增加“机器学习”和“数据分析”的块

2024-11-29 08:53:29 +08:00
parent 28552b0a22
commit ead77c4b89
8 changed files with 141 additions and 28 deletions
--- a/boards/default_src/python_mixpy/blocks/data.js
+++ b/boards/default_src/python_mixpy/blocks/data.js
@@ -1247,16 +1247,12 @@ export const pandas_drop_columns = {
        this.appendValueInput('DATAFRAME')
            .appendField('从数据集');
        this.appendValueInput('COLUMNS')
-            .appendField('中删除列');
-        this.appendDummyInput()
-            .appendField('沿着axis')
-            .appendField(new Blockly.FieldDropdown([
-                ['行', '0'],
-                ['列', '1']
-            ]), 'AXIS');
+            .appendField('中删除列')
+            .setCheck(String);
+        this.setInputsInline(true);
        this.setOutput(true);
-        this.setTooltip('Drops columns from dataframe.');
-    }
+        this.setTooltip('从数据框中删除指定的列。用逗号分隔多个列名。');
+    },
 };

 export const numpy_ones = {
--- a/boards/default_src/python_mixpy/blocks/sklearn.js
+++ b/boards/default_src/python_mixpy/blocks/sklearn.js
@@ -405,6 +405,49 @@ export const sklearn_GaussianNB = {
    }
 };

+//sklearn 初始化PCA降维
+export const sklearn_pca = {
+    init: function () {
+        this.appendDummyInput()
+            .appendField("sklearn 初始化 PCA 算法");
+        this.appendValueInput("model_name")
+            .setCheck(null)
+            .setAlign(Blockly.inputs.Align.RIGHT)
+            .appendField(Blockly.Msg.MODEL_NAME);
+        this.appendValueInput("n_components")
+            .setCheck(null)
+            .setAlign(Blockly.inputs.Align.RIGHT)
+            .appendField(Blockly.Msg.SKLEARN_PCA_N_COMPONENTS);
+        this.setInputsInline(false);
+        this.setPreviousStatement(true, null);
+        this.setNextStatement(true, null);
+        this.setColour(SKLEARN_HUE);
+        this.setTooltip("");
+        this.setHelpUrl("");
+    }
+};
+
+//sklearn PCA拟合并转换数据
+export const sklearn_pca_fit_transform = {
+    init: function () {
+        this.appendDummyInput()
+            .appendField("sklearn PCA 降维");
+        this.appendValueInput("model_name")
+            .setCheck(null)
+            .setAlign(Blockly.inputs.Align.RIGHT)
+            .appendField(Blockly.Msg.MODEL_NAME);
+        this.appendValueInput("train_data")
+            .setCheck(null)
+            .setAlign(Blockly.inputs.Align.RIGHT)
+            .appendField(Blockly.Msg.EIGENVALUES);
+        this.setInputsInline(true);
+        this.setOutput(true, null);
+        this.setColour(SKLEARN_HUE);
+        this.setTooltip("");
+        this.setHelpUrl("");
+    }
+};
+
 //sklearn 初始化K-均值聚类
 export const sklearn_KMeans = {
    init: function () {
@@ -426,11 +469,29 @@ export const sklearn_KMeans = {
            .setCheck(null)
            .setAlign(Blockly.inputs.Align.RIGHT)
            .appendField(Blockly.Msg.RANDOM_SEED);
-        this.appendValueInput("n_jobs")
+        this.setInputsInline(false);
+        this.setPreviousStatement(true, null);
+        this.setNextStatement(true, null);
+        this.setColour(SKLEARN_HUE);
+        this.setTooltip("");
+        this.setHelpUrl("");
+    }
+};
+
+//sklearn KMeans拟合数据
+export const sklearn_KMeans_fit = {
+    init: function () {
+        this.appendDummyInput()
+            .appendField("sklearn K-均值聚类");
+        this.appendValueInput("model_name")
            .setCheck(null)
            .setAlign(Blockly.inputs.Align.RIGHT)
-            .appendField(Blockly.Msg.SKLEARN_THREADS);
-        this.setInputsInline(false);
+            .appendField(Blockly.Msg.MODEL_NAME);
+        this.appendValueInput("train_data")
+            .setCheck(null)
+            .setAlign(Blockly.inputs.Align.RIGHT)
+            .appendField(Blockly.Msg.EIGENVALUES);
+        this.setInputsInline(true);
        this.setPreviousStatement(true, null);
        this.setNextStatement(true, null);
        this.setColour(SKLEARN_HUE);
--- a/boards/default_src/python_mixpy/generators/data.js
+++ b/boards/default_src/python_mixpy/generators/data.js
@@ -519,8 +519,7 @@ export const pandas_drop_columns = function (block, generator) {
    generator.definitions_.import_pandas = "import pandas";
    var dataframe = generator.valueToCode(block, 'DATAFRAME', generator.ORDER_ATOMIC) || 'df';
    var columns = generator.valueToCode(block, 'COLUMNS', generator.ORDER_ATOMIC) || '[]';
-    var axis = block.getFieldValue('AXIS') || '0';
-    var code = dataframe + '.drop(columns=' + columns + ', axis=' + axis + ')';
+    var code = dataframe + '.drop(columns=' + columns + ', axis=1)';
    return [code, generator.ORDER_ATOMIC];
 }

--- a/boards/default_src/python_mixpy/generators/sklearn.js
+++ b/boards/default_src/python_mixpy/generators/sklearn.js
@@ -147,18 +147,40 @@ export const sklearn_GaussianNB = function (_, generator) {
    return code;
 }

+// sklearn 初始化PCA降维
+export const sklearn_pca = function (_, generator) {
+    var value_model_name = generator.valueToCode(this, 'model_name', generator.ORDER_ATOMIC) || 'pca';
+    var value_n_components = generator.valueToCode(this, 'n_components', generator.ORDER_ATOMIC) || '2';
+    generator.definitions_['import_sklearn_pca'] = 'from sklearn.decomposition import PCA';
+    var code = value_model_name + ' = PCA(n_components=' + value_n_components + ')\n';
+    return code;
+}
+
+export const sklearn_pca_fit_transform = function(block, generator) {
+    var value_model_name = generator.valueToCode(block, 'model_name', generator.ORDER_ATOMIC);
+    var value_train_data = generator.valueToCode(block, 'train_data', generator.ORDER_ATOMIC);
+    var code = value_model_name + '.fit_transform(' + value_train_data + ')';
+    return [code, generator.ORDER_ATOMIC];
+};
+
 // sklearn 初始K-均值聚类
 export const sklearn_KMeans = function (_, generator) {
    var value_model_name = generator.valueToCode(this, 'model_name', generator.ORDER_ATOMIC) || 'model';
    var value_n_clusters = generator.valueToCode(this, 'n_clusters', generator.ORDER_ATOMIC) || '8';
    var value_max_iter = generator.valueToCode(this, 'max_iter', generator.ORDER_ATOMIC) || '300';
    var value_random_state = generator.valueToCode(this, 'random_state', generator.ORDER_ATOMIC) || 'None';
-    var value_n_jobs = generator.valueToCode(this, 'n_jobs', generator.ORDER_ATOMIC) || 'None';
    generator.definitions_['import_sklearn_KMeans'] = 'from sklearn.cluster import KMeans';
-    var code = value_model_name + ' = KMeans(n_clusters = ' + value_n_clusters + ',max_iter = ' + value_max_iter + ',random_state = ' + value_random_state + ',n_jobs = ' + value_n_jobs + ')\n';
+    var code = value_model_name + ' = KMeans(n_clusters = ' + value_n_clusters + ',max_iter = ' + value_max_iter + ',random_state = ' + value_random_state + ')\n';
    return code;
 }

+export const sklearn_KMeans_fit = function(block, generator) {
+    var value_model_name = generator.valueToCode(block, 'model_name', generator.ORDER_ATOMIC);
+    var value_train_data = generator.valueToCode(block, 'train_data', generator.ORDER_ATOMIC);
+    var code = value_model_name + '.fit(' + value_train_data + ')\n';
+    return code;
+};
+
 // sklearn 训练模型
 export const sklearn_fit = function (_, generator) {
    var value_model_name = generator.valueToCode(this, 'model_name', generator.ORDER_ATOMIC) || 'model';
--- a/boards/default_src/python_mixpy/template.xml
+++ b/boards/default_src/python_mixpy/template.xml
@@ -2527,12 +2527,6 @@
          <field name="VAR">df</field>
        </shadow>
      </value>
-      <value name="COLUMNS">
-        <shadow type="variables_get">
-          <field name="VAR">columns</field>
-        </shadow>
-      </value>
-      <field name="AXIS">0</field>
    </block>
    <block type="variables_set">
      <field name="VAR">x</field>
@@ -2727,6 +2721,13 @@
        </shadow>
      </value>
    </block>
+    <block type="pylab_imshow">
+      <value name="ARRAY">
+        <shadow type="variables_get">
+          <field name="VAR">myArray</field>
+        </shadow>
+      </value>
+    </block>
    <block type="series_create">
      <value name="SER">
        <shadow type="variables_get">