lightgbm gpu在ubuntu/linuxmint下的安装与验证的真正方法

ゞ浴缸里的玫瑰 2023-06-24 09:27 83阅读 0赞

垃圾教程太多

sudo apt-get install cmake
    sudo apt-get install libboost-dev

下载项目  
[https://github.com/microsoft/LightGBM][https_github.com_microsoft_LightGBM]  
解压zip

cd  Lightgbm-master
    mkdir build ; cd build
    cmake -DOpenCL_LIBRARY=/usr/local/cuda/lib64/libOpenCL.so -DOpenCL_INCLUDE_DIR=/usr/local/cuda/include -DUSE_GPU=1 ..
    make -j4

依据具体情况修改上述libOpenCL.so和include的路径  
安装python版

cd ../python-package
    python setup.py install --precompile

验证

cd ../
    vim test.py

输入

import lightgbm as lgb
    import time
    
    #讀取data資料
    train_data = lgb.Dataset(r'LightGBM-master/examples/parallel_learning/binary.train')
    
    #設定lightgbm參數，gpu版本測試
    params = {'max_bin': 63,
    'num_leaves': 255,
    'learning_rate': 0.1,
    'tree_learner': 'serial',
    'task': 'train',
    'is_training_metric': 'false',
    'min_data_in_leaf': 1,
    'min_sum_hessian_in_leaf': 100,
    'ndcg_eval_at': [1,3,5,10],
    'sparse_threshold': 1.0,
    'device': 'gpu',
    'gpu_platform_id': -1,
    'gpu_device_id': -1}
    t0 = time.time()
    
    #執行模型訓練
    gbm = lgb.train(params, train_set=train_data, num_boost_round=10,
    valid_sets=None, valid_names=None,
    fobj=None, feval=None, init_model=None,
    feature_name='auto', categorical_feature='auto',
    early_stopping_rounds=None, evals_result=None,
    verbose_eval=True,
    keep_training_booster=False, callbacks=None)
    t1 = time.time()
    #顯示執行時間
    print('gpu version elapse time: {}'.format(t1-t0))
    
    #設定lightgbm參數，cpu版本
    params = {'max_bin': 63,
    'num_leaves': 255,
    'learning_rate': 0.1,
    'tree_learner': 'serial',
    'task': 'train',
    'is_training_metric': 'false',
    'min_data_in_leaf': 1,
    'min_sum_hessian_in_leaf': 100,
    'ndcg_eval_at': [1,3,5,10],
    'sparse_threshold': 1.0,
    'device': 'cpu'
    }
    t0 = time.time()
    
    #執行模型訓練
    gbm = lgb.train(params, train_set=train_data, num_boost_round=10,
    valid_sets=None, valid_names=None,
    fobj=None, feval=None, init_model=None,
    feature_name='auto', categorical_feature='auto',
    early_stopping_rounds=None, evals_result=None,
    verbose_eval=True,
    keep_training_booster=False, callbacks=None)
    t1 = time.time()

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3NoZXJwYWh1_size_16_color_FFFFFF_t_70]

[https_github.com_microsoft_LightGBM]: https://github.com/microsoft/LightGBM
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3NoZXJwYWh1_size_16_color_FFFFFF_t_70]: https://img-blog.csdnimg.cn/20191222164435905.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3NoZXJwYWh1,size_16,color_FFFFFF,t_70