colaboratory | Jupyter CoLaboratory , goto google colab | Machine Learning library

Alright, until this problem get solved, I did this trick for my project:
I shared which files I need (like datasets) with my other accounts. For this, you should:

1. Go to your google drive (where your file is stored) then right-click on it and choose "Share"
2. Click on "Change to anyone with the link"
3. Copy link and open it in new window
4. In top-right side, click on your google accounts list and select which one you need
5. At the opened window, in top-right side click on "Add shortcut to Drive" and choose location where you want to save file in it
6. Your file now is accessible in account you did choose

I solved the same problem by running pip install statsmodels just before pip install pmdarima.

It looks like a versions conflict.

I'm pretty sure you can connect your Colab with your google drive, so you'll have easy access to your images

And then maybe you can use something like zappier to connect your google drive to you GitHub

Google Colab has nltk v3.2.5 installed, but nltk.lm (Language Modeling package) was added in v3.4.

In your Google Colab run:

Searching online; there semes to be many causes for this same problem.

In my case, setting Accelerator to None in Google Colaboratory solved this.

I managed to sidestep these numerical errors by doing the following:

1. Rescale my dependent variables and parameters so everything is within a few orders of magnitude of each other.
2. Increase the number of independent variable values at which the ODE solver returns dependent variable values. I only need 20 evenly spaced points in time to calculate my loss function, but if I want to calculate the gradient of that loss function wrt my parameters the ODE solver must return at least 60 evenly spaced points in time. I just throw out the extra points.

The training data has 60000 instances, if you split them into batches of size 256 you are left a smaller batch of size 60000 % 256 which is 96. Keras also assumes this as a batch if you dont drop it. So in train_step for this batch of size 96, the shape of real_output will be (96, 1) and the shape of fake_output will be (256, 1). As you set reduction to None in cross_entropy loss, the shape will be retained, so shape of real_loss will (96,) and shape of fake_loss will be (256,) then adding them will definitely result in an error.

You may solve this problem this way -

Just enabling GPU or TPU won't help your problem, you need to explicitly code them to run on GPU if you are not using any frameworks or libraries.

Just to make sure, your functions are located in py files not ipydb files correct?

If they are in py files, you can do one of the two methods below.

You need to change the directory to where the functions are located %cd "mnt/My Drive/Pyfunctions". Make sure you know where the python files you are trying to import are located.

You can also achieve this using the sys library,