.mae(preds.lu_pred, preds.rating)
assert lu_mae == approx(0.73, abs=0.025)
cd_mae = pm.mae(preds.cd_pred, preds.rating)
assert cd_mae == approx(0.73, abs=0.025)
user_rmse = preds.groupby('user').apply(lambda df: pm.rmse(df.lu_pred, df.rating))
assert user_rmse.mean() == approx(0.91, abs=0.05)
user_rmse = preds.groupby('user').apply(lambda df: pm.rmse(df.cd_pred, df.rating))
assert user_rmse.mean() == approx(0.91, abs=0.05)
def test_mae_series_two():
mae = pm.mae(pd.Series([1, 2]), pd.Series([1, 2]))
assert isinstance(mae, float)
assert mae == approx(0)
mae = pm.mae(pd.Series([1, 1]), pd.Series([2, 2]))
assert mae == approx(1)
mae = pm.mae(pd.Series([1, 3]), pd.Series([3, 1]))
assert mae == approx(2)
def test_mae_array_two():
mae = pm.mae(np.array([1, 2]), np.array([1, 2]))
assert isinstance(mae, float)
assert mae == approx(0)
mae = pm.mae(np.array([1, 1]), np.array([2, 2]))
assert mae == approx(1)
mae = pm.mae(np.array([1, 3]), np.array([3, 1]))
assert mae == approx(2)
def test_mae_two():
mae = pm.mae([1, 2], [1, 2])
assert isinstance(mae, float)
assert mae == approx(0)
mae = pm.mae([1, 1], [2, 2])
assert mae == approx(1)
mae = pm.mae([1, 3], [3, 1])
assert mae == approx(2)
mae = pm.mae([1, 3], [3, 2])
assert mae == approx(1.5)
def test_fsvd_batch_accuracy():
from lenskit.algorithms import basic
from lenskit.algorithms import bias
import lenskit.crossfold as xf
from lenskit import batch
import lenskit.metrics.predict as pm
ratings = lktu.ml100k.ratings
svd_algo = svd.FunkSVD(25, 125, damping=10)
algo = basic.Fallback(svd_algo, bias.Bias(damping=10))
def eval(train, test):
_log.info('running training')
algo.fit(train)
_log.info('testing %d users', test.user.nunique())
return batch.predict(algo, test)
folds = xf.partition_users(ratings, 5, xf.SampleFrac(0.2))
preds = pd.concat(eval(train, test) for (train, test) in folds)
mae = pm.mae(preds.prediction, preds.rating)
assert mae == approx(0.74, abs=0.025)
user_rmse = preds.groupby('user').apply(lambda df: pm.rmse(df.prediction, df.rating))
assert user_rmse.mean() == approx(0.92, abs=0.05)
def test_ii_batch_accuracy():
from lenskit.algorithms import basic
import lenskit.crossfold as xf
from lenskit import batch
import lenskit.metrics.predict as pm
ratings = lktu.ml100k.ratings
ii_algo = knn.ItemItem(30)
algo = basic.Fallback(ii_algo, basic.Bias())
def eval(train, test):
_log.info('running training')
algo.fit(train)
_log.info('testing %d users', test.user.nunique())
return batch.predict(algo, test, n_jobs=4)
preds = pd.concat((eval(train, test)
for (train, test)
in xf.partition_users(ratings, 5, xf.SampleFrac(0.2))))
mae = pm.mae(preds.prediction, preds.rating)
assert mae == approx(0.70, abs=0.025)
user_rmse = preds.groupby('user').apply(lambda df: pm.rmse(df.prediction, df.rating))
assert user_rmse.mean() == approx(0.90, abs=0.05)
def test_tf_bmf_batch_accuracy(tf_session):
from lenskit.algorithms import basic
from lenskit.algorithms import bias
import lenskit.crossfold as xf
from lenskit import batch
import lenskit.metrics.predict as pm
ratings = lktu.ml100k.ratings
algo = lktf.BiasedMF(25,
damping=10,
batch_size=1024,
epochs=20,
rng_spec=42)
algo = basic.Fallback(algo, bias.Bias(damping=10))
def eval(train, test):
_log.info('running training')
algo.fit(train)
_log.info('testing %d users', test.user.nunique())
return batch.predict(algo, test)
folds = xf.partition_users(ratings, 5, xf.SampleFrac(0.2))
preds = pd.concat(eval(train, test) for (train, test) in folds)
mae = pm.mae(preds.prediction, preds.rating)
assert mae == approx(0.83, abs=0.025)
user_rmse = preds.groupby('user').apply(
lambda df: pm.rmse(df.prediction, df.rating))
assert user_rmse.mean() == approx(1.03, abs=0.05)
def test_tf_isvd(ml20m):
algo = lenskit_tf.IntegratedBiasMF(20)
def eval(train, test):
_log.info('running training')
algo.fit(train)
_log.info('testing %d users', test.user.nunique())
return batch.predict(algo, test)
folds = xf.sample_users(ml20m, 2, 5000, xf.SampleFrac(0.2))
preds = pd.concat(eval(train, test) for (train, test) in folds)
mae = pm.mae(preds.prediction, preds.rating)
assert mae == approx(0.60, abs=0.025)
user_rmse = preds.groupby('user').apply(
lambda df: pm.rmse(df.prediction, df.rating))
assert user_rmse.mean() == approx(0.92, abs=0.05)
def test_global_metric():
import lenskit.crossfold as xf
import lenskit.batch as batch
from lenskit.algorithms.bias import Bias
train, test = next(
xf.sample_users(lktu.ml_test.ratings, 1, 200, xf.SampleFrac(0.5)))
algo = Bias()
algo.fit(train)
preds = batch.predict(algo, test)
rmse = pm.global_metric(preds)
assert rmse == pm.rmse(preds.prediction, preds.rating)
mae = pm.global_metric(preds, metric=pm.mae)
assert mae == pm.mae(preds.prediction, preds.rating)
def test_uu_batch_accuracy():
from lenskit.algorithms import basic
import lenskit.crossfold as xf
import lenskit.metrics.predict as pm
ratings = lktu.ml100k.ratings
uu_algo = knn.UserUser(30)
algo = basic.Fallback(uu_algo, basic.Bias())
folds = xf.partition_users(ratings, 5, xf.SampleFrac(0.2))
preds = [__batch_eval((algo, train, test)) for (train, test) in folds]
preds = pd.concat(preds)
mae = pm.mae(preds.prediction, preds.rating)
assert mae == approx(0.71, abs=0.028)
user_rmse = preds.groupby('user').apply(lambda df: pm.rmse(df.prediction, df.rating))
assert user_rmse.mean() == approx(0.91, abs=0.055)
def test_user_metric():
import lenskit.crossfold as xf
import lenskit.batch as batch
from lenskit.algorithms.bias import Bias
train, test = next(
xf.sample_users(lktu.ml_test.ratings, 1, 200, xf.SampleFrac(0.5)))
algo = Bias()
algo.fit(train)
preds = batch.predict(algo, test)
rmse = pm.user_metric(preds)
u_rmse = preds.groupby('user').apply(
lambda df: pm.rmse(df.prediction, df.rating))
assert rmse == approx(u_rmse.mean())
mae = pm.user_metric(preds, metric=pm.mae)
u_mae = preds.groupby('user').apply(
lambda df: pm.mae(df.prediction, df.rating))
assert mae == approx(u_mae.mean())
def test_als_batch_accuracy():
from lenskit.algorithms import basic
import lenskit.crossfold as xf
import lenskit.metrics.predict as pm
ratings = lktu.ml100k.load_ratings()
svd_algo = als.BiasedMF(25, iterations=20, damping=5)
algo = basic.Fallback(svd_algo, basic.Bias(damping=5))
def eval(train, test):
_log.info('running training')
algo.fit(train)
_log.info('testing %d users', test.user.nunique())
return test.assign(prediction=algo.predict(test))
folds = xf.partition_users(ratings, 5, xf.SampleFrac(0.2))
preds = pd.concat(eval(train, test) for (train, test) in folds)
mae = pm.mae(preds.prediction, preds.rating)
assert mae == approx(0.73, abs=0.025)
user_rmse = preds.groupby('user').apply(lambda df: pm.rmse(df.prediction, df.rating))
assert user_rmse.mean() == approx(0.91, abs=0.05)