def peplot():
X, y = load_concrete()
X_train, X_test, y_train, y_test = tts(X, y, test_size=0.2)
oz = PredictionError(Lasso(), ax=newfig())
oz.fit(X_train, y_train)
oz.score(X_test, y_test)
savefig(oz, "prediction_error")
def alpha_selection(ax=None):
data = load_concrete(return_dataset=True)
X, y = data.to_pandas()
alphas = np.logspace(-10, 1, 400)
viz = AlphaSelection(LassoCV(alphas=alphas), ax=ax)
return tts_plot(viz, X, y)
def residuals():
X, y = load_concrete()
X_train, X_test, y_train, y_test = tts(X, y, test_size=0.2)
oz = ResidualsPlot(Ridge(), ax=newfig())
oz.fit(X_train, y_train)
oz.score(X_test, y_test)
savefig(oz, "residuals")
def manifold(dataset, manifold):
if dataset == "concrete":
X, y = load_concrete()
elif dataset == "occupancy":
X, y = load_occupancy()
else:
raise ValueError("unknown dataset")
oz = Manifold(manifold=manifold, ax=newfig())
oz.fit_transform(X, y)
savefig(oz, "{}_{}_manifold".format(dataset, manifold))
def dataset_example(dataset="occupancy",
manifold="all",
path="images/",
quick=False,
**kwargs):
if manifold == "all":
if path is not None and not os.path.isdir(path):
"please specify a directory to save examples to"
for algorithm in MANIFOLD_ALGORITHMS:
if algorithm in SKIP:
continue
print("generating {} {} manifold".format(dataset, algorithm))
fpath = os.path.join(
path, "{}_{}_manifold.png".format(dataset, algorithm))
try:
dataset_example(dataset, algorithm, fpath)
except Exception as e:
print("could not visualize {} manifold on {} data: {}".format(
algorithm, dataset, e))
continue
# Break here!
return
# Create single example
_, ax = plt.subplots(figsize=(9, 6))
# Checks for the dataset which user wants to use
if dataset == "occupancy":
X, y = load_occupancy()
elif dataset == "concrete":
X, y = load_concrete()
else:
raise Exception("unknown dataset '{}'".format(dataset))
# Check if the quick method is called
if quick:
oz = manifold_embedding(X, y, manifold=manifold, show=False, **kwargs)
oz.show(outpath=path)
else:
oz = Manifold(ax=ax, manifold=manifold, **kwargs)
oz.fit_transform(X, y)
oz.show(outpath=path)
def test_integration_coef(self):
"""
Integration test of visualizer with coef param
"""
# Load the test dataset
dataset = load_concrete(return_dataset=True)
X, y = dataset.to_numpy()
features = dataset.meta["features"]
fig = plt.figure()
ax = fig.add_subplot()
reg = Lasso(random_state=42)
features = list(map(lambda s: s.title(), features))
viz = FeatureImportances(reg, ax=ax, labels=features, relative=False)
viz.fit(X, y)
viz.finalize()
# Appveyor and Linux conda non-text-based differences
self.assert_images_similar(viz, tol=16.2)
def test_numpy_integration(self):
"""
Test on concrete dataset with numpy arrays
"""
data = load_concrete(return_dataset=True)
X, y = data.to_numpy()
assert isinstance(X, np.ndarray)
assert isinstance(y, np.ndarray)
_, ax = plt.subplots()
viz = CooksDistance(ax=ax).fit(X, y)
assert_fitted(viz)
assert viz.distance_.sum() == pytest.approx(1.2911900571300652)
assert viz.p_values_.sum() == pytest.approx(1029.9999525376425)
assert viz.influence_threshold_ == pytest.approx(0.003883495145631068)
assert viz.outlier_percentage_ == pytest.approx(7.3786407766990285)
viz.finalize()
self.assert_images_similar(viz)
def test_with_fitted(self):
"""
Test that visualizer properly handles an already-fitted model
"""
X, y = load_concrete(return_dataset=True).to_numpy()
model = Lasso().fit(X, y)
with mock.patch.object(model, "fit") as mockfit:
oz = FeatureImportances(model)
oz.fit(X, y)
mockfit.assert_not_called()
with mock.patch.object(model, "fit") as mockfit:
oz = FeatureImportances(model, is_fitted=True)
oz.fit(X, y)
mockfit.assert_not_called()
with mock.patch.object(model, "fit") as mockfit:
oz = FeatureImportances(model, is_fitted=False)
oz.fit(X, y)
mockfit.assert_called_once_with(X, y)
def dataset_example(dataset="occupancy",
manifold="all",
path="images/",
**kwargs):
if manifold == "all":
if path is not None and not os.path.isdir(path):
"please specify a directory to save examples to"
for algorithm in MANIFOLD_ALGORITHMS:
if algorithm in SKIP:
continue
print("generating {} {} manifold".format(dataset, algorithm))
fpath = os.path.join(
path, "{}_{}_manifold.png".format(dataset, algorithm))
try:
dataset_example(dataset, algorithm, fpath)
except Exception as e:
print("could not visualize {} manifold on {} data: {}".format(
algorithm, dataset, e))
continue
# Break here!
return
# Create single example
_, ax = plt.subplots(figsize=(9, 6))
oz = Manifold(ax=ax, manifold=manifold, **kwargs)
if dataset == "occupancy":
X, y = load_occupancy()
elif dataset == "concrete":
X, y = load_concrete()
else:
raise Exception("unknown dataset '{}'".format(dataset))
oz.fit(X, y)
oz.poof(outpath=path)
from yellowbrick.datasets import load_energy # Load dataset X, y = load_energy() # make our dataset read as x and y axis values somehow and replace this dataset with ours # X = [] # makes a list # y = [] # makes a list # data = pd.read_csv(r'C:\Users\Jujin\Desktop\cs-4641-group-44\cleaned_encoded_COVID_Data_Copy.csv') # for row in data: # X.append(row[1]) # selects data from the ith row # y.append(row[2]) # selects data from the ith row # Use the quick method and immediately show the figure alphas(LassoCV(random_state=0), X, y) # Load a regression dataset X, y = load_concrete() # same as above #X = pd.read_csv(r'C:\Users\Jujin\Desktop\cs-4641-group-44\training_data.csv') #y = pd.read_csv(r'C:\Users\Jujin\Desktop\cs-4641-group-44\test_data.csv') # X_train = [] # makes a list # y_train = [] # makes a list # d = pd.read_csv(r'C:\Users\Jujin\Desktop\cs-4641-group-44\training_data.csv') # for row in d: # X_train.append(row[1]) # selects data from the ith row # y_train.append(row[2]) # selects data from the ith row # # X_test = [] # makes a list # y_test = [] # makes a list # da = pd.read_csv(r'C:\Users\Jujin\Desktop\cs-4641-group-44\test_data.csv') # for row in da:
def binning():
_, y = load_concrete()
oz = BalancedBinningReference(ax=newfig())
oz.fit(y)
savefig(oz, "balanced_binning_reference")
def alphas():
X, y = load_concrete()
alphas = np.logspace(-10, 1, 400)
oz = AlphaSelection(LassoCV(alphas=alphas), ax=newfig())
oz.fit(X, y)
savefig(oz, "alpha_selection")
def jointplot():
X, y = load_concrete()
oz = JointPlot(columns=["cement", "splast"], ax=newfig())
oz.fit_transform(X, y)
savefig(oz, "jointplot")
fig.set_size_inches(7, 5)
fig.savefig("5._Autoregression_Model.png")
# Xenia: Showing all my plots as an output
plt.show()
# %%
# Residuals Plot (Trying new things)
# The residuals plot shows how the model is injecting error, the bold \
# horizontal line at residuals = 0 is no error, and any point above or below \
# that line, indicates the magnitude of error.
# (https://www.scikit-yb.org/en/latest/quickstart.html#installation)
# Load a regression dataset
X, y = load_concrete()
# Create training and test sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1)
visualizer = ResidualsPlot(LinearRegression())
visualizer.fit(X_train, y_train) # Fit the training data to the visualizer
visualizer.score(X_test, y_test) # Evaluate the model on the test data
visualizer.show() # Finalize and render the figure
# Xenia: Saving my plots
plt.show()
fig.set_size_inches(7, 5)
plt.savefig("6._Residuals_Plot.png")
fig.savefig("6._Residuals_Plot.png")
def residuals_plot(ax=None):
data = load_concrete(return_dataset=True)
X, y = data.to_pandas()
viz = ResidualsPlot(Ridge(), ax=ax)
return tts_plot(viz, X, y)
def prediction_error(ax=None):
data = load_concrete(return_dataset=True)
X, y = data.to_pandas()
viz = PredictionError(Lasso(), ax=ax)
return tts_plot(viz, X, y)
