Question

我有一个大型图像数据集，我想在其上运行predict_generator。由于内存问题，我不能立刻运行所有这些。想法是

通过循环显示图像范围并对其进行预测，将小组图像迭代地提供给生成器。

将预测保存到文件

稍后在循环中打开文件，以读取代码中提到的计算概率的所有预测。

validation_generator = ImageDataGenerator(rescale=1./255).flow_from_directory(path, target_size=(img_width, img_height),
                                                        batch_size=6,shuffle=False)


print("generator built")
print (counter)
#file = open('Failed.py', 'w')
#for x in file:
 #   for i in range(counter):
  #      features = model.predict_generator(validation_generator,steps=2)



print("features found")

model = Sequential()
model.add(Flatten(input_shape=(3, 3, 1536)))
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(6, activation='softmax'))
model.load_weights(top_model_weights_path)
print("top model loaded")
prediction_proba = model.predict_proba(features)
prediction_classes = model.predict_classes(features)
print(prediction_proba)
print(prediction_classes)
print("original file names")
print(validation_generator.filenames)

问题是如何将不同的预测保存在一个文件中。我试过为文件创建一个for循环，但不知道它应该如何工作？如果有人可以为所定义的目标提供暗示，那就太好了。

Answer 1

预测和保存

i = 0
maximumPredictions = ??
for x,y in generator: #if the generator doesn't have y, use only "for x in..."
    predictions = model.predict(x)
    numpy.save('predictions/prediction' + str(i) +".npy", predictions)
    i+=1

    if i == maximumPredictions:
        break;

加载和处理

files = [name for name in os.listdir('predictions')]
for file in files:
    prediction = numpy.load('predictions/"+file)

    #do what you want with the loaded predictions.

如何通过循环将predict_generator输出写入文件？

1 个答案: