AI
haikuc
这个作者很懒,什么都没留下…
展开
-
tf.train.batch,tf.train.batch_join
用到的测试数据如下:t1 = tf.constant(np.array([0])) # (1, ) datat2 = tf.constant(np.array([3]))t3 = tf.constant(np.array([4]))t4 = tf.constant(np.array([5]))t0 = tf.constant(np.array([[0], [3], [4], [5]]))...原创 2019-02-02 00:11:10 · 387 阅读 · 0 评论 -
训练Inceptionv3遇到的坑
InvalidArgumentError (see above for traceback): Cannot assign a device for operation ‘gradients/aux_loss/xentropy_grad/LogSoftmax’: Could not satisfy explicit device specification ‘/device:GPU:0’ beca...转载 2019-01-24 22:12:07 · 926 阅读 · 0 评论 -
MTCNN 测试源码
代码实现(MXNet):https://github.com/pangyupo/mxnet_mtcnn_face_detection从代码主体main.py开始detector = MtcnnDetector(model_folder='model', ctx=mx.cpu(0), num_worker = 4 , accurate_landmark = False) # 创建检测器 ...原创 2019-01-10 20:17:40 · 876 阅读 · 0 评论 -
吴恩达 yolo.h5生成方法
yolo.h5生成方法git clone https://github.com/allanzelener/yad2k.gitcd yad2k方法一 复制 conda environment:conda env create -f environment.ymlsource activate yad2k方法二 安装python包pip install numpy h5py pillo...原创 2018-12-31 20:44:01 · 2406 阅读 · 0 评论