import torch import numpy as np from sklearn.metrics import accuracy_score, precision_recall_fscore_support, confusion_matrix, roc_auc_score def evaluate_model(model, dataloader, device="cpu"): """ Standard comprehensive evaluation calculating Accuracy, Precision, Recall, Specificity, F1, and AUC. """ model.eval() all_preds = [] all_targets = [] all_probs = [] with torch.no_grad(): for inputs, targets in dataloader: inputs = inputs.to(device) outputs, _ = model(inputs) probs = torch.softmax(outputs, dim=1)[:, 1] preds = torch.argmax(outputs, dim=1) all_preds.extend(preds.cpu().numpy()) all_targets.extend(targets.numpy()) all_probs.extend(probs.cpu().numpy()) all_preds = np.array(all_preds) all_targets = np.array(all_targets) all_probs = np.array(all_probs) acc = accuracy_score(all_targets, all_preds) prec, rec, f1, _ = precision_recall_fscore_support(all_targets, all_preds, average='binary') tn, fp, fn, tp = confusion_matrix(all_targets, all_preds).ravel() spec = tn / (tn + fp) auc_val = roc_auc_score(all_targets, all_probs) return { "accuracy": acc, "precision": prec, "recall_sensitivity": rec, "specificity": spec, "f1_score": f1, "roc_auc": auc_val, "confusion_matrix": {"TP": int(tp), "TN": int(tn), "FP": int(fp), "FN": int(fn)} }