Python hyper-parametric auto-search module GRIDSEARCHCV (Favorites)

Source: Internet
Author: User
Tags generator svm

1. Introduction

When we run the machine learning program, especially when adjusting the network parameters, there are usually many parameters to be adjusted, the combination of parameters is more complicated. In accordance with the principle of attention > Time > Money, manual adjustment of attention costs by manpower is too high and is not worth it. The For loop or for loop-like approach is constrained by too-distinct levels, concise and flexible, with high attention costs and error-prone. This paper introduces the GRIDSEARCHCV module of the Sklearn module, which can automatically search for different model combinations with different parameters within the specified range to effectively liberate the attention.

2. Introduction to the GRIDSEARCHCV module

This module is a sub-module of the Sklearn module, and the import method is very simple

from import GRIDSEARCHCV  
Function Prototypes:
classsklearn.model_selection. GRIDSEARCHCV (Estimator, Param_grid, Scoring=none, Fit_params=none, N_jobs=1, Iid=true, Refit=true, Cv=None, Verbose=0, Pre_dispatch='2*n_jobs', error_score='Raise', Return_train_score=true)

Where the CV can be an integer or a cross-validation generator or an iterator, the 4 types of input for CV parameters are listed below:

    1. None: Default parameter, function uses default 30 percent cross-validation
    2. Integer k:k cross-validation. For categorical tasks, use Stratifiedkfold (category balancing, which has as many training sets per class as you can see in the official documentation). For other tasks, use Kfold
    3. Cross-validation generator: Get your own sketch builder
    4. Iterators that can generate training sets and test sets
3. Automatic saving of analysis results

comma-separated values (comma-separated values,CSV, and sometimes also referred to as character-delimited value , because delimited characters can also be not commas), whose files store tabular data (numbers and text) in plain text. Plain text means that the file is one and does not contain data that must be interpreted like a binary number. A CSV file consists of any number of records separated by a newline character, each record consists of a field, and the delimiter between the fields is another character or string, most commonly a comma or tab. Typically, all records have exactly the same sequence of fields.

CSV file has a prominent advantage, can be opened with Excel and other software, compared to Notepad and Matlab, Python and other programming language interface, easy to view, production reports, post-finishing and so on.

In the GRIDSEARCHCV module, the combination of different hyper-parameters and their computational results are saved in a dictionary in Clf.cv_results_, and the Python Pandas module provides a way to efficiently organize the data, requiring only 3 lines of code to solve the problem.

In the GRIDSEARCHCV module, the combination of different hyper-parameters and their computational results are saved in a dictionary in Clf.cv_results_, and the Python Pandas module provides a way to efficiently organize the data, requiring only 3 lines of code to solve the problem.
Cv_result =
With Open ('Cv_result.csv','W') as F:
Cv_result.to_csv (f)
4. Complete routines

the code is clear and understandable without explanation. Https://github.com/JiJingYu/tensorflow-exercise/tree/master/svm_grid_search

ImportPandas as PD fromSklearnImportSVM, Datasets fromsklearn.model_selectionImportGRIDSEARCHCV fromSklearn.metricsImportClassification_report 
Iris =Datasets.load_iris ()parameters = {'Kernel':('Linear', 'RBF'), 'C': [1, 2, 4],'Gamma': [0.125, 0.25, 0.5, 1, 2, 4]}SVR =SVM. SVC ()CLF = GRIDSEARCHCV (SVR, parameters, N_jobs=-1)Clf.fit (Iris.data, Iris.target)Cv_result =PD. Dataframe.from_dict (CLF.CV_RESULTS_)With Open ('Cv_result.csv','W') as F: cv_result.to_csv (f)     Print('The parameters of the best model is:')Print(CLF.BEST_PARAMS_)y_pred =clf.predict (iris.data)Print(Classification_report (Y_true=iris.target, y_pred=y_pred))


Python Hyper-parameter auto-search module GRIDSEARCHCV (favorites)

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.