Module src.visualization.world
Expand source code
from src.data.world import world_features
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
def plot_residuals(ypred: np.ndarray, ytest: np.ndarray, dataset: str = "train"):
# create id line
# TODO: Make this more robust
# min_y = np.minimum(ypred.min(), ytest.min())
# max_y = np.minimum(ypred.max(), ytest.max())
if dataset == "train":
id_line = np.linspace(-3, 3, 50)
elif dataset == "test":
id_line = np.linspace(-6, -2, 50)
else:
raise ValueError(f"Unrecognized dataset: {dataset}")
fig, ax = plt.subplots(figsize=(10, 10))
ax.scatter(ypred, ytest, marker="o")
ax.plot(id_line, id_line, linewidth=3, color="black")
return fig, ax
def plot_test_residuals(ypred: pd.DataFrame, ytest: pd.DataFrame):
# create id line
# TODO: Make this more robust
# min_y = np.minimum(ypred.min(), ytest.min())
# max_y = np.minimum(ypred.max(), ytest.max())
id_line = np.linspace(-3, 3, 50)
fig, ax = plt.subplots(figsize=(10, 10))
ax.scatter(ypred.values.ravel(), ytest.values.ravel(), marker="o")
ax.plot(id_line, id_line, linewidth=3, color="black")
return fig, ax
Functions
def plot_residuals(ypred: numpy.ndarray, ytest: numpy.ndarray, dataset: str = 'train')
-
Expand source code
def plot_residuals(ypred: np.ndarray, ytest: np.ndarray, dataset: str = "train"): # create id line # TODO: Make this more robust # min_y = np.minimum(ypred.min(), ytest.min()) # max_y = np.minimum(ypred.max(), ytest.max()) if dataset == "train": id_line = np.linspace(-3, 3, 50) elif dataset == "test": id_line = np.linspace(-6, -2, 50) else: raise ValueError(f"Unrecognized dataset: {dataset}") fig, ax = plt.subplots(figsize=(10, 10)) ax.scatter(ypred, ytest, marker="o") ax.plot(id_line, id_line, linewidth=3, color="black") return fig, ax
def plot_test_residuals(ypred: pandas.core.frame.DataFrame, ytest: pandas.core.frame.DataFrame)
-
Expand source code
def plot_test_residuals(ypred: pd.DataFrame, ytest: pd.DataFrame): # create id line # TODO: Make this more robust # min_y = np.minimum(ypred.min(), ytest.min()) # max_y = np.minimum(ypred.max(), ytest.max()) id_line = np.linspace(-3, 3, 50) fig, ax = plt.subplots(figsize=(10, 10)) ax.scatter(ypred.values.ravel(), ytest.values.ravel(), marker="o") ax.plot(id_line, id_line, linewidth=3, color="black") return fig, ax