Source code

#/******************************************************************************
# 
#  Copyright (c) 2018 Antillia.com TOSHIYUKI ARAI. ALL RIGHTS RESERVED.
#
#    This program is free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation, either version 3 of the License, or
#    (at your option) any later version.
#
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    You should have received a copy of the GNU General Public License
#    along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
#******************************************************************************/
 
# 2019/09/13

#  ZTochImageClassifierView.py

# encodig: utf-8

import sys
import os
import time
import traceback
import numpy as np

import torch
import torch.nn as nn
import torch.optim as optim
from torch.autograd import Variable
import torchvision
from torchvision import datasets, models, transforms
import json
import numpy as np
from PIL import Image

sys.path.append('../../')

from SOL4Py.ZImageClassifierView import *

############################################################
# Classifier View

class ZTorchImageClassifierView(ZImageClassifierView):  
  # Class variables

  # ClassifierView Constructor
  def __init__(self, title, x, y, width, height, datasets={"ImageModel": 0}):
    super(ZTorchImageClassifierView, self).__init__(title, x, y, width, height, datasets)
    self.resize = 256
    self.crop   = 224

    pass


  def preprocess(self, image, resize=256, crop=224):
    normalize = transforms.Normalize(
                  mean=[0.485, 0.456, 0.406],
                  std =[0.229, 0.224, 0.225])

    preprocessor = transforms.Compose([
                 transforms.Resize(resize),
                 transforms.CenterCrop(crop),
                 transforms.ToTensor(),
                 normalize
                 ])
    return preprocessor(image)


  def image_crop(self, image, resize=256, crop=224):
    crop_preprocessor = transforms.Compose([
       transforms.Resize(resize),
       transforms.CenterCrop(crop)
    ])
    return crop_preprocessor(image)


  def load_file(self, filename):
    self.ndarray = None

    try:      
      # 1 Open an original image file by PIL Image class.
      self.image = Image.open(filename)
            
      array = np.array(self.image)

      self.image_view.set_image(np.array(self.image)) 
      self.image_view.update_scale()  #2019/09/13

      # <modified date="2019/09/13">
      array = self.remove_alpha_channel(array)
      self.image = Image.fromarray(np.uint8(array))
      #</modified>
      
      self.image_tensor = self.preprocess(self.image, self.resize, self.crop)
      self.image_tensor.unsqueeze_(0)
  
      self.set_filenamed_title(filename)
      
      # 2 Crop the image.  
      self.cropped_image = self.image_crop(self.image, self.resize, self.crop)
      
      # 3 Convert the self.image to numpy ndarray. 
      self.ndarray  = np.array(self.cropped_image)

      # 4 Set self.nadarryy to the test_image_view.
      self.test_image_view.set_image(self.ndarray)

    except:
      self.write(formatted_traceback())


  def classify(self):
    self.write("------------------------------------------------------------")
    self.write("classify start.")
    self.write(self.filename)

    predictions = self.model(Variable(self.image_tensor))

    predictions = nn.functional.softmax(predictions, dim=1)
    
    TOP_FIVE = 5
    # Get top 5 predictions
    results = predictions.topk(TOP_FIVE)
    #for result in results:
    #  self.write("result {}".format(result))

    scores   = results[0].data.numpy()
    classids = results[1].data.numpy()
    maxid    = classids[0]

    score    = scores[0]
    for i in range(TOP_FIVE):
      label    = self.classes[maxid[i]]
      prob     = score[i]
      self.write("({},  {})".format(label, prob))

    self.write("Classify end.")


 
############################################################
#    
if main(__name__):

  try:
    app_name  = os.path.basename(sys.argv[0])
    applet    = QApplication(sys.argv)
  
    main_view = ZImageClassifierView(app_name, 40, 40, 900, 500)
    main_view.show ()

    applet.exec_()

  except:
    traceback.print_exc()