From 8f5c6574343d3100516ccfa6b40aa2a876b9d950 Mon Sep 17 00:00:00 2001 From: Maddy Underwood <167196745+madeline-underwood@users.noreply.github.com> Date: Wed, 1 Jan 2025 04:59:42 +0000 Subject: [PATCH 1/3] Post-production tweaks. --- .../pytorch-digit-classification-arch-training/intro.md | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/intro.md b/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/intro.md index 0d70ca8c4c..683b9eb999 100644 --- a/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/intro.md +++ b/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/intro.md @@ -11,7 +11,7 @@ layout: "learningpathall" Meta AI have designed an Open Source deep learning framework called PyTorch, that is now part of the Linux Foundation. -PyTorch provides a flexible and efficient platform for building and training neural networks. It has a dynamic computational graph that allows users to modify the architecture during runtime, making debugging and experimentation easier, and therefore makes it popular among developers. +PyTorch provides a flexible and efficient platform for building and training neural networks. It has a dynamic computational graph that allows users to modify the architecture during runtime, making debugging and experimentation easier, and therefore making it popular among developers. PyTorch provides a more flexible, user-friendly deep learning framework that reduces the limitations of static computational graphs found in earlier tools, such as TensorFlow. @@ -110,8 +110,11 @@ python3 -m ipykernel install --user --name=pytorch-env To ensure everything is set up correctly, follow these next steps: 1. Open Visual Studio Code. + 2. Click **New file**, and select `Jupyter Notebook .ipynb Support`. + 3. Save the file as `pytorch-digits.ipynb`. + 4. Select the Python kernel you created earlier, `pytorch-env`. To do so, click **Kernels** in the top right-hand corner. Then, click **Jupyter Kernel...**, and you will see the Python kernel as shown below: ![img1 alt-text#center](Figures/1.png "Figure 1: Python kernel.") From 61f60d9c3c6de1774fa5684f1e047a6a5a007e1d Mon Sep 17 00:00:00 2001 From: Maddy Underwood <167196745+madeline-underwood@users.noreply.github.com> Date: Wed, 1 Jan 2025 05:01:51 +0000 Subject: [PATCH 2/3] Reduce subheading size by 1. --- .../pytorch-digit-classification-arch-training/model.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/model.md b/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/model.md index 1db5d1e793..e189a5c708 100644 --- a/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/model.md +++ b/content/learning-paths/cross-platform/pytorch-digit-classification-arch-training/model.md @@ -25,7 +25,7 @@ The total number of trainable parameters for this network is calculated as follo In total, the network has 102,762 trainable parameters. -# Implementation +## Implementation To implement the model, supplement the `pytorch-digits.ipynb` notebook with the following statements: @@ -132,7 +132,7 @@ The output is still a probability distribution over the 10 digit classes (0-9), Technically, the code will run without errors as long as you provide it with an input image of the correct dimensions, which is 28x28 pixels. The model can accept input, pass it through the layers, and return a prediction - a vector of 10 probabilities. However, the results are not useful until the model is trained. -# What have you learned so far? +## What have you learned so far? You have successfully defined and initialized a feedforward neural network using PyTorch. From 2026daf4d01c1189349aceb55f4e8f3c7a34e05b Mon Sep 17 00:00:00 2001 From: Maddy Underwood <167196745+madeline-underwood@users.noreply.github.com> Date: Thu, 2 Jan 2025 03:41:22 +0000 Subject: [PATCH 3/3] Update find-hardware.md --- .../servers-and-cloud-computing/intro/find-hardware.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/content/learning-paths/servers-and-cloud-computing/intro/find-hardware.md b/content/learning-paths/servers-and-cloud-computing/intro/find-hardware.md index 1a2e5e80ff..0cceff7c99 100644 --- a/content/learning-paths/servers-and-cloud-computing/intro/find-hardware.md +++ b/content/learning-paths/servers-and-cloud-computing/intro/find-hardware.md @@ -7,7 +7,7 @@ Server hardware, based on Arm Neoverse processors, is available from cloud servi ## Cloud Service Providers -Creating an account with a cloud service provider (CSPs) is the easiest way to get started with Arm. CSPs offer introductory free credits to start learning cloud services. A pay-as-you-go model makes it easy to try out Arm hardware at little to no cost. +Creating an account with a cloud service provider (CSP) is the easiest way to get started with Arm. CSPs offer introductory free credits to start learning cloud services. A pay-as-you-go model makes it easy to try out Arm hardware at little cost, if any. Software developers often try Arm hardware as a way to reduce cost and improve application performance. @@ -21,7 +21,7 @@ Cloud providers offer Arm instances based on Neoverse processors. For example: - [Scaleway](https://www.scaleway.com/en/cost-optimized-instances-based-on-arm/) - [Hetzner Cloud](https://www.hetzner.com/news/arm64-cloud) -Free tier offers are currently available. +Free tier offers are currently available: - [Amazon EC2 t4g.small instances powered by AWS Graviton2 processors are free until Dec 31st 2024](https://aws.amazon.com/ec2/instance-types/t4/) - [Oracle free tier includes up to 4 instances of ARM Ampere A1 Compute which are always free](https://www.oracle.com/cloud/free/) @@ -46,7 +46,7 @@ Below is a list of Arm cloud partners that have existing academic offers for stu ## Arm SystemReady Certified hardware -[Arm SystemReady](https://www.arm.com/architecture/system-architectures/systemready-certification-program) is a program that certifies that systems meet the SystemReady standards, giving confidence that operating systems (OS) and subsequent layers of software just work. +[Arm SystemReady](https://www.arm.com/architecture/system-architectures/systemready-certification-program) is a program that certifies that systems meet the SystemReady standards, giving you confidence that operating systems (OS) and subsequent layers of software just work. You can find a full list of SystemReady SR Certified Systems on the [Arm website](https://www.arm.com/architecture/system-architectures/systemready-certification-program/sr) along with links to purchase Arm servers. @@ -56,6 +56,6 @@ Servers typically run the Linux operating system. Popular distributions for serv If the GNU compiler is not pre-installed with your distribution, you can [install it using a package manager](/install-guides/gcc/native/). -You may also wish to install [Arm Compiler for Linux](/install-guides/acfl/) for HPC applications. +You can also install [Arm Compiler for Linux](/install-guides/acfl/) for HPC applications. [Migrating applications to Arm servers](/learning-paths/servers-and-cloud-computing/migration/) is a good place to start analyzing existing applications and reviewing guidance for developers interested in trying Arm hardware.