ceshine

radekosmulski

Summary of Fastai.ai's in-depth discussion of types of normalization

# simulated batch of images
x = torch.rand(64, 3, 224, 224)
# or some number of layers up the convolutional stack
x = torch.rand(64, 256, 32, 32)

BatchNorm:

premek

# Put this function to your .bashrc file.
# Usage: mv oldfilename
# If you call mv without the second parameter it will prompt you to edit the filename on command line.
# Original mv is called when it's called with more than one argument.
# It's useful when you want to change just a few letters in a long name.
#
# Also see:
# - imv from renameutils
# - Ctrl-W Ctrl-Y Ctrl-Y (cut last word, paste, paste)

AFRUITPIE

#!/bin/bash

# This is a quick installer
# script I made to build and install the latest version of
# fish on my Raspberry Pi.
#
# Use at your own risk as I have made no effort to make
# this install safe!

set -e

tntwist

# Requirements
#1. Enable IPv6 for the docker daemon:
#$ cat /etc/docker/daemon.json
#{
#  "ipv6": true,
#  "fixed-cidr-v6": "2001:db8:1::/64"
#}
#2. reload docker :
#$ sudo systemctl reload docker
#3. start:

chrisalbon

# -*- coding: utf-8 -*-

""" Deletes all tweets below a certain retweet threshold.
"""

import tweepy
from datetime import datetime

# Constants
CONSUMER_KEY = ''

jph00

#!/usr/bin/env bash
###
# NB: You probably don't want this gist any more.
#     Instead, use this version from `fastsetup`:
#     https://github.com/fastai/fastsetup/blob/master/setup-conda.sh
###

set -e

cd

nikoheikkila

envsource

⚠️ NOTE (20.5.2023): I don't really use this function at all anymore since there are now tools such as Taskfile, which reads the .env file for me sparing me the need to pollute my session with environment variables.

---

I've been using Fish shell for years which is great and all, but one thing that has got me frustrated is using it with .env files.

When attempting to run source .env in a project, I usually encounter this problem:

ceshine

from fastcore.basics import patch_to


class Demo:
    val = 10
    def __init__(self, val):
        self.val = val
    
# ====================
# The default mode

jph00

# I added all the imports by hand
import torch
from torchvision.datasets import ImageFolder
from torchvision import transforms,models
from torch import nn,optim

# For all functions including this one, I wrote the name and docstring, and sometimes also the param names
def finetune(folder, model):
    """fine tune pytorch model using images from folder and report results on validation set"""
    if not os.path.exists(folder): raise ValueError(f"{folder} does not exist")