Python joblib.dump方法代码示例

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def save_classifier_model(df_question_train, df_question_class, model_type="linearSVC"):
    
    """
    FIXME: Although the classifier is being saved in Pickle file. It is not being used to predict.
    Since, Support Vector Classifier, fails when it encounters new features it failed to see while training.
    """

    classifier_model = None
    training_model_path = os.path.join(CORPUS_DIR, QUESTION_CLASSIFICATION_MODEL)

    if model_type == "linearSVC":
        classifier_model = support_vector_machine(df_question_train, df_question_class)
    else:
        logger.error("Undefined Classifier")

    if classifier_model is not None:
        joblib.dump(classifier_model, training_model_path)
        logger.info("Model saved at {0}".format(training_model_path))
    else:
        logger.error("Model empty") 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def save_predictions(train_ids, train_predictions, meta_test, out_of_fold_test_predictions):
    averaged_mask_predictions_test = np.mean(np.array(out_of_fold_test_predictions), axis=0)

    LOGGER.info('Saving predictions')
    out_of_fold_train_predictions_path = os.path.join(EXPERIMENT_DIR, 'out_of_fold_train_predictions.pkl')
    joblib.dump({'ids': train_ids,
                 'images': train_predictions}, out_of_fold_train_predictions_path)

    out_of_fold_test_predictions_path = os.path.join(EXPERIMENT_DIR, 'out_of_fold_test_predictions.pkl')
    joblib.dump({'ids': meta_test[ID_COLUMN].tolist(),
                 'images': averaged_mask_predictions_test}, out_of_fold_test_predictions_path)


#  .___  ___.      ___       __  .__   __.
#  |   \/   |     /   \     |  | |  \ |  |
#  |  \  /  |    /  ^  \    |  | |   \|  |
#  |  |\/|  |   /  /_\  \   |  | |  . `  |
#  |  |  |  |  /  _____  \  |  | |  |\   |
#  |__|  |__| /__/     \__\ |__| |__| \__|
# 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def store_models(self) -> None:
        """
        Save the learners to disk.

        :returns: None
        :rtype: None
        """

        # Iterate over the learner types (for which there will be
        # separate instances for each sub-experiment of the
        # cross-validation experiment)
        for learner_name in self.cv_learners_:
            loginfo('Saving {0} model files to disk...'.format(learner_name))
            for i, estimator in enumerate(self.cv_learners_[learner_name]):
                loginfo('Saving {0} model file #{1}'.format(learner_name, i + 1))
                joblib.dump(estimator,
                            self.model_path_template_.format(learner_name, i + 1)) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def save(self, model_fname):
        """
        Save the model into file.

        Args:
            model_fname (str): Filename of the model.
        """
        joblib.dump(self.model, '%s.pkl' % model_fname) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def persist(self, filepath):
        joblib.dump(self.embedding_matrix, filepath) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def persist(self, filepath):
        object_pickle = {'char_level': self.char_level,
                         'maxlen': self.maxlen,
                         'num_words': self.num_words,
                         'tokenizer': self.tokenizer}
        joblib.dump(object_pickle, filepath) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def persist(self, filepath):
        joblib.dump({}, filepath) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def persist(self, filepath):
        joblib.dump(self.estimator, filepath) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def persist(self, filepath):
        params = {'x_columns': self.x_columns,
                  'y_columns': self.y_columns
                  }
        joblib.dump(params, filepath) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def persist(self, filepath):
        params = {'loader_params': self.loader_params}
        joblib.dump(params, filepath) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def save(self, filepath):
        params = {'loader_params': self.loader_params}
        joblib.dump(params, filepath) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def script_run():
    # 产生keyword
    kw_list = build_key_word("train.txt")
    # 保存数据
    fp = open("new_word.txt", encoding="utf-8", mode="w")
    for word in kw_list:
        fp.write(word + "\n")
    fp.close()
    # kw_list = load_key_words("word.txt")
    feature, label = get_feature("train.txt", kw_list)
    gnb = MultinomialNB()  # 多项式贝叶斯
    gnb = gnb.fit(feature, label)
    joblib.dump(gnb, 'model/gnb.model')
    print("训练完成") 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def tofile(file_name, model):
    joblib.dump(model, file_name) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def save_model(self, path):
        assert self.n_ensemble == len(self.model)
        for i in range(self.n_ensemble):
            joblib.dump(self.model[i], path + str(i) + '.pkl')
        if self.feature_type == 'descriptors':
            np.save(path + 'desc_mean.npy', self.desc_mean) 

# 需要导入模块: from sklearn.externals import joblib [as 别名]
# 或者: from sklearn.externals.joblib import dump [as 别名]
def make_example_classifier(filename):
    # Create a dummy RF model for train/classify testing
    rf = RandomForestClassifier()
    p, n_class = 42, 2
    n = n_class * 5
    X = np.random.rand(n, p)
    y = np.repeat(range(n_class), n / n_class)
    rf.fit(X, y)
    jl.dump(rf, filename)


# EXAMPLE DATASETS