Unfurling Cheetah

1,  Run one of the dogs and cats classification task on my computer( intel i5) for almost a week, and then i realized that the program stopped when it entered into the sleep mode !! I had to reset the power setting to run it again. Hopefully it will end soon in several days. 2,  It is big surprise to find that in Scikit family,  there are so many useful members there already, waiting for me to explore THEM!  I found it is almost a wonderful wheel which i previously would like to reinvent: scikit-spectra. 3, Even though I think i still want to write some GUI to facilitate my works.

Borrowed this book from library.  <Python Playground: Geeky Projects for the Curious Programmer> Really there is fun to run those codes.  Here is the one i showed to my kids:

The simple definition of epoch is , An epoch is  one forward pass  and  one backward pass  of  all  training examples. Note : epoch and iterations are two  different  things. For example , for a set of 1000 images and a batch size of 10, each iteration would process 10 images for a total of 100 such iterations to go over the entire set. This is called  one epoch . Training can go on for 100s of epochs.(https://stackoverflow.com/questions/31155388/meaning-of-an-epoch-in-neural-networks-training): One epoch consists of  one  full training cycle on the training set. Once every sample in the set is seen, you start again - marking the beginning of the 2nd epoch.

1, Ellipsis >>> a = np.arange(6).reshape(2,3) >>> a array([[0, 1, 2], [3, 4, 5]]) >>> for x in np.nditer(a, op_flags=['readwrite']):                                                       x[...] = 2 * x >>> a array([[ 0, 2, 4], [ 6, 8, 10]]) 2, array or matrix Operator  * ,  dot() , and  multiply() : For  array ,  ‘``*``’ means element-wise multiplication , and the  dot()  function is used for matrix multiplication. For  matrix ,  ‘``*``’ means matrix multiplication , and the  multiply()  function is used for element-wise multiplication. Handling of vectors (one-dimensional arrays) For  array , the  vector shapes 1xN, Nx1, and N are all different things . Operations like  A[:,1]  return a one-dimensional array of shape N, not...

1, From last week, I started to watch AndrewNg's RNN. 2, Also found that Hung-yi Lee's(NTU) youtube channel very informative.      https://www.youtube.com/watch?v=rTqmWlnwz_0 3, Plan to Run Keras on Kaggle! (hopefully can run "Dogs and Cats" existing codes and get result.)

The  Siamese method , or  De la Loubère method , is a simple method to construct any size of  n -odd  magic squares  (i.e. number squares in which the sums of all rows, columns and diagonals are identical). (from wiki) In  linear algebra , the  Gram matrix  ( Gramian matrix  or  Gramian ) of a set of vectors  {\displaystyle v_{1},\dots ,v_{n}}  in an  inner product space  is the  Hermitian matrix  of  inner products , whose entries are given by  {\displaystyle G_{ij}=\langle v_{i},v_{j}\rangle } . [1] An important application is to compute  linear independence : a set of vectors is linearly independent if and only if the  Gram determinant  (the  determinant  of the Gram matrix) is non-zero. It is named after  Jørgen Pedersen Gram .(from wiki)