auc_roc.py

def roc_curve(y,pred):
    TPR = [0.0]
    FPR = [0.0]
    thresholds = np.arange(0.01,1.00+0.01,0.01)
    for th in thresholds:
        TPR.append(recall(y,pred,th))
        FPR.append(fpr(y,pred,th))
        
    return TPR,FPR,thresholds
  
def auc_score(y,pred):
    TPR,FPR,thresholds = roc_curve(y,pred)
    TPR = TPR/TPR[-1]
    FPR = FPR/FPR[-1]
    tpr_diff = np.r_[np.diff(TPR),0.0]
    fpr_diff = np.r_[np.diff(FPR),0.0]
    auc = np.dot(TPR,fpr_diff) + np.dot(tpr_diff,fpr_diff)/2
    
def plot_roc_curve(y,pred):
    TPR,FPR,Thresholds = roc_curve(y,pred)
    roc_auc = auc_score(y,pred)
    plt.figure(figsize=(10,10))
    plt.plot(FPR,TPR,linewidth=2,label='{} (AUC={:.3f})'.format(label, roc_auc))
    plt.plot([0.0,1.0], [0.0,1.0], linestyle='dashed', color='red', linewidth=2, label='random')
    plt.xlim(0.0, 1.0)
    plt.ylim(0.0, 1.0)
    plt.xlabel("FPR",fontsize=12)
    plt.ylabel("TPR",fontsize=12)
    plt.legend(fontsize=10, loc='best')
    plt.title("ROC Curve",fontsize=12)
    plt.savefig('ROC_Curve.png')