Ryan Reece rreece

MacOs quick setup 🚀

Getting everything ready

1 - Xcode/Ruby/Command line tools

You need to have Xcode installed to proceed.

xcode-select --install
sudo xcodebuild -license accept

2 - Brew

Lifetime of python supported versions

Scientific Python: SPEC 0
Numpy: NEP 29

COO 2017 dataset

Ryan downloaded the COCO data from here: https://cocodataset.org/#download

wget http://images.cocodataset.org/zips/val2017.zip
wget http://images.cocodataset.org/zips/test2017.zip
wget http://images.cocodataset.org/annotations/annotations_trainval2017.zip

ImageNet-1k dataset

Download instructions: https://towardsdatascience.com/downloading-and-using-the-imagenet-dataset-with-pytorch-f0908437c4be

Create a kaggle account
Login and create an authentication token
Add the token to a file: ~/.kaggle/kaggle.json

The file has one line like this:

	"""
	openai-chat-client.py

	OpenAI compatible chat client in the terminal

	See:
	https://github.com/openai/openai-python
	https://github.com/vllm-project/vllm/blob/main/examples/openai_chat_completion_client.py
	"""

	import yaml

	def read_yaml(path):
	with open(path, 'r') as f:
	config = yaml.safe_load(f)
	return config

	import subprocess

	def run_shell(cmd):
	out = subprocess.check_output(cmd, stderr=subprocess.STDOUT, shell=True, encoding="UTF-8")
	return out

	"""
	Writes float16 data to a tfrecord as raw bytes and reads it back.

	Based on:
	https://stackoverflow.com/questions/40184812/tensorflow-is-it-possible-to-store-tf-record-sequence-examples-as-float16
	"""

	import argparse
	import numpy as np
	import tensorflow as tf

	def path_is_in_git(repo, path):
	"""
	Check if path is tracked by git.
	"""
	returncode = None
	try:
	cmd = 'git ls-files --error-unmatch %s' % (path)
	_ = subprocess.check_output(cmd,
	cwd=repo,
	shell=True,

	import os
	path_of_this_file = os.path.abspath(__file__)
	dir_of_this_file = os.path.dirname(os.path.abspath(__file__))