Hi,

I am having an issue on the second step of creating an environment.

I am currently working on a M2 Mac.

I believe I downloaded the correct version of Miniconda, because when I try the platform command, it shows "macOS-14.1.1-arm64-arm-64bit."

I downloaded the yml file from the your link and tried to mirror your folders to run the same command that was in your YouTube Video: https://youtu.be/VEDy-c5Sk8Y. Also, I did place the yml file on my desktop and I am still getting the following error:
"EnvironmentFileNotFound: '/Users/tedmalloy/torch-conda.yml' file not"

Am I missing a step to make the system find the YML File?

Below: is a view of my whole terminal script:

Last login: Wed Jan 10 10:37:21 on ttys000
(base) @Teds-MacBook-Pro ~ % python
Python 3.11.5 (main, Sep 11 2023, 08:31:25) [Clang 14.0.6 ] on darwin
Type "help", "copyright", "credits" or "license" for more information.

import platform
platform.platform()
'macOS-14.1.1-arm64-arm-64bit'
quit()
(base) @Teds-MacBook-Pro ~ % conda install -y jupyter
Channels:

• defaults
  Platform: osx-arm64
  Collecting package metadata (repodata.json): done
  Solving environment: done

All requested packages already installed.

(base) tedmalloy@Teds-MacBook-Pro ~ % conda env create -f torch-
conda.yml
EnvironmentFileNotFound: '/Users/tedmalloy/torch-conda.yml' file not
found
(base) @Teds-MacBook-Pro ~ %

Running on wsl conda environment created via the install/torch-cuda.yml
Change from:

# Make use of a GPU or MPS (Apple) if one is available.
has_mps = torch.backends.mps.is_built()
device = "mps" if has_mps else "gpu" if torch.cuda.is_available() else "cpu"
print(f"Using device: {device}")

to

# Make use of a GPU or MPS (Apple) if one is available.
has_mps = torch.backends.mps.is_built()
device = "mps" if has_mps else "cuda" if torch.cuda.is_available() else "cpu"
print(f"Using device: {device}")

Where is cnn_history coming from? It's used in your notebook but is not defined and throws an error.

github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_10_4_seasonal.ipynb

• Update kernel registration in pytorch-install-aug-2023.ipynb based on the yml files.

• Noticed that the torch-conda.yml for mac still uses 3.9, but others are 3.11. suggest upgrading and renaming file to torch-m-series.yml?

• pytorch-install-aug-2023.ipynb is duplicated in the top directory and in the install folder, suggest removing from top directory

Change from:

python -m ipykernel install --user --name pytorch --display-name "Python 3.9 (torch)"

to:

python -m ipykernel install --user --name pytorch --display-name "Python 3.11 (torch)"

Hi,

Thanks for sharing this course! This is a gem!

There is a link to ImageDatagenerator in the notebook https://github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_05_4_generators.ipynb, which still points to a Keras page. However, in the subsequent lines of code, you make use of Pytorch transforms from torchvision for image augmentation examples.

When I try to install pytorch, at the end I want to check if everything is alright using the pytorch-install-aug-2023.ipynb.
But I received a warning:

Maybe 'has_mps' has been replaced by 'torch.backends.mps.is_built()'?