fake-news | Lists of fake news sites | Mock library

I wrote a function for this purpose in my package rwhatsapp.

As your example is a whatsapp dataset, you can test it directly using the package (install via remotes::install_github("JBGruber/rwhatsapp"))

You are iterating over each word and appending them one at a time to the list, which is why it is flattening. Instead of appending each word you need to append the filtered list. This is probably clearer if you do it as a list comprehension:

I think you have one feature and 3608 records, but the code thinks there is one sample with 3608 features.

change the code where x and y are defined as follows.

So from my understanding, you are trying to close the window when the "DONE" button is clicked, if so: You can remove the onclick attribute from the button tag and attach an EventListener to the button itself like so:

There are some things you could try to speed things up:

It is usually faster, check available layers here keras.layers.CuDNNLSTM is what you are after.

You can use 1 dimensional convolution with kernel_size specifying how many words should be taken into account and stride specifying the jump of moving window. For kernel_size=3 and stride=3, padding="SAME" it would drop your dimensionality three times.

You may stack more convolutional layers.

On top of that you can still employ LSTM normally.

You may go with 1d convolutions and pooling for classification, RNNs are not the only way.

On the upside: you will not encounter vanishing gradients (could be mitigated a little by Bidirectional LSTM as well).

On the downside: you will lose strict dependence between words, though it shouldn't be much of a problem for binary classification (I suppose it's your goal).

Your JSON has 2 double quotes within a string value. See screenshot.

I believe the JSON when returned from the source is incorrect or the character encoding you are providing is not right.

Why are you expecting those two command have the same result, if fake-news2 doesn't exist locally?