cxf712

Inssue in custom dataset traing

1

When I trainon my own data set,I encountered the following problem: tensorflow.python.framework.errors_impl.InvalidArgumentError: Found Inf or NaN global norm. : Tensor had NaN values [[node VerifyFinite/CheckNumerics (defined at ../lib/fast_rcnn/train.py:107) = CheckNumerics[T=DT_FLOAT,...

about code

2

在代码里gen_disc（）生成的训练数据没有被 h_disc.hier_train(disc_config, evl_config)函数用到，不是说 gen_disc（)函数是生成判别器的训练数据的吗，而且代码中h_disc.hier_train(disc_config, evl_config)需要用到训练集和验证集。同时在训练的过程中，一直使用while true语句，没有判断训练停止的时候吗？

minist reweight

3

why ex_wts_a is always zeros in minist.reweight.py? and it will cause the calculation of the gradient to be zero,why? ex_wts_a = tf.zeros([bsize_a], dtype=tf.float32) ex_wts_b = tf.ones([bsize_b], dtype=tf.float32) / float(bsize_b) w_dict,...