parser.add_argument("--ep", default=100, type=int, help="训练的总轮数") parser.add_argument('--re', default=False, help='是否从checkpoint处开始训练') parser.add_argument("--ckpt", default="./checkpoints/res/CNN-LSTM-60.t7", help="checkpoint的地址") parser.add_argument("--gpu", default=False, help="是否使用GPU进行训T练") parser.add_argument("--cuda", default="0,1,2,3", type=str, help="用于训练GPU的代号") cfg, unknown = parser.parse_known_args() os.environ["CUDA_VISIBLE_DEVICES"] = cfg.cuda # 配置环境中的GPU start_epoch = 0 # 开始训练的轮数 best_test_acc = 0 # 最好的测试准确率 best_test_acc_epoch = 0 # 最好的测试准确率时的轮数 print("--------------加载训练数据中------------") trainRoot = cfg.dr + "/train" X_train, y_train, mask_train = dataLoader.window_process(trainRoot, "./data/r.txt", cfg.dl, 20, cfg.ts, cfg.lb) X_train = np.array(X_train) y_train = np.array(y_train) mask_train = np.array(mask_train) y_train = to_categorical(y_train, 22) # 对数据进行随机打乱 permutation = np.random.permutation(X_train.shape[0]) X_train = X_train[permutation, :, :, :] y_train = y_train[permutation, :] mask_train = mask_train[permutation, :, :] trainData_sum = len(X_train) # 训练样本总数 print("--------------加载训练数据完成------------") print("--------------加载测试数据中------------")
parser.add_argument('--re', default=True, help='是否从checkpoint处开始训练') parser.add_argument("--ckpt", default="./checkpoints/res1_52/CNN-LSTM-82.t7", help="checkpoint的地址") parser.add_argument("--gpu", default=False, help="是否使用GPU进行训T练") parser.add_argument("--cuda", default="0,1,2,3", type=str, help="用于训练GPU的代号") cfg, unknown = parser.parse_known_args() os.environ["CUDA_VISIBLE_DEVICES"] = cfg.cuda # 配置环境中的GPUF start_epoch = 0 # 开始训练的轮数 best_test_acc = 0 # 最好的测试准确率 best_test_acc_epoch = 0 # 最好的测试准确率时的轮数 print("--------------加载测试数据中------------") testRoot = cfg.dr + "/test" X_test, y_test, mask_test = dataLoader.window_process(testRoot, cfg.dl, 160, cfg.ts, cfg.lb) X_test = np.array(X_test) y_test = np.array(y_test) mask_test = np.array(mask_test) y_test = to_categorical(y_test, 22) # 对数据进行随机打乱 permutation = np.random.permutation(X_test.shape[0]) X_test = X_test[permutation, :, :, :] y_test = y_test[permutation, :] mask_test = mask_test[permutation, :, :] testData_sum = len(X_test) # 训练样本总数 print("--------------加载测试数据完成------------") use_cuda = torch.cuda.is_available() # 环境中是否有GPU
parser.add_argument("--ckpt", default="./checkpoints/res1_52/CNN-LSTM-82.t7", help="checkpoint的地址") parser.add_argument("--gpu", default=False, help="是否使用GPU进行训T练") parser.add_argument("--cuda", default="0,1,2,3", type=str, help="用于训练GPU的代号") cfg, unknown = parser.parse_known_args() os.environ["CUDA_VISIBLE_DEVICES"] = cfg.cuda # 配置环境中的GPUF start_epoch = 0 # 开始训练的轮数 best_test_acc = 0 # 最好的测试准确率 best_test_acc_epoch = 0 # 最好的测试准确率时的轮数 print("--------------加载训练数据中------------") trainRoot = cfg.dr + "/train" X_train, y_train, mask_train = dataLoader.window_process( trainRoot, cfg.rr, cfg.dl, cfg.ts, cfg.lb) X_train = np.array(X_train) y_train = np.array(y_train) mask_train = np.array(mask_train) y_train = to_categorical(y_train, 2) # 对数据进行随机打乱 permutation = np.random.permutation(X_train.shape[0]) X_train = X_train[permutation, :, :, :] y_train = y_train[permutation, :] mask_train = mask_train[permutation, :, :] trainData_sum = len(X_train) # 训练样本总数 print("--------------加载训练数据完成------------") print("--------------加载测试数据中------------")
if iter == 100: P = P / 4. # Return solution return Y if __name__ == "__main__": testRoot = "./data/test" dl = 0 ts = 52 lb = 1 print("--------------加载测试数据中------------") X_test, y_test, mask_test = dataLoader.window_process( testRoot, dl, 160, ts, lb) X_test = np.array(X_test) y_test = np.array(y_test) mask_test = np.array(mask_test) y_test = to_categorical(y_test, 22) # 对数据进行随机打乱 permutation = np.random.permutation(X_test.shape[0]) X_test = X_test[permutation, :, :, :] y_test = y_test[permutation, :] mask_test = mask_test[permutation, :, :] testData_sum = len(X_test) # 训练样本总数 print("--------------加载测试数据完成------------") inputs = torch.Tensor(X_test)