运行KD_DAGFM时报错：No module named 'lightgbm'

[jobs08]

使用ML-1M数据集运行KD_DAGFM_train时，报错如下：
File "D:\Tools\PyCharm2024\WorkSpace\Recbole\RecBole-master\recbole\model\exlib_recommender_init_.py", line 1, in
from recbole.model.exlib_recommender.lightgbm import LightGBM
File "D:\Tools\PyCharm2024\WorkSpace\Recbole\RecBole-master\recbole\model\exlib_recommender\lightgbm.py", line 11, in
import lightgbm as lgb
ModuleNotFoundError: No module named 'lightgbm'

recbole版本：1.1.1
ymal文件如下：

model config

model: KD_DAGFM
dataset: ml-1m
embedding_size: 10
reg_weight: 2
seed: 2020
teacher: CrossNet
phase: teacher_training
type: outer
t_depth: 3
depth: 3
alpha: 0.1
beta: 921.6
t_cin: [200, 200, 200]

dataset config

field_separator: "\t" # 指定数据集field的分隔符
seq_separator: " " # 指定数据集中token_seq或者float_seq域里的分隔符
USER_ID_FIELD: user_id # 指定用户id域（目标域）
ITEM_ID_FIELD: item_id # 指定物品id域（武器域）
RATING_FIELD: rating # 指定打分rating域
TIME_FIELD: timestamp # 指定时间域
NEG_PREFIX: neg_ # 指定负采样前缀
LABEL_FIELD: label # 指定标签域

threshold:
rating: 3

制定从什么文件里读什么列，这里就是从ml-1m.inter里面读取user_id,item_id,rating,timestamp这四列

load_col:
inter: [user_id, item_id, rating]
user: [user_id, age, gender, occupation]
item: [item_id,movie_title, release_year, genre]

training settings

epochs: 300 # 训练的最大轮数
train_batch_size: 4096 # 训练的batch_size
learner: adam # 使用的pytorch内置优化器
learning_rate: 0.001 # 学习率
train_neg_sample_args: ~

eval_step: 1 # 每次训练后做evalaution的次数
stopping_step: 30 # 控制训练收敛的步骤数，在该步骤内若选取的评测标准没有什么变化，就可以提前终止

evalution settings

eval_args:
split: {'RS': [0.8,0.1,0.1]} # 划分比例
order: RO
guoup_by: ~
mode: labeled

metrics: ["AUC","Logloss"] # 评测标准
valid_metric: AUC
eval_batch_size: 4096 # 评测的batch_size