Lesson 1 - Intro and watch party

Timings: 18/3/2020 7pm-9pm IST

Zoom Webinar Link : https://zoom.us/j/112279140

Official Jupyter notebook: https://github.com/fastai/course-v4/blob/master/nbs/01_intro.ipynb

Starter Notebook : https://jovian.ml/aakashns/fastai2-lesson1
(simpler version prepared by Jovian.ml team)

Asking questions
During the livestream, if you have any questions, please post them on this thread (NOT on Zoom!!). We will compile and post these questions on the official FastAI forum, and post the answers from the FastAI team back here. The questions will also be discussed during the review session on Saturday.

Recordings
Please note that the session will NOT be recorded, so this is your only chance to watch it live, until the online MOOC is released officially in July.

22 Likes

Hello Everyone, like this message just so we can know how many people are in the live session and use this thread for all the discussions.

56 Likes

Hi I am Prasad K from Bangalore. I am an entrepreneur and resell data analytics software

1 Like

How’s the re-stream ? Are both audio and video are okay?

7 Likes

yes perfect audio and video

1 Like

IMPORTANT NOTE
In today’s lecture, Jeremy spent the fist 45 minutes talking about Covid-19 (not related to deep learning). That part of the video will be edited and put up online by the FastAI team in a day or two, so that’s why we are not streaming it.

We are starting the stream from the point where they start the course officially.

6 Likes

Hit like if you pledge to be a “decent human being” and not misuse the available jupyter version of the book. :angel:

8 Likes

What are the prerequisites for this MOOC? What background knowledge and experience do we need?

Only basic python. Rest is learn as you go.

Here is the link to the full textbook (in the form of Jupyter notebooks):

4 Likes

We’ll be taking a short 5 minutes break at 8:25pm.

We will soon re-post setup instructions from the main FastAI forum here, for platforms like AWS, Google Cloud, Paperspace etc.

In the meantime, you can use Kaggle Notebooks to get started. Try following the instructions in this starter notebook: https://jovian.ml/aakashns/fastai2-lesson1

1 Like

This is the link to video where Jeremy talks about data analysis done on COVID-19 https://youtu.be/GZ0yNMnvwqY . Watch it if you’re interested, this was skipped during the re-streaming.

2 Likes

Any idea how much difference is there between v3 and v4?

I don’t think I can learn effectively through these live videos given my current schedule. Since v4 videos won’t be available till July, I just want to understand if I can learn from v3 till then instead.

1 Like

I will post this question on the FastAI forum.

Here are my thoughts:
Since the library has been rewritten from scratch for v4 of the course, there are some changes to the top-level API and also how the packages are organized. Also, every year the course is updated with state-of-the-art models & techniques, so v3 is a bit outdated since it is almost 15 months old. If you can’t join live, consider learning via the textbook: https://github.com/fastai/fastbook

4 Likes

Setup instructions from the official forum have been posted on the Setup thread: Setup Instructions - FastAI

2 Likes

Before running notebooks install the dependency to avoid any errors while running them. with requirements.txt file !pip install -r requirements.txt which is inside the git repo. :+1:

1 Like

Download of https://s3.amazonaws.com/fast-ai-imageclas/oxford-iiit-pet.tgz has failed after 5 retries
Fix the download manually:

File downloaded is broken. Remove C:\Users\hipra.fastai\archive\oxford-iiit-pet.tgz and try again.

1 Like

Thanks everyone for joining the livestream, and huge thanks to @PrajwalPrashanth for hosting & @init27 for the tips! Let’s get to work and come up with some interesting project ideas.

Please post your questions here!

2 Likes