1.没用下载训练集导致出错，爆错如下。

这个时候需要去比赛官网下载对应的初赛训练集

unzip -d /mnt/workspace/sais_third_new_energy_baseline/data /mnt/workspace/sais_third_new_energy_baseline/初赛训练集.zip

在命令行执行这个命令解压

2.没定义测试集

test = pd.concat(data, axis=0).reset_index(drop=True)

利用这行代码构建test测试集。

3.预测时提供的数据特征数量（8 个特征）与训练模型时使用的数据特征数量（10 个特征）不一致

[LightGBM] [Fatal] The number of features in data (8) is not the same as it was in training data (10).
You can set ``predict_disable_shape_check=true`` to discard this error, but please be aware what you are doing.
LightGBMError: The number of features in data (8) is not the same as it was in training data (10).
You can set ``predict_disable_shape_check=true`` to discard this error, but please be aware what you are doing.

test = feature_combine(test)
test = test.dropna().reset_index(drop=True)

之前未建立test，所有未给test做特征数量的改造，别忘了这点。