Use the method of reading mnist dataset under tensorflow always read emnist dataset show that more than a few tags

CodePudding user response:

+ 1 is also don't know how to join in the neural network

CodePudding user response:

 
The import numpy as np 
The import matplotlib. Pyplot as PLT 
The import gzip 

"" "
Just look the "parse_emnist API" and "load_data API 
"
Parse_emnist: parse the emnist file with the extension - 'gz' to.. NPZ 
Load_data: load the NPZ (the data structure is same as the 'mnist dataset of keras) 

"" "


Def read_idx3 (filename) : 
"" "
Read the given file with its name 
: param filename: extension the name of the file is' gz '
: return: images data, shape - & gt; Num rows, cols 
"" "
With gzip. Open (filename) as fo: 
Print (' Reading images') 
Buf=fo the read () 

Offset=0 # pointer 
The header=np. Frombuffer (buf, '& gt; I ', 4, offset) 
Magic_number, num_images num_rows, num_cols=header 
Print (" the magic number: {} \ nnumber of images: {}, \ nnumber of rows: {}, \ nnumber of the columns: {} "\ 
. The format (magic_number, num_images num_rows, num_cols)) 

Offset +=header. The header size *. Itemsize 
Data=https://bbs.csdn.net/topics/np.frombuffer (buf, '& gt; B ', num_rows num_images * * num_cols, offset). Reshape (
(num_images num_rows, num_cols)) 

The return data 


Def read_idx1 (filename) : 
"" "
Read the given file with its name 
: param filename: extension the name of the file is' gz '
: return: labels 
"" "
With gzip. Open (filename) as fo: 
Print (' Reading labels') 
Buf=fo the read () 

Offset=0 
The header=np. Frombuffer (buf, '& gt; I ', 2, offset) 
Magic_number, num_labels=header 
Print (" the magic number: {}, \ nnumber of labels: {} "\ 
. The format (magic_number, num_labels)) 

Offset +=header. The header size *. Itemsize 

Data=https://bbs.csdn.net/topics/np.frombuffer (buf, '& gt; B 'num_labels, offset) 
The return data 


Def show (images, labels, letter_mapping, window=(3, 4)) : 
FIG, axes=PLT. Subplots (* window, figsize=(15, 15)) 
# FIG. Set_figheight (15) 
# FIG. Set_figwidth (15) 

For a row in the range (window [0]) : 
For the column in the range (window [1]) : 
Ind window [1]=* row + column 
X=images (ind) 
Y=labels (ind) 

Axes/row (column). Imshow (x, cmap=PLT. Cm. Gray) 
Axes/row (column). Set_title (letter_mapping [y], fontsize=30) 
Axes [row] [r]. The column axis (' off ') 
PLT. Tight_layout () 

PLT. The show () 


Def parse_emnist (output_path, * arg) : 
"" "
: param output_path: 
: param arg: four path - & gt; Train_img train_label, test_img, test_label 
: return: None 
"" "
Train_img_path train_label_path, test_img_path, test_label_path=arg 
Train_img_data=https://bbs.csdn.net/topics/read_idx3 (train_img_path) 
Train_label_data=https://bbs.csdn.net/topics/read_idx1 (train_label_path) 
Test_img_data=https://bbs.csdn.net/topics/read_idx3 (test_img_path) 
Test_label_data=https://bbs.csdn.net/topics/read_idx1 (test_label_path) 

If np. Min (train_label_data)==1: 
Train_label_data=1 
 https://bbs.csdn.net/topics/train_label_data -Test_label_data=1 
 https://bbs.csdn.net/topics/test_label_data -
Np. Savez (output_path, 
X_train=train_img_data, 
Y_train=train_label_data, 
X_test=test_img_data, 
Y_test=test_label_data) 
Print (' Congratulations, your job has had done! Go to have a rest. ') 


Def load_data (path) : 
"" "Loads the EMNIST dataset. 

# the Arguments 
Path: the path where to load the dataset 

# Returns 
A Tuple of Numpy arrays: ` (x_train, y_train), (x_test, y_test) `. 
"" "
F=np. The load (path) 
X_train, y_train=f [' x_train], [' y_train] f 
X_test, y_test=f [' x_test], [' y_test] f 
F. lose () 
Return (x_train y_train), (x_test y_test) 

CodePudding user response:

I am always more than labels, training data,