Error in predictrms

使用外部数据来验证cox回归模型的诺莫图时报了Error in predictrms这样的错误解决不了，请教各位大神

#Setup cox nomogram
f2 <- cph(Surv(Survival, Status_New) ~ Age_Group+Gender+Histology+Grade+T_Stage+N+, x=T, y=T, surv=T, data=a_train, time.inc=36)
nom <- nomogram(f2, fun=list(function(x) surv(12, x),
                             function(x) surv(36, x),
                             function(x) surv(60, x)),
                funlabel=c("1-year Survival Probability", "3-year Survival Probability", "5-year Survival Probability"))
plot(nom, xfrac=.2)

#External Valdation
pre <- predict(f2, newdata=a_test)
#然后就报了这个错误：
Error in predictrms: Values in Gender not in Female Male :
 [1] Male   Male   Male   Female Male   Male   Male   Male   Male   Male   Male   Male   Male   Male  
[15] Male   Male   Male   Male   Male   Female Male   Male   Male   Male   Male   Male   Male   Male  
[29] Male   Male   Male   Male   Male   Male   Male   Male   Male   Female Male   Female Male   Male  
[43] Male   Male   Male   Male   Male   Male   Male   Female Male   Male   Male   Male   Male   Male

我找到原因了，数值里面有字符串格式，直接用Excel打开清楚所有格式，使用常规格式

predictrms这个是你自己定义的，自己写的？如果是的话看看里面Gender字段的值是不是有拼写错误

遇到同样的问题，还是没解决Error in predictrms(object, newdata, type, se.fit, conf.int, conf.type, :
分类变量出现Error in predictrms: Values in PVTT not in 0 1 :