Welcome to Part 1 (v2)

Can I run all the code in the video(jupyter notebook) in a ubuntu 16.04 server with 2 k80 gpu, 12 Intel® Xeon® CPU E5-2620 and 64G ram, 128G disk? I think the script in the
first notebook is quite different from the code I see in tensorflow or keras environment, what software package do I need to install to run those notebook?

I am a couple lessons in on the old (v1) version of the course. For myself and those of us in a similar spot, would you recommend rewatching the v2 version of the videos?

Also, will the v2 version of the course be doable on a custom built pc as the v1 course was or will it be reliant on the online tools mentioned above (paperspace and crestle)?

Edit: Started watching the first v2 video and found that the recommended approach (paperspace) sets up a server from scratch so it should be no problem using a custom PC. Additionally, the relatively low investment made so far into the v1 of the course makes me feel like it would be worth at least watching the new videos, if not following any additional assignments as well.

Does anyone know where can I find the machine learning course videos?

Jeremy has mentioned them consistently, but I can’t find the videos. Thank you!

Jeremy was referring to this course.

1 Like

Great! thanks a lot!

Yeap. See this post: Another treat! Early access to Intro To Machine Learning videos

Is anyone from London up for a (potentially weekly) hangout/study group?

I’m looking to following along with the videos using my local machine with an NVIDIA gpu (gtx 1060):

Theoretically if I have Anaconda 3, and Python 3 installed, and I download all the course files from the website I’d be able to follow along with the videos?

OS: Windows 10
Processor: i7
GPU: GTX 1060

If anyone else has done this before and can offer some advice, that’d be extremely helpful! :slight_smile:

Theoretical that’s possible plus you need to download some other library’s also…
This will make it complete…

But being on windows itself is an invitation to problems…(personal experience)

Ok, let’s talk about what is required to work with part1_v2 on a personal local workstation.

I think a good starting point would be https://github.com/fastai/fastai/blob/master/environment.yml

I’d like to hear specific recommendations from Jeremy and/or Rachel. Thanks a lot!

1 Like

Hi Jeremy,

Are you going to be releasing an updated wiki to go along with v2 of Part I video/website? Any idea of the timeline for that?


Try this:

conda create -n fast_ai python=3.6 numpy cython statsmodels opencv

pip install fastai

conda install -c peterjc123 pytorch cuda90

Keep us posted. I never managed to get pytorch working on w10.


The only thing that failed to download was the bcolz wheel, which caused a command to fail.

bcolz wheel failure:
command failure:

Then I tried following along with the lesson notebook, a bunch of bugs kept popping up.

A few days ago I got the amazon instance to work from the very first version of the video. I guess I’ll just wait till I get paid next month and use crestle or the other option to learn data science. I spent money on this computer specifically so I could use it for learning machine learning, but oh well.

Thanks for your help balnazzar, those commands made trying this out a lot quicker. The other day it took me 3 hours to configure that amazon ec2 server for my windows machine. Did you end up using the tools suggested in the V2 introductory video, or did you get it running on your local machine?

I am running Pytorch (.3) successfully on Win 10

Were you able to follow along with the first lesson?

No problem, Richard. I’ll keep you posted, should I make any further progress.

I’m fairly convinced I should be able to configure a workstation for deep learning no matter what libraries I use, since that’s a valuable skill to learn by its own.

So you managed to get pytorch working properly on windows 10, which is substantial! :slight_smile: Do you have other libraries installed and/or other versions of cuda?

1 Like

Did I manage to run lesson1 nbs? Yes, but with cpu-only pytorch, wich leaves me dissatisfied.

I could switch to linux, but for a number of reasons, I’d rather follow with windows 10, primarily since I have other things to do on that windows box, and rebooting continuously is a nonoption.

Thanks for your feedback Aditya.

Do you have other versions of cuda installed and/or other DL libraries in other anaconda envs? Thanks!

Yep Cuda 8 and cudnn 6.0

Have Keras, Tensorflow, Pytorch in my environment

Had installed Pytorch from Anaconda Peterjc channel

Thanks Aditya.

Could you attach the output of “conda list” for your env? Furthermore, I suppose you are using Cuda 8.0 for both TF and Pytorch, am I right?