拟合 keras 模型时发生无法识别的关键字参数

Question

I am trying to use ImageDataGenerator to label the data. However, when I fitting models, an error showed.

TypeError: Unrecognized keyword arguments: {'generator': <keras_preprocessing.image.dataframe_iterator.DataFrameIterator object at 0x0000021CC81BBE48>}

Here is my full code :

import area:

import numpy as np
import keras  
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras.callbacks import ModelCheckpoint
import pandas as pd
from keras_preprocessing.image import ImageDataGenerator

Data loading and labeling:

dataset=pd.read_csv('labelset.csv')
columns=['Agonist','Antagonist','Binding']
datagen=ImageDataGenerator(rescale=1./255.)
test_datagen=ImageDataGenerator(rescale=1./255.)

train_generator=datagen.flow_from_dataframe(
dataframe=dataset[:1500],
directory ='C:/Users/j7042/Desktop/Training set/' ,
x_col="Filenames",
y_col=columns,
batch_size=1,
seed=42,
shuffle=True,
class_mode="raw",
target_size=(500,500))

test_generator=test_datagen.flow_from_dataframe(
dataframe=dataset[1500:],
directory='C:/Users/j7042/Desktop/Training set/',
x_col="Filenames",
y_col=columns,
batch_size=1,
seed=42,
shuffle=True,
class_mode="raw",
target_size=(500,500))

Model area :

model = Sequential()
model.add(Conv2D(8, kernel_size=(3, 3),activation='relu',input_shape=[500,500,3]))
model.add(Conv2D(16,(3,3),activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dense(128, activation='relu'))
model.add(Dense(3, activation='softmax'))
model.compile(loss=keras.losses.categorical_crossentropy,
              optimizer=keras.optimizers.Adadelta(),
              metrics=['acc'])

Fitting:

STEP_SIZE_TRAIN=train_generator.n//train_generator.batch_size
STEP_SIZE_VALID=valid_generator.n//test_generator.batch_size


model.fit(generator=train_generator,steps_per_epoch=STEP_SIZE_TRAIN,validation_data=valid_generator,validation_steps=STEP_SIZE_VALID,batch_size=1, epochs=10)

Full error message :

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-59-e499c6763112> in <module>
      3 
      4 
----> 5 model.fit(generator=train_generator,steps_per_epoch=STEP_SIZE_TRAIN,validation_data=valid_generator,validation_steps=STEP_SIZE_VALID,batch_size=1, epochs=10)

~\anaconda3\envs\keras environment\lib\site-packages\keras\engine\training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)
   1116             epochs = kwargs.pop('nb_epoch')
   1117         if kwargs:
-> 1118             raise TypeError('Unrecognized keyword arguments: ' + str(kwargs))
   1119 
   1120         if x is None and y is None and steps_per_epoch is None:

TypeError: Unrecognized keyword arguments: {'generator': <keras_preprocessing.image.dataframe_iterator.DataFrameIterator object at 0x0000021CC81BBE48>}

Here is a part of my labelset.csv: enter image description here

I have no idea about this error, hoping someone can help me.

Answer 1

The error is because you're using generator parameter on model.fit , which is not a valid argument (see Keras documentation on Fit ).

You can call model.fit with a generator, but the generator is your x parameter:

model.fit(x=train_generator,steps_per_epoch=STEP_SIZE_TRAIN,validation_data=valid_generator,validation_steps=STEP_SIZE_VALID,batch_size=1, epochs=10)