1. cat_features must be integer or string, real number values and NaN values should be converted to string.
catfeat = [33, 34, 36, 49, 50]
model_k.fit(train_df, label, cat_features = catfeat)
主要问题是 train_df[catfeat]
里面有的数不是整数或者string。需要先对数据做一次转换
for i in catfeat:
train_df[i] = train_df[i].astype(np.int)