Upload 3 files

app.py

@@ -0,0 +1,245 @@
+import gradio as gr
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+from scipy.stats import f_oneway
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.model_selection import cross_val_score
+from sklearn.naive_bayes import GaussianNB
+from sklearn.neighbors import KNeighborsClassifier
+from sklearn.svm import SVC
+from sklearn.tree import DecisionTreeClassifier
+
+
+def add_max_score_to_list(temp_scores, current_score, selected_indices, selected_indices_list):
+    max_score_index = np.argmax(np.array(temp_scores))
+    current_score.append(temp_scores[max_score_index])
+    selected_indices.add(max_score_index)
+    selected_indices_list.append(max_score_index)
+
+
+def fs(data, method, num_fea_int, clf):
+    data = pd.read_csv(data.name)
+    X = data.iloc[:, :-1].values
+    y = data['Label'].values
+    num_fea_int = int(num_fea_int)
+    if method == 'MRMR_FCD':
+        num_features = len(X[0])
+
+        f_test_scores = [f_oneway(X[:, i], y)[0] for i in range(num_features)]
+        # 添加起始特征的分数到current_score
+        current_score = [max(f_test_scores)]
+
+        # 索引从0开始
+        # start_feature_index = random.randint(0, num_features - 1)
+        # 索引从最高分数的特征开始
+        start_feature_index = f_test_scores.index(max(f_test_scores))
+        selected_indices = set()
+        selected_indices_list = []
+        selected_indices.add(start_feature_index)
+        selected_indices_list.append(start_feature_index)
+
+        pearson_score_matrix = np.zeros((num_features, num_features))
+
+        for _ in range(num_fea_int - 1):
+            temp_scores = []
+            for i in range(num_features):
+                if i in selected_indices:
+                    temp_scores.append(-float('inf'))
+                else:
+                    f_test_score = f_test_scores[i]
+                    diff = 0
+                    for j in selected_indices:
+                        # pearson score
+                        if j > i:
+                            if pearson_score_matrix[i][j] == 0:
+                                pearson_score_matrix[i][j] = np.corrcoef(X[:, i], X[:, j])[0, 1]
+                            diff += pearson_score_matrix[i][j]
+                        else:
+                            if pearson_score_matrix[j][i] == 0:
+                                pearson_score_matrix[j][i] = np.corrcoef(X[:, i], X[:, j])[0, 1]
+                            diff += pearson_score_matrix[j][i]
+                        # diff += np.corrcoef(X[:,i], X[:,j])[0, 1]
+                    temp_scores.append(f_test_score - diff / len(selected_indices))
+            add_max_score_to_list(temp_scores, current_score, selected_indices, selected_indices_list)
+        combined = list(zip(selected_indices_list, current_score))
+        # 使用sorted()函数对合并后的列表进行排序，key参数指定按照分数排序，reverse=True表示降序排序
+        sorted_combined = sorted(combined, key=lambda x: x[1], reverse=True)
+        inde = []
+        scores = []
+
+        for indy in sorted_combined:
+            inde.append(str(indy[0] + 1))
+            scores.append(indy[1])
+        fig = plt.figure(figsize=(24, 12))
+        ax1 = fig.add_subplot(211)
+        ax1.set_title("mRMR-FCD()")
+        ax1.plot(inde, scores)
+
+        # 设置x轴和y轴的标签
+        ax1.set_xlabel('Feature Index')
+        ax1.set_ylabel('Feature Score')
+
+        ff = []
+        for fire in inde:
+            ff.append(int(fire) - 1)
+
+        if clf == 'RF':
+            clf = RandomForestClassifier(n_jobs=-1)
+        elif clf == 'KNN':
+            clf = KNeighborsClassifier()
+        elif clf == 'DT':
+            clf = DecisionTreeClassifier()
+        elif clf == 'SVM':
+            clf = SVC()
+        elif clf == 'Naive Bayes':
+            clf = GaussianNB()
+
+        acc = []
+        # 对于index列表中的每个特征索引
+        for i in range(len(ff)):
+            # 使用前i个特征进行交叉验证
+            selected_features = X[:, [int(j) - 1 for j in ff[:i + 1]]]
+            scores = cross_val_score(clf, selected_features, y, cv=5)
+            # 计算平均准确率并添加到acc列表中
+            acc.append(scores.mean())
+        max_acc = max(acc)
+        max_index = acc.index(max_acc) + 1
+
+        ax2 = fig.add_subplot(212)
+        ax2.set_title("IFS_mRMR_FCD_Accuracy")
+        ax2.plot(max_index, max_acc, 'ro')
+        ax2.plot(acc)
+        ax2.annotate(f'({max_index}, {max_acc})', (max_index, max_acc), textcoords="offset points", xytext=(-5, -5),
+                     ha='center')
+        # 设置x轴和y轴的标签
+        ax2.set_xlabel('Top n features')
+        ax2.set_ylabel('Accuracy')
+        plt.grid(True)
+        plt.savefig('output.png')
+        return 'output.png'
+
+    elif method == 'MRMR_FCQ':
+        num_fea_inttures = len(X[0])
+        f_test_scores = [f_oneway(X[:, i], y)[0] for i in range(num_fea_inttures)]
+
+        # 添加起���特征的分数到current_score
+        current_score = [max(f_test_scores)]
+
+        # 索引从0开始
+        # start_feature_index = random.randint(0, num_features - 1)
+        # 索引从最高分数的特征开始
+        start_feature_index = f_test_scores.index(max(f_test_scores))
+
+        selected_indices = set()
+        selected_indices_list = []
+        selected_indices.add(start_feature_index)
+        selected_indices_list.append(start_feature_index)
+        pearson_score_matrix = np.zeros((num_fea_inttures, num_fea_inttures))
+        for _ in range(num_fea_int - 1):
+            temp_scores = []
+            for i in range(num_fea_inttures):
+                if i in selected_indices:
+                    temp_scores.append(-float('inf'))
+                else:
+                    f_test_score = f_test_scores[i]
+                    q = 0
+                    for j in selected_indices:
+                        # pearson score
+                        if j > i:
+                            if pearson_score_matrix[i][j] == 0:
+                                pearson_score_matrix[i][j] = np.corrcoef(X[:, i], X[:, j])[0, 1]
+                            q += pearson_score_matrix[i][j]
+                        else:
+                            if pearson_score_matrix[j][i] == 0:
+                                pearson_score_matrix[j][i] = np.corrcoef(X[:, i], X[:, j])[0, 1]
+                            q += pearson_score_matrix[j][i]
+                    temp_scores.append(f_test_score / (q / len(selected_indices)))
+            add_max_score_to_list(temp_scores, current_score, selected_indices, selected_indices_list)
+        combined = list(zip(selected_indices_list, current_score))
+
+        # 使用sorted()函数对合并后的列表进行排序，key参数指定按照分数排序，reverse=True表示降序排序
+        sorted_combined = sorted(combined, key=lambda x: x[1], reverse=True)
+        inde = []
+        scores = []
+
+        for indy in sorted_combined:
+            inde.append(str(indy[0] + 1))
+            scores.append(indy[1])
+        fig = plt.figure(figsize=(24, 12))
+        ax1 = fig.add_subplot(211)
+        ax1.set_title(str(method))
+        ax1.plot(inde, scores)
+
+        # 设置x轴和y轴的标签
+        ax1.set_xlabel('Feature Index')
+        ax1.set_ylabel('Feature Score')
+
+        ff = []
+        for fire in inde:
+            ff.append(int(fire) - 1)
+
+        if clf == 'RF':
+            clf = RandomForestClassifier(n_jobs=-1)
+        elif clf == 'KNN':
+            clf = KNeighborsClassifier()
+        elif clf == 'DT':
+            clf = DecisionTreeClassifier()
+        elif clf == 'SVM':
+            clf = SVC()
+        elif clf == 'Naive Bayes':
+            clf = GaussianNB()
+
+        acc = []
+        # 对于index列表中的每个特征索引
+        for i in range(len(ff)):
+            # 使用前i个特征进行交叉验证
+            selected_features = X[:, [int(j) - 1 for j in ff[:i + 1]]]
+            scores = cross_val_score(clf, selected_features, y, cv=5)
+            # 计算平均准确率并添加到acc列表中
+            acc.append(scores.mean())
+        max_acc = max(acc)
+        max_index = acc.index(max_acc) + 1
+
+        ax2 = fig.add_subplot(212)
+        ax2.set_title("IFS_"+str(method)+"_Accuracy")
+        ax2.plot(max_index, max_acc, 'ro')
+        ax2.plot(acc)
+        ax2.annotate(f'({max_index}, {max_acc})', (max_index, max_acc), textcoords="offset points", xytext=(-5, -5),
+                     ha='center')
+        # 设置x轴和y轴的标签
+        ax2.set_xlabel('Top n features')
+        ax2.set_ylabel('Accuracy')
+        plt.grid(True)
+        plt.savefig('output.png')
+        return 'output.png'
+
+    elif method == 'CFS':
+
+        pass
+    elif method == 'Lasso':
+        pass
+    elif method == 'Ensemble':
+        pass
+
+
+iface = gr.Interface(
+    fn=fs,
+    inputs=["file",
+            gr.inputs.Radio(['MRMR_FCD', 'MRMR_FCQ', 'CFS', 'Lasso', 'Ensemble', 'CI']),
+            gr.inputs.Number(),
+            gr.inputs.Radio(['RF', 'SVM', 'KNN', 'DT', 'Naive Bayes']),
+
+            ],
+    outputs="image",
+    examples=[
+        ["example_data.csv", 'MRMR_FCQ', 20, 'RF'],
+        ["example_data.csv", 'MRMR_FCD', 10, 'SVM'],
+        ["example_data.csv", 'MRMR_FCD', 30, 'KNN'],
+        ["example_data.csv", 'MRMR_FCQ', 50, 'DT'],
+        ["example_data.csv", 'MRMR_FCQ', 40, 'Naive Bayes'],
+
+    ],
+)
+
+iface.launch()

requirements.txt

@@ -0,0 +1,6 @@
+gradio==3.44.4
+matplotlib==3.7.2
+numpy==1.24.4
+pandas==2.1.1
+scikit_learn==1.3.1
+scipy==1.11.2