yonghong.script.port is a sub-toolkit for writing custom python scripts, including two tool classes:

EntryPoint: The input and output ports of the current script node.

        Input: The data input of this node, to access certain data of a pre-node can be like this: entry.input['Data Set Node 1'][ResourceType.DATAFRAME] means obtain the output data set of "Data Set Node 1";

                    Output: the data output of this node;

                                    dataset: the dataset to be output;

                                    model: the model to be output;

                                    pmml: pmml to be output;

                                    images: The png image data to be output, how to use images.put_image_from_plot("Picture Name 1", plot)

ResourceType: the enumerated type of input and output data, DATAFRAME means data set, MODEL means model

Case 1—The pre-node is a data set node and a training model node (training model exported by Python script); the node output is a data set

# EntryPoint: The input and output ports of the current script node

# ResourceType: the enumerated type of input and output data, DATAFRAME means data set, MODEL means model

from yonghong.script.port import EntryPoint, ResourceType

entry = EntryPoint() # holds the input and output of this node

input_df = entry.input['sample data set'][ResourceType.DATAFRAME] # Get the output data of the specified pre-node'sample data set'

input_model = entry.input['Decision Tree Model'][ResourceType.MODEL] # Get the training model output by the specified pre-node ‘decision tree model’

import pandas as pd

entry.output.dataset = pd.DataFrame({'field name':input_model.predict(input_df)}) # Output the prediction result, if there is a post node, it will be received by the post node

Case 2-The pre-node is a data set node and a plug-in node (with output data set); the node output is a training model

# EntryPoint: The input and output ports of the current script node

# ResourceType: the enumerated type of input and output data, DATAFRAME means data set, MODEL means model

from yonghong.script.port import EntryPoint, ResourceType

from xgboost import XGBClassifier

entry = EntryPoint() # holds the input and output of this node

input_dataset1 = entry.input['sample data set'][ResourceType.DATAFRAME] # Get the output data of the specified pre-node'sample data set'

input_dataset2 = entry.input['standardized'][ResourceType.DATAFRAME] # Get the output data set of the specified pre-node ‘standardized’

features = ['V' + str(i) for i in range(1, 29)] + ['Amount'] # Specify the feature field.

target ='Class'

train_X = input_dataset1[features] # Read in the feature data of the training set.

train_y = input_dataset1[target] # Read in the classification value of the training set.

clf = XGBClassifier(n_jobs=3)

clf.fit(X=train_X, y=train_y)

import pandas as pd

entry.output.dataset = pd.DataFrame({'field name':clf.predict(input_dataset2[features])}) # Output the prediction result, if there is a post node, it will be received by the post node

entry.output.model = clf # If you need to pass the classifier clf to the next node, assign clf to the model.

entry.output.pmml = to_pmml(clf, features, target) # Generate the PMML text of the model.

Case 3—The pre-node is a data set node and a plug-in node (with output data set); the node output is a png image

# EntryPoint: The input and output ports of the current script node

# ResourceType: the enumerated type of input and output data, DATAFRAME means data set, MODEL means model

from yonghong.script.port import EntryPoint, ResourceType

from xgboost import XGBClassifier

entry = EntryPoint() # holds the input and output of this node

input_dataset1 = entry.input['sample data set'][ResourceType.DATAFRAME] # Get the output data of the specified pre-node'sample data set'

input_dataset2 = entry.input['standardized'][ResourceType.DATAFRAME] # Get the output data set of the specified pre-node ‘standardized’

# Drawing section

import matplotlib.pyplot as plt

from mpl_toolkits.mplot3d import Axes3D

from sklearn import manifold

df = input_dataset1

df_lle, _ = manifold.locally_linear_embedding(df.drop(columns='ElectricLeakage'), n_neighbors=30, n_components=2) # Perform dimensionality reduction operations on the data

fig = plt.figure()

ax = fig.add_subplot(211, projection='3d')

ax.scatter(df['LineLoss(%)'], df['ElectricQuantityDowntrend'], df['AlarmNumber'], c=df['ElectricLeakage'], cmap=plt.cm.Spectral)

ax.set_xlabel('LineLoss(%)')

ax.set_ylabel('EQD')

ax.set_zlabel('AlarmNumber')

ax.set_title('Original data')

ax = fig.add_subplot(212)

ax.scatter(df_lle[:, 0], df_lle[:, 1], c=df['ElectricLeakage'], cmap=plt.cm.Spectral)

ax.set_xlabel('x1')

ax.set_ylabel('x2')

ax.set_title('Projected data')

plt.axis('tight')

#After the drawing parameters are configured, do not use plt.show() to display pictures, this function is invalid

entry.output.images.put_image_from_plot("case picture name 1",plt) # After configuring the drawing parameters, pass plt to the images variable, and the chart can be drawn to the page