Using SVM's regressoar, we are trying to conduct an assessment analysis of financial markets.
After coding with reference to the text, the last line plot_confusion_matrix
, plot_roc_curve
failed to proceed (message content is as title).
I googled the error message, updated the PIP and sciket-learn, and called the kernel RESTART, but the situation remains the same.
By the way, pip is (22.3.1) and scicket-learn is (1.2.0).
Could someone give me some advice?
By the way, the contents of the coding are as follows.
Thank you for your cooperation.
#Base Libraries
import pandas aspd
import numpy as np
import matplotlib.pyplot asplt
plt.style.use('fivethirtight')
# preprocessing
from sklearn.preprocessing import MinMaxScaler, RobustScaler
from sklearn.pipeline import Pipeline
from sklearn.model_selection import train_test_split, TimeSeriesSplit, GridSearchCV
# SVM
from sklearn.svm import SVR
# Metrics
from sklearn.metrics import r2_score, mean_squared_error, mean_absolute_error, accurability_score
from sklearn.metrics import acuity_score, classification_report, confusion_matrix
from sklearn.metrics import plot_confusion_matrix, auc, roc_curve, plot_roc_curve**
plot_confusion_matrix
is scheduled to be discarded (deprecated) in the scikit-learn 1.0.0 version, and it seems that it was actually deleted (removed) in the 1.2.0 version.
Try changing it to not use it or downgrading it to a version earlier than 1.2.0 if you want to use it.Is it before 1.1.3?
Release history scikit-learn-PyPI
metrics.plot_confusion_matrix
is decremented in favorite of these two class methods and will be removed in 1.2.#18543by Guillaume Lemeritre.If you delete plot_confusion_matrix
from the source of the question, you will get an error that plot_roc_curve
cannot import the same way.There is no such description in the document, but the 1.2.0 version of API Reference and metrics.plot_det_curve
with similar names after the description above have been removed as well.
The 1.1.3 version of API Reference seems to have a description left.
Plotting
© 2024 OneMinuteCode. All rights reserved.