we are an epsilon away from writing our model in bf16 as well, in add… #101

	name: Build and test

	on:
	push:
	branches:
	- master
	pull_request:
	branches:
	- master

	jobs:
	build-and-test-cpu:
	strategy:
	matrix:
	os: [ubuntu-latest, macos-latest]

	runs-on: ${{ matrix.os }}

	steps:
	- name: Checkout code
	uses: actions/checkout@v3

	- name: Install OpenMP
	run: \|
	if [ "${{ runner.os }}" == "Linux" ]; then
	sudo apt-get update && sudo apt-get install -y libomp-dev
	elif [ "${{ runner.os }}" == "macOS" ]; then
	brew install libomp
	fi

	- name: Install dependencies
	run: pip install -r requirements.txt

	- name: Run preprocessing
	run: python prepro_tinyshakespeare.py

	- name: Train model
	run: python train_gpt2.py --device=cpu

	- name: Compile training and testing program
	run: make test_gpt2 train_gpt2

	- name: Execute testing program (With OpenMP)
	run: OMP_NUM_THREADS=8 ./test_gpt2

	- name: Compile training and testing program without OpenMP
	run: NO_OMP=1 make test_gpt2 train_gpt2

	- name: Execute testing program (No OpenMP)
	run: ./test_gpt2

	build-with-cuda-fp32:
	runs-on: ubuntu-latest # Host OS, Docker will run on top of this
	container:
	image: nvidia/cuda:11.2.2-devel-ubuntu20.04 # Example CUDA development image with nvcc

	steps:
	- name: Checkout code
	uses: actions/checkout@v3

	- name: Build project
	run: make train_gpt2fp32cu test_gpt2fp32cu

Provide feedback