Using flow_from_dataframe what is the correct "value" for y_col

user_6510273 0
Viewed 6134

I am reading in a csv file with pandas, and give the column names stored in colname 

colnames=['file', 'label'] 
# Read data from file 
data = pd.read_csv('./Hand_Annotations_2.csv',names=colnames, header=None)
# Preview the first 5 lines of the loaded data 
data.head()

Then, I use ImageDataGenerator() and flow_fromdataframe() to get batches of data

train_generator=datagen.flow_from_dataframe(dataframe=data,
                                            directory=None, 
                                            x_col=colnames[0], 
                                            y_col=colnames[1],
                                            class_indices=IDmap,
                                            class_mode="categorical", target_size=(224,224), batch_size=32)

But I get an error, as below:

TypeError: If class_mode="categorical", y_col="label" column values must be type string, list or tuple.

But my y_col is a string. I get the same error if I just enter "label". It also seems to work with x_col.

Can someone point me to my mistake?

Thanks

Solution

read csv with dtype e.g. str:

data = pd.read_csv('./Hand_Annotations_2.csv',dtype=str,names=colnames, header=None)
tf.keras keras tensorflow generator
2 Answers

You can check the code in github for what it is actually being checked, specially here. Keras preprocessing is checking that all the values have allowed types (str, list or tuple), so the check is failing, you have to check that all values of the y_col column have the right type.

user_3491300

If your labels are integer, set class_mode='raw'

user_151815570
Related