update

1d-qsar/cuda/extract_log_CV_MSE.py

@@ -0,0 +1,49 @@
+import re
+import matplotlib.pyplot as plt
+
+def extract_cv_mse(log_path):
+    """
+    从日志文件中提取特征数和对应的CV MSE。
+    日志行格式示例：
+    "Current number of features: 156, CV MSE: 4.9253"
+    """
+    feature_counts = []
+    cv_mses = []
+    pattern = re.compile(r"Current number of features:\s*(\d+),\s*CV MSE:\s*([\d\.]+)")
+    with open(log_path, "r") as f:
+        for line in f:
+            match = pattern.search(line)
+            if match:
+                feature_counts.append(int(match.group(1)))
+                cv_mses.append(float(match.group(2)))
+    return feature_counts, cv_mses
+
+def plot_cv_mse(log_path, output_image="cv_mse_curve.png"):
+    """
+    根据日志文件中的数据绘制 CV MSE 曲线，并保存图片
+    """
+    feature_counts, cv_mses = extract_cv_mse(log_path)
+    
+    if not feature_counts:
+        print("没有在日志文件中找到有效的 'Current number of features' 数据。")
+        return
+
+    plt.figure(figsize=(8,6))
+    plt.plot(feature_counts, cv_mses, marker='o', linestyle='-')
+    plt.xlabel("Number of Features")
+    plt.ylabel("CV MSE")
+    plt.title("CV MSE vs. Number of Features")
+    plt.grid(True)
+    plt.savefig(output_image, dpi=300)
+    plt.close()
+    print(f"CV MSE 曲线图已保存为: {output_image}")
+
+if __name__ == "__main__":
+    import sys
+    if len(sys.argv) < 2:
+        print("用法: python script.py <log_file_path>")
+    else:
+        log_file = sys.argv[1]
+        plot_cv_mse(log_file)
+
+# python 1d-qsar/cuda/RFE_cuml_permutation_update.log