def test_partial_fit_regression(self):
data = make_regression(100, 2, 1)
df = pd.DataFrame(data[0])
df["label"] = data[1]
with df.model() as m:
fit_partial = m.fit(
SkModel(
MLPRegressor(max_iter=1, random_state=42),
FeaturesAndLabels(features=[0, 1], labels=['label'])
),
FittingParameter(
naive_splitter(0.3),
batch_size=10,
fold_epochs=10
)
)
with df.model() as m:
fit = m.fit(
SkModel(
MLPRegressor(max_iter=10, random_state=42),
FeaturesAndLabels(features=[0, 1], labels=['label'])
),
FittingParameter(naive_splitter(0.3))
)
self.assertAlmostEqual(df.model.predict(fit.model).iloc[0,-1], df.model.predict(fit_partial.model).iloc[0,-1], 4)
def provide_linear_regression_model(self):
from sklearn.linear_model import LinearRegression
from sklearn.neural_network import MLPRegressor
from pandas_ml_utils import FeaturesAndLabels, SkModel
return [
(
SkModel(LinearRegression(), FeaturesAndLabels(["x"], ["y"])),
FittingParameter(epochs=1, fold_epochs=1, context="LinearRegression")
),
(
SkModel(
MLPRegressor(10, learning_rate_init=0.01, max_iter=9000, validation_fraction=0),
FeaturesAndLabels(["x"], ["y"])
),
FittingParameter(epochs=1, fold_epochs=1, context="MLPRegressor")
),
(
SkModel(
MLPRegressor(10, learning_rate_init=0.01, max_iter=1, validation_fraction=0, warm_start=True),
FeaturesAndLabels(["x"], ["y"])
),
FittingParameter(epochs=9000, fold_epochs=1, context="MLPRegressor partial fit")
)
]
def test_partial_fit_classification(self):
data = make_classification(100, 2, 1, 0, n_clusters_per_class=1)
df = pd.DataFrame(data[0])
df["label"] = data[1]
with df.model() as m:
fit_partial = m.fit(
SkModel(
MLPClassifier(max_iter=1, random_state=42),
FeaturesAndLabels(features=[0, 1], labels=['label']),
classes=np.unique(data[1])
),
FittingParameter(
stratified_random_splitter(0.3),
batch_size=10,
fold_epochs=10,
)
)
with df.model() as m:
fit = m.fit(
SkModel(
MLPClassifier(max_iter=10, random_state=42),
FeaturesAndLabels(features=[0, 1], labels=['label'])
),
FittingParameter(stratified_random_splitter(0.3))
)
self.assertAlmostEqual(df.model.predict(fit.model).iloc[0,-1], df.model.predict(fit_partial.model).iloc[0,-1], 4)
def test_simple_regression_model(self):
df = DF_TEST.copy()
fit = df.model.fit(
SkModel(
MLPRegressor(activation='tanh',
hidden_layer_sizes=(60, 50),
random_state=42,
max_iter=2),
FeaturesAndLabels(features=[
lambda df: df["Close"].ta.rsi().ta.rnn(28), lambda df:
(df["Volume"] / df["Volume"].ta.ema(14) - 1).ta.rnn(28)
],
labels=[
lambda df:
(df["Close"] / df["Open"] - 1).shift(-1),
]),
summary_provider=RegressionSummary), FittingParameter())
print(fit)
html = fit._repr_html_()
prediction = df.model.predict(fit.model)
print(prediction)
self.assertIsInstance(prediction[PREDICTION_COLUMN_NAME, 0].iloc[-1],
(float, np.float, np.float32, np.float64))
backtest = df.model.backtest(fit.model)
def test_simple_classification_model(self):
df = DF_NOTES.copy()
with df.model() as m:
fit = m.fit(
SkModel(
MLPClassifier(activation='tanh', hidden_layer_sizes=(20, 12), random_state=42, max_iter=2),
FeaturesAndLabels(
features=["variance", "skewness", "kurtosis", "entropy"],
labels=["authentic"],
label_type=bool
)
),
FittingParameter(stratified_random_splitter())
)
print(fit)
html = fit._repr_html_()
prediction = df.model.predict(fit.model)
print(prediction)
self.assertGreaterEqual(prediction[PREDICTION_COLUMN_NAME].iloc[-1].values, 0.68)
backtest = df.model.backtest(fit.model)
self.assertIn(FEATURE_COLUMN_NAME, backtest.df)
self.assertIn(LABEL_COLUMN_NAME, backtest.df)
np.testing.assert_array_almost_equal(prediction[PREDICTION_COLUMN_NAME].iloc[-1].values,
backtest.df[PREDICTION_COLUMN_NAME].iloc[-1].values)
# test multiple samples
samples = df.model.predict(fit.model, samples=2)
self.assertIsInstance(samples[PREDICTION_COLUMN_NAME].iloc[-1, 0], list)
self.assertEqual(2, len(samples[PREDICTION_COLUMN_NAME].iloc[-1, 0]))
def test_regressor(self):
"""given some toy regression data"""
df = pd.DataFrame({
"a": [-1.0, 0.0, 1.0, 2.0, 3.0, 4.0],
"b": [-2.0, 1.0, 4.0, 7.0, 10.0, 13.0]
})
"""and a model"""
model = self.provide_regression_model(
FeaturesAndLabels(features=["a"], labels=["b"]))
"""when we fit the model"""
batch_size, epochs = self.provide_batch_size_and_epoch()
with df.model() as m:
fit = m.fit(model,
FittingParameter(splitter=naive_splitter(0.3),
batch_size=batch_size,
epochs=epochs),
verbose=0)
print(fit.training_summary.df)
self.assertEqual(4, len(fit.training_summary.df))
self.assertEqual(2, len(fit.test_summary.df))
"""then we can predict"""
prediction = df.model.predict(fit.model)
np.testing.assert_array_almost_equal(prediction.iloc[:, 0].values,
df["b"].values, 1)
"""and save and load the model"""
temp = os.path.join(tempfile.gettempdir(), str(uuid.uuid4()))
try:
fit.model.save(temp)
copy = Model.load(temp)
pd.testing.assert_frame_equal(df.model.predict(fit.model),
df.model.predict(copy),
check_less_precise=True)
finally:
os.remove(temp)
def test_multindex_row_multi_samples(self):
"""given some toy regression data while we provide a multiindex for the rows"""
df = pd.DataFrame(
{
"a":
[-1.0, 0.0, 1.0, 2.0, 3.0, 4.0, -1.0, 0.0, 1.0, 2.0, 3.0, 4.0],
"b": [
-2.0, 1.0, 4.0, 7.0, 10.0, 13.0, -2.0, 1.0, 4.0, 7.0, 10.0,
13.0
]
},
index=pd.MultiIndex.from_product([["A", "B"], range(6)]))
"""and a model"""
model = self.provide_regression_model(
FeaturesAndLabels(features=["a"], labels=["b"]))
"""when we fit the model"""
batch_size, epochs = self.provide_batch_size_and_epoch()
with df.model() as m:
fit = m.fit(model,
FittingParameter(splitter=random_splitter(
0.3, partition_row_multi_index=True),
batch_size=batch_size,
epochs=epochs),
verbose=0)
self.assertEqual(8, len(fit.training_summary.df))
self.assertEqual(4, len(fit.test_summary.df))
prediction = df.model.predict(fit.model, samples=2)
self.assertEqual(2, len(prediction.iloc[:, 0]._.values))
self.assertEqual((6, 2), prediction.loc["A"].iloc[:, 0]._.values.shape)
self.assertEqual((6, 2), prediction.loc["B"].iloc[:, 0]._.values.shape)
def test_simple_classification_cross_validation(self):
df = DF_NOTES.copy()
with df.model() as m:
fit = m.fit(
SkModel(
MLPClassifier(activation='tanh', hidden_layer_sizes=(20, 12), random_state=42, max_iter=2),
FeaturesAndLabels(
features=["variance", "skewness", "kurtosis", "entropy"],
labels=["authentic"],
label_type=bool
)
),
FittingParameter(
splitter=random_splitter(),
cross_validation=KFold(3, random_state=42, shuffle=True)
)
)
print(fit)
html = fit._repr_html_()
prediction = df.model.predict(fit.model)
print(prediction)
self.assertGreaterEqual(prediction[PREDICTION_COLUMN_NAME].iloc[-1].values, 0.65)
def test_regularized_loss(self):
df = pd.DataFrame({
"f": np.sin(np.linspace(0, 12, 40)),
"l": np.sin(np.linspace(5, 17, 40))
})
class TestModel(PytorchNN):
def __init__(self):
super().__init__()
self.net = nn.Sequential(nn.Linear(1, 3), nn.ReLU(),
nn.Linear(3, 2), nn.ReLU(),
nn.Linear(2, 1), nn.Sigmoid())
def forward_training(self, x):
return self.net(x)
def L2(self) -> Dict[str, float]:
return {'**/2/**/weight': 99999999999.99}
fit = df.model.fit(
PytorchModel(TestModel, FeaturesAndLabels(["f"], ["l"]),
nn.MSELoss, Adam),
FittingParameter(epochs=1000, splitter=naive_splitter(0.5)))
print(fit.model._current_model.net.net[2].weight.detach().numpy())
print(
fit.model._current_model.net.net[2].weight.norm().detach().item())
self.assertLess(
fit.model._current_model.net.net[2].weight.norm().detach().item(),
0.1)
def test_linear_model(self):
df = DF_NOTES.copy()
with df.model() as m:
fit = m.fit(
SkModel(
Lasso(),
FeaturesAndLabels(
features=[
lambda df: df["variance"],
lambda df: (df["skewness"] / df["kurtosis"]).rename("engineered")
],
labels=[
'authentic'
]
)
),
FittingParameter(naive_splitter())
)
print(fit)
prediction = df.model.predict(fit.model)
print(prediction)
backtest = df.model.backtest(fit.model)
self.assertLess(backtest.model.sk_model.coef_[0], 1e-5)
def test_make_model(self):
notebooks_path = os.path.join(PWD, '..', 'examples')
df = pd.read_csv(os.path.join(notebooks_path, 'SPY.csv'))
with df.model("/tmp/pijsfnwuacpa.model") as m:
from torch import nn
from torch.optim import SGD
from pandas_ml_common.utils.column_lagging_utils import lag_columns
from pandas_ml_utils import FeaturesAndLabels, RegressionSummary, FittingParameter
from pandas_ml_utils_torch import PytorchModel
from pandas_ml_utils_torch.merging_cross_folds import take_the_best
def net_provider():
from pandas_ml_utils_torch import PytorchNN
class Net(PytorchNN):
def __init__(self):
super().__init__()
self.net = nn.Sequential(
nn.Linear(10, 4),
nn.Tanh(),
nn.Linear(4, 4),
nn.Tanh(),
nn.Linear(4, 1),
nn.Tanh(),
)
def L1(self):
# path to the parameters which should be regularized
# the path is constructed from self.named_parameters() and allows the use of wildcards
return {'net/0/**/weight': 0.02}
def L2(self):
return {
'net/0/**/weight': 0.02,
'net/2/**/weight': 0.05
}
def forward_training(self, x):
return self.net(x)
return Net()
fit = m.fit(
PytorchModel(
net_provider,
FeaturesAndLabels(
[lambda df: lag_columns(df["Close"].pct_change(), range(10))],
[lambda df: df["Close"].pct_change().shift(-1)]),
nn.MSELoss,
lambda params: SGD(params, lr=0.01, momentum=0.0),
merge_cross_folds=take_the_best,
summary_provider=RegressionSummary
),
FittingParameter(epochs=2),
verbose=1
)
def provide_non_linear_regression_model(self):
from sklearn.neural_network import MLPRegressor
from pandas_ml_utils import FeaturesAndLabels, SkModel
return [
(
SkModel(
MLPRegressor(200, learning_rate_init=0.001, max_iter=5000, validation_fraction=0),
FeaturesAndLabels(["x"], ["y"])
),
FittingParameter(epochs=1, context="epoch 1 fit"),
),
(
SkModel(
MLPRegressor(200, learning_rate_init=0.001, max_iter=1, validation_fraction=0, warm_start=True),
FeaturesAndLabels(["x"], ["y"])
),
FittingParameter(epochs=5000, context="partial fit"),
)
]
def provide_non_linear_regression_model(self):
from pandas_ml_utils_torch import PytorchModel, PytorchNN
from pandas_ml_utils import FeaturesAndLabels
from torch.optim import Adagrad
from torch import nn
import torch as t
# t.manual_seed(0)
class Net(PytorchNN):
def __init__(self):
super().__init__()
self.net = nn.Sequential(nn.Linear(1, 200), nn.ReLU(),
nn.Linear(200, 200), nn.ReLU(),
nn.Linear(200, 200), nn.ReLU(),
nn.Linear(200, 1), nn.ReLU())
def forward_training(self, *input) -> t.Tensor:
return self.net(input[0])
t.manual_seed(0)
model = PytorchModel(Net, FeaturesAndLabels(["x"], ["y"]), nn.MSELoss,
Adagrad)
return [(
model,
FittingParameter(epochs=600,
batch_size=64,
context="epoch fit batched"),
), (
model,
FittingParameter(epochs=600, context="epoch fit"),
),
(
model,
FittingParameter(epochs=1,
fold_epochs=600,
context="fold epoch fit"),
)]
def test_probabilistic(self):
def create_sine_data(n=300):
np.random.seed(32)
n = 300
x = np.linspace(0, 1 * 2 * np.pi, n)
y1 = 3 * np.sin(x)
y1 = np.concatenate(
(np.zeros(60), y1 + np.random.normal(0, 0.15 * np.abs(y1), n),
np.zeros(60)))
x = np.concatenate(
(np.linspace(-3, 0, 60), np.linspace(0, 3 * 2 * np.pi, n),
np.linspace(3 * 2 * np.pi, 3 * 2 * np.pi + 3, 60)))
y2 = 0.1 * x + 1
y = y1 + y2
return x, y
df = pd.DataFrame(np.array(create_sine_data(300)).T,
columns=["x", "y"])
with df.model() as m:
from pandas_ml_utils import FeaturesAndLabels
from pandas_ml_utils_torch import PytorchNN, PytorchModel
from pandas_ml_utils_torch.loss import HeteroscedasticityLoss
from pandas_ml_common.sampling.splitter import duplicate_data
from torch.optim import Adam
from torch import nn
class Net(PytorchNN):
def __init__(self):
super().__init__()
self.l = nn.Sequential(
nn.Linear(1, 20),
nn.ReLU(),
nn.Linear(20, 50),
nn.ReLU(),
nn.Linear(50, 20),
nn.ReLU(),
nn.Linear(20, 2),
)
def forward_training(self, x):
return self.l(x)
fit = m.fit(
PytorchModel(Net,
FeaturesAndLabels(["x"], ["y"]),
HeteroscedasticityLoss,
Adam,
restore_best_weights=True),
FittingParameter(batch_size=128,
epochs=10,
splitter=duplicate_data()))
def provide_linear_regression_model(self):
from pandas_ml_utils_torch import PytorchModel, PytorchNN
from pandas_ml_utils import FeaturesAndLabels
from torch.optim import Adam
from torch import nn
import torch as t
class Net(PytorchNN):
def __init__(self):
super(Net, self).__init__()
self.net = nn.Linear(1, 1)
def forward_training(self, *input) -> t.Tensor:
return self.net(input[0])
return [
(
PytorchModel(Net, FeaturesAndLabels(["x"], ["y"]), nn.MSELoss,
Adam),
FittingParameter(epochs=5000, context="epoch fit"),
),
(
PytorchModel(Net, FeaturesAndLabels(["x"], ["y"]), nn.MSELoss,
Adam),
FittingParameter(epochs=5000,
batch_size=64,
context="epoch fit batched"),
),
(
PytorchModel(Net, FeaturesAndLabels(["x"], ["y"]), nn.MSELoss,
Adam),
FittingParameter(epochs=1,
fold_epochs=5000,
context="fold epoch fit"),
),
]
def test_simple_classification_model_with_all_options(self):
df = DF_NOTES.copy()
with df.model() as m:
fit = m.fit(
SkModel(
MLPClassifier(activation='tanh', hidden_layer_sizes=(20, 12), random_state=42, max_iter=2),
FeaturesAndLabels(
features=["variance", "skewness", "kurtosis", "entropy"],
sample_weights=["variance"],
gross_loss=["kurtosis"],
targets=["entropy"],
labels=["authentic"],
label_type=bool
)
),
FittingParameter(stratified_random_splitter())
)
# should not thro an error
html = fit._repr_html_()
# fit resulting columns
print(fit.test_summary.df)
self.assertIn(GROSS_LOSS_COLUMN_NAME, fit.training_summary.df)
self.assertIn(FEATURE_COLUMN_NAME, fit.training_summary.df)
self.assertIn(LABEL_COLUMN_NAME, fit.training_summary.df)
self.assertIn(TARGET_COLUMN_NAME, fit.training_summary.df)
self.assertIn(FEATURE_COLUMN_NAME, fit.test_summary.df)
self.assertIn(LABEL_COLUMN_NAME, fit.test_summary.df)
self.assertIn(TARGET_COLUMN_NAME, fit.test_summary.df)
# prediction resulting columns
prediction = df.model.predict(fit.model)
print(prediction)
self.assertIn(FEATURE_COLUMN_NAME, prediction)
self.assertIn(TARGET_COLUMN_NAME, prediction)
# backtest resulting columns
backtest = df.model.backtest(fit.model)
print(backtest.df)
self.assertEqual(len(df), len(backtest.df))
self.assertIn(FEATURE_COLUMN_NAME, backtest.df)
self.assertIn(LABEL_COLUMN_NAME, backtest.df)
self.assertIn(TARGET_COLUMN_NAME, backtest.df)
self.assertIn(GROSS_LOSS_COLUMN_NAME, backtest.df)
def test_mult_epoch_cross_validation(self):
df = pd.DataFrame({
"a": [
1,
0,
1,
0,
1,
0,
1,
0,
],
"b": [
0,
1,
0,
1,
1,
0,
1,
0,
],
})
with df.model() as m:
class NN(PytorchNN):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.nn = nn.Sequential(
nn.Linear(1, 2),
nn.ReLU(),
nn.Linear(2, 1),
)
def forward_training(self, x):
return self.nn(x)
fit = m.fit(
PytorchModel(NN, FeaturesAndLabels(["a"], ["b"]), nn.MSELoss,
Adam),
FittingParameter(splitter=naive_splitter(0.5),
epochs=2,
fold_epochs=10,
batch_size=2))
print(fit)
def test_multindex_row(self):
"""given some toy regression data while we provide a multiindex for the rows"""
df = pd.DataFrame(
{
"a":
[-1.0, 0.0, 1.0, 2.0, 3.0, 4.0, -1.0, 0.0, 1.0, 2.0, 3.0, 4.0],
"b": [
-2.0, 1.0, 4.0, 7.0, 10.0, 13.0, -2.0, 1.0, 4.0, 7.0, 10.0,
13.0
]
},
index=pd.MultiIndex.from_product([["A", "B"], range(6)]))
"""and a model"""
model = self.provide_regression_model(
FeaturesAndLabels(features=["a"], labels=["b"]))
"""when we fit the model"""
batch_size, epochs = self.provide_batch_size_and_epoch()
with df.model() as m:
fit = m.fit(model,
FittingParameter(splitter=random_splitter(
0.3, partition_row_multi_index=True),
batch_size=batch_size,
epochs=epochs),
verbose=0)
prediction = df.model.predict(fit.model)
print(fit)
# fit.training_summary.df.to_pickle('/tmp/multi_index_row_summary.df')
# print(fit._repr_html_())
"""then we get a prediction for A and B rows"""
self.assertEqual(8, len(fit.training_summary.df))
self.assertEqual(4, len(fit.training_summary.df.loc["A"]))
self.assertEqual(4, len(fit.training_summary.df.loc["B"]))
self.assertEqual(4, len(fit.test_summary.df))
self.assertEqual(2, len(fit.test_summary.df.loc["A"]))
self.assertEqual(2, len(fit.test_summary.df.loc["B"]))
self.assertEqual(6, len(prediction.loc["A"]))
self.assertEqual(6, len(prediction.loc["B"]))
np.testing.assert_array_almost_equal(prediction.iloc[:, 0].values,
df["b"].values, 1)
def test_multi_objective_loss(self):
df = pd.DataFrame(
np.array([
# train
[0, 0, 0],
[0, 1, 1],
[1, 0, 1],
[1, 1, 1],
# test
[0, 0, 0],
[0, 1, 1],
[1, 0, 1],
[1, 1, 1],
]),
columns=["f1", "f2", "l"])
class XorModule(PytorchNN):
def __init__(self):
super().__init__()
self.x1 = nn.Linear(2, 1)
self.s1 = nn.Sigmoid()
self.x2 = nn.Linear(2, 1)
self.s2 = nn.Sigmoid()
self.s = nn.Softmax()
def forward_training(self, x):
return self.s1(self.x1(x)), self.s2(self.x2(x))
def forward_predict(self, x):
return self.s1(self.x1(x))
fit = df.model.fit(
PytorchModel(
XorModule, FeaturesAndLabels(["f1", "f2"], ["l"]),
lambda: MultiObjectiveLoss(
(1, nn.MSELoss(reduction='none')),
(1, nn.L1Loss(reduction='none')),
on_epoch=lambda criterion, epoch: criterion.update_weights(
(0, 1.1))), Adam),
FittingParameter(splitter=naive_splitter(0.5)))
print(fit.test_summary.df)
def test_no_test_data(self):
"""given some toy regression data"""
df = pd.DataFrame({
"a": [-1.0, 0.0, 1.0, 2.0, 3.0, 4.0],
"b": [-2.0, 1.0, 4.0, 7.0, 10.0, 13.0]
})
"""and a model"""
model = self.provide_regression_model(
FeaturesAndLabels(features=["a"], labels=["b"]))
"""when we fit the model"""
batch_size, epochs = self.provide_batch_size_and_epoch()
with df.model() as m:
fit = m.fit(model,
FittingParameter(splitter=naive_splitter(0),
batch_size=batch_size,
epochs=epochs),
verbose=0)
# print(fit.training_summary.df)
print(fit.test_summary.df)
"""then we have an empty test data frame"""
self.assertEqual(len(fit.training_summary.df), len(df))
self.assertEqual(len(fit.test_summary.df), 0)
def test_multi_sample_regressor(self):
"""given some toy regression data"""
df = pd.DataFrame({
"a": [-1.0, 0.0, 1.0, 2.0, 3.0, 4.0],
"b": [-2.0, 1.0, 4.0, 7.0, 10.0, 13.0]
})
"""and a model"""
model = self.provide_regression_model(
FeaturesAndLabels(features=["a"], labels=["b"]))
"""when we fit the model"""
batch_size, epochs = self.provide_batch_size_and_epoch()
with df.model() as m:
fit = m.fit(model,
FittingParameter(splitter=naive_splitter(0.3),
batch_size=batch_size,
epochs=epochs),
verbose=0)
print(fit.training_summary.df)
"""then we can predict"""
prediction = df.model.predict(fit.model, samples=2)
np.testing.assert_array_almost_equal(
prediction.iloc[:, 0]._.values,
np.concatenate([df[["b"]].values, df[["b"]].values], axis=1), 1)
def test_auto_encoder(self):
"""given the implementation can handle auto encoders"""
model = self.provide_auto_encoder_model(
FeaturesAndLabels(features=["a", "b"],
labels=["a", "b"],
latent=["x"]))
if model is None:
return
"""and some toy classification data"""
df = pd.DataFrame({
"a": [1, 0] * 10,
"b": [0, 1] * 10,
})
"""when we fit the model"""
batch_size, epochs = self.provide_batch_size_and_epoch()
with df.model() as m:
fit = m.fit(model,
FittingParameter(splitter=naive_splitter(0.49),
batch_size=batch_size,
epochs=epochs),
verbose=0)
print(fit.training_summary.df)
"""then we can predict Autoencoded"""
auto_encoded_prediction = df.model.predict(fit.model)
self.assertEqual((20, 2), auto_encoded_prediction["prediction"].shape)
"""and we can encode"""
encoded_prediction = df.model.predict(fit.model.as_encoder())
print(encoded_prediction)
self.assertEqual((20, 1), encoded_prediction["prediction"].shape)
"""and we can decode"""
decoded_prediction = encoded_prediction["prediction"].model.predict(
fit.model.as_decoder())
print(decoded_prediction)
np.testing.assert_array_almost_equal(
decoded_prediction["prediction"].values > 0.5, df[["a",
"b"]].values)
"""and we can encoder and decode after safe and load"""
temp = os.path.join(tempfile.gettempdir(), str(uuid.uuid4()))
try:
fit.model.save(temp)
copy = Model.load(temp)
pd.testing.assert_frame_equal(df.model.predict(
fit.model.as_encoder()),
df.model.predict(copy.as_encoder()),
check_less_precise=True)
pd.testing.assert_frame_equal(
encoded_prediction.model.predict(fit.model.as_decoder()),
encoded_prediction.model.predict(copy.as_decoder()),
check_less_precise=True)
finally:
os.remove(temp)
# try to save only as encoder model
try:
fit.model.as_encoder().save(temp)
copy = Model.load(temp)
finally:
os.remove(temp)
def test_classifier(self):
"""given some toy classification data"""
df = pd.DataFrame({
"a": [
1,
0,
1,
0,
1,
0,
1,
0,
],
"b": [
0,
0,
1,
1,
0,
0,
1,
1,
],
"c": [
1,
0,
0,
1,
1,
0,
0,
1,
]
})
"""and a model"""
model = self.provide_classification_model(
FeaturesAndLabels(features=["a", "b"],
labels=["c"],
label_type=int))
temp = os.path.join(tempfile.gettempdir(), str(uuid.uuid4()))
"""when we fit the model"""
batch_size, epochs = self.provide_batch_size_and_epoch()
with df.model(temp) as m:
fit = m.fit(model,
FittingParameter(splitter=naive_splitter(0.49),
batch_size=batch_size,
epochs=epochs),
verbose=0)
print(fit.training_summary.df)
# fit.training_summary.df.to_pickle('/tmp/classifier.df')
# print(fit._repr_html_())
"""then we get a html summary and can predict"""
self.assertIn('<style>', fit.training_summary._repr_html_())
prediction = df.model.predict(fit.model)
binary_prediction = prediction.iloc[:, 0] >= 0.5
np.testing.assert_array_equal(
binary_prediction,
np.array([
True,
False,
False,
True,
True,
False,
False,
True,
]))
"""and load the model"""
try:
copy = Model.load(temp)
pd.testing.assert_frame_equal(df.model.predict(fit.model),
df.model.predict(copy),
check_less_precise=True)
# test using context manager and ForecastProvider
pd.testing.assert_frame_equal(
df.model(temp).predict(forecast_provider=Forecast).df,
df.model.predict(copy),
check_less_precise=True)
finally:
os.remove(temp)
def test_soft_dtw_loss(self):
df = TEST_DF[["Close"]][-21:].copy()
class LstmAutoEncoder(PytorchNN):
def __init__(self):
super().__init__()
self.input_size = 1
self.seq_size = 10
self.hidden_size = 2
self.num_layers = 1
self.num_directions = 1
self._encoder =\
nn.RNN(input_size=self.input_size, hidden_size=self.hidden_size, num_layers=self.num_layers,
batch_first=True)
self._decoder =\
nn.RNN(input_size=self.hidden_size, hidden_size=self.input_size, num_layers=self.num_layers,
batch_first=True)
def forward_training(self, x):
# make sure to treat single elements as batches
x = x.view(-1, self.seq_size, self.input_size)
batch_size = len(x)
hidden_encoder = nn.Parameter(
t.zeros(self.num_layers * self.num_directions, batch_size,
self.hidden_size))
hidden_decoder = nn.Parameter(
t.zeros(self.num_layers * self.num_directions, batch_size,
self.input_size))
x, _ = self._encoder(x, hidden_encoder)
x = t.repeat_interleave(x[:, -2:-1], x.shape[1], dim=1)
x, hidden = self._decoder(x, hidden_decoder)
return x.squeeze()
def encode(self, x):
x = x.reshape(-1, self.seq_size, self.input_size)
batch_size = len(x)
with t.no_grad():
hidden = nn.Parameter(
t.zeros(self.num_layers * self.num_directions,
batch_size, self.hidden_size))
# return last element of sequence
return self._encoder(x, hidden)[0][:, -1]
def decode(self, x):
x = x.reshape(-1, self.seq_size, self.hidden_size)
batch_size = len(x)
with t.no_grad():
hidden = nn.Parameter(
t.zeros(self.num_layers * self.num_directions,
batch_size, self.input_size))
return self._decoder(x.float(), hidden)[0]
model = PytorchAutoEncoderModel(
LstmAutoEncoder,
PostProcessedFeaturesAndLabels(
df.columns.to_list(),
[lambda df: lag_columns(df, 10).dropna()],
df.columns.to_list(),
[lambda df: lag_columns(df, 10).dropna()],
["condensed-a", "condensed-b"]), SoftDTW, Adam)
with df.model() as m:
fit = m.fit(model, FittingParameter(epochs=100))
print(fit.test_summary.df)
encoded = df.model.predict(fit.model.as_encoder())
print(encoded)
def test_probabilistic_model_with_callback(self):
try:
pandas_ml_quant_data_provider = importlib.import_module(
"pandas_ml_quant")
from pandas_ml_quant import PricePredictionSummary
from pandas_ml_quant.model.summary.price_prediction_summary import PriceSampledSummary
except:
print("pandas_ml_quant not found, skipping!")
return
df = pd.DataFrame({
"Returns":
np.random.normal(-0.02, 0.03, 500) +
np.random.normal(0.03, 0.02, 500)
})
fl = PostProcessedFeaturesAndLabels(
features=["Returns"],
feature_post_processor=lambda df: df.ta.rnn(20),
labels=[
lambda df: df["Returns"].shift(-1).rename("Future_Returns")
],
targets=lambda df: (1 + df["Returns"]).cumprod().rename("Close"))
model_factory = PytorchNNFactory.create(
nn.Sequential(
nn.Linear(20, 10),
nn.Tanh(),
nn.Linear(10, 6),
LambdaSplitter(
lambda x: T.softmax(x[..., :2], dim=1),
lambda x: T.exp(x[..., 2:4]),
# enforce one mean positive and the other negativ
lambda x: T.cat([T.exp(x[..., 4:5]), -T.exp(x[..., 5:6])],
dim=1),
)),
predictor=lambda n, i: T.cat(n(i), dim=1),
trainer=lambda n, i: n(i))
def dist(probs, scales, locs):
return MixtureSameFamily(Categorical(probs=probs),
Normal(loc=locs, scale=scales))
def loss(y_pred):
probs, scales, locs = y_pred
return dist(probs, scales, locs)
def cdf_cb(arg):
probs, scales, locs = arg[..., :2], arg[..., 2:4], arg[..., 4:6]
return dist(probs, scales, locs)
summary_provider = PriceSampledSummary.with_reconstructor(
sampler=wrap_applyable(lambda params, samples: cdf_cb(params).
sample([int(samples.item())]),
nr_args=2),
samples=100,
confidence=0.8)
model = PytorchModel(module_provider=model_factory,
features_and_labels=fl,
criterion_provider=lambda: DistributionNLL(
loss, penalize_toal_variance_lambda=1.1),
optimizer_provider=Adam,
summary_provider=summary_provider)
fit = df.model.fit(
model,
FittingParameter(epochs=10,
batch_size=6,
splitter=naive_splitter(0.25)),
#verbose=1,
callbacks=[
TestConfidenceInterval(
TestConfidenceInterval.CdfConfidenceInterval(
wrap_applyable(
lambda params, val: cdf_cb(params).cdf(val),
nr_args=2),
interval=0.8),
wrap_applyable(lambda params: cdf_cb(params).variance),
early_stopping=True)
])
print(fit.test_summary.calc_scores())