When researching which algorithm to use in the text classification I stumbled upon usage of DNNClassifier. As I understand it is different from text classifiers from nltk. I already have my text embedded, but can`t quite figure what is the 'input_fn' argument passed to fit function. What is this function and how do I use DNNClassifier properly?
input_fn is a function where you convert your data to 'tf.data.Dataset' and configure how your data is served to the model (shuffling, repeats, etc.).
Here's an example how can it be defined:
editeddef train_input_fn():
dataset = tf.data.Dataset.from_tensor_slices((dict(x_train), y_train))
dataset = dataset.shuffle(1000).repeat().batch(10)
return dataset
You can find out more in this article: https://medium.com/datadriveninvestor/tensorflow-dnnclassifier-4e68df3df00
Hope, that helps. Cheers.
Just drop us an email to ...
Show more 
Log in with Google