def main(): args = parse_args() model = load_checkpoint(args.checkpoint) cat_to_name = load_cat_name = load_cat_names(args.category_names) img_path = args.filepath probs, classes = predict(img_path, model, int(args.top_k),gpu) lables = [cat_to_name[str(index)] for index in classes] probability = probs print ('File selected: ' + img_path) print (labels) print (probability) i=0 while i < len(labels): print ("{} with a probability of {}".format(labels[i],probability[i])) i = i + 1
def main(): args = parse_args() gpu = args.gpu model = load_checkpoint(args.checkpoint) cat_to_name = load_cat_names(args.category_names) img_path = args.filepath probs, classes = predict(img_path, model, int(args.top_k), gpu) labels = [cat_to_name[str(index)] for index in classes] probability = probs print('File selected: ' + img_path) print(labels) print(probability) i = 0 # this prints out top k classes and probs as according to user while i < len(labels): print("{} with a probability of {}".format(labels[i], probability[i])) i += 1 # cycle through
def main(): args = parse_args() gpu = args.gpu model = load_checkpoint(args.checkpoint) cat_to_name = load_cat_names(args.category_names) if args.filepath == None: img_num = random.randint(1, 102) image = random.choice( os.listdir('./flowers/test/' + str(img_num) + '/')) image_path = './flowers/test/' + str(img_num) + '/' + image prob, classes = predict(image_path, model, int(args.top_k), gpu) print('Image selected: ' + str(cat_to_name[str(img_num)])) else: image_path = args.filepath prob, classes = predict(image_path, model, int(args.top_k), gpu) print('File selected: ' + image_path) print(prob) print(classes) print([cat_to_name[x] for x in classes])
def main(): args = parse_args() gpu = args.gpu model = load_checkpoint(args.checkpoint) cat_to_name = load_cat_names(args.category_names) if args.filepath: img_path = args.filepath else: print('Cannot run prediction ..') img_path = input("Please provide path to image: ") probs, classes = predict(img_path, model, int(args.top_k), gpu) print('\n======') print('The filepath of the selected image is: ' + img_path, '\n') print('The top K CLASSES for the selected image are: \n', classes, '\n') print('The top K PROBABILITIES for the selected image are: \n ', probs, '\n') print('The top K CATEGORY NAMES for the selected image are: \n', [cat_to_name[x].title() for x in classes]) print('======\n')
def main(): in_arg = get_input_args() gpu = in_arg.gpu model = load_checkpoint(in_arg.checkpoint) cat_to_name = load_cat_names(in_arg.category_names) if in_arg.filepath == None: image_num = random.randint(1, 102) image = random.choice(os.listdir('./flowers/test/' + str(image_num) + '/')) img_path = './flowers/test/' + str(image_num) + '/' + image prob, classes = predict(img_path, model, in_arg.top_k, gpu) print("Selected Image is: " + str(cat_to_name[str(image_num)])) else: # Show random predicted image of the original image (displayed above) from a particular subfolder #image = random.choice(os.listdir('./flowers/test/' + str(image_num) + '/')) img_path = in_arg.filepath prob, classes = predict(img_path, model, in_arg.top_k, gpu) print("Selected Image is: " + img_path) print("\nProbabilities are: \n", prob) print("\nClasses are: \n", classes) print("\nFlowers names are: \n", [cat_to_name[i] for i in classes])
img_torch = img_torch.float() if gpu == 'gpu': with torch.no_grad(): output = model.forward(img_torch.cuda()) else: with torch.no grad(): output = model.forward(img_torch) probability = F.softmax(output.data,dim=1) probos = np.array(probability.topk(topk)[0][0]) index-to_class = {val: key for key, val in model.class_to_idx()} top_classes = [np.int(index_to_class[each] for np.array(probability.topk(topk)[1][0])] return probs, top_classes def main(): args = parse_arg() model = load.checkpoint(args.checkpoint) cat_to_name = load_cat_name = load_cat_names(args.category_names) img_path = args.filepath probs, classes = predict(img_path, model, int(args.top_k),gpu) lables = [cat_to_name[str(index)] for index in classes] probability = probs print ('File selected: ' + img_path) print (labels) print (probability) i=0 while i < len(labels): print ("{} with a probability of {}.format(labels[i], probability[i]])) i = i + 1 if __name__ == "__main__": main()