hwasiti
(Haider Alwasiti)
October 24, 2018, 6:05pm
797
@jeremy In your notebook output, resnet34 learn.fit_one_cycle(4) took around 2 minutes in the main video notebook (using sagemaker)
While I can see, it took for you around 1 minute in the github lesson 1 notebook.
Can you please mention what was the specs for the video lesson nb training and what was for the github nb?
This could be useful for all of us as a baseline to check whether our local setup is working well.
For me I am getting much lower speed (12 min) for gtx1080ti with an old cpu (Xeon® W3503) which is running 100%, and from time to time the gpu utilization pops up from 2% to 70-90%
The 1st thing I should do is to get rid of the gpu risers that is lowering the pcie2 x16 into pcie2 x1 which is the only way to install multiple gpu for me.
jeremy
(Jeremy Howard)
October 24, 2018, 6:09pm
798
For github repo I’m using a 1080ti. In the lesson it’s a V100
3 Likes
hwasiti
(Haider Alwasiti)
October 24, 2018, 6:16pm
800
In the lesson it is twice as slow. V100 should be at least 1.5x faster than gtx1080ti
Do you suspect that the cpu is the bottleneck? What are the cpu specs?
SSD vs hdd has no role in training I think, right?
1 Like
isarth
(Sarthak)
October 24, 2018, 6:22pm
801
Anyone facing this error while training --> https://www.evernote.com/l/AQMUf-g95YxE04J4bJ2NTNPhgMzBVdyDgWg
OS: Mac OS and installed using pip
Ralph
October 24, 2018, 6:24pm
802
You might want to look at this benchmark thread:
Let’s try to keep this thread on topic…
The purpose of this thread is to share your experience running the “lesson-1-pets.ipynb” jupyter notebook on various platforms. Many people are eager to get their local servers up and running or even building a brand new box with the latest hardware. Others may prefer to use a paid cloud option. Either way, it costs money upfront to build a local server, or you will pay as you go with a cloud option (once credits run out). Hopefully, in the end, people wi…
1 Like
Are you using Colab? Check this thread.
1 Like
kofi
(Kofi Asiedu Brempong)
October 24, 2018, 8:17pm
804
Hi, @jeremy @rachel
I applied lesson 1 to another fine grain classification problem
And i’m getting pretty high accuracy, but the images come organised in folders where the folder names are aliases of the original names of the species. The original names are specified in a separate text file.
Is there a way to replace the class names in the data bunch with the the real names?
1 Like
tsail
(Larry)
October 24, 2018, 8:20pm
805
Have you been able to find any documentation? I’m struggling to find the docs for “ImageDataBunch” as well.
EDIT: found it… http://docs.fast.ai/vision.data.html#class-imagedatabunch
1 Like
kofi
(Kofi Asiedu Brempong)
October 24, 2018, 8:27pm
807
Hi guys, I wrote an article about my experiences learning AI over the summer, please let me know your thoughts.
1 Like
jeremy
(Jeremy Howard)
October 24, 2018, 8:29pm
808
Kofi you might want to share here:
Show us what you’ve created with what you learned in fast.ai! It could be a blog post, a jupyter notebook, a picture, a github repo, a web app, or anything else. Some tips:
Probably the easiest way to blog is on Medium . If you use Medium, make sure you add your twitter username to your Medium profile, so that sharing will automatically credit you
The easiest way to share a notebook on github is to install the gist it extension . This will only be possible if you use a platform th…
1 Like
I noticed Jeremy sets up random seed in Lesson-1 will this take care of reproducibility for the weight initialization for CNN
Hi kofi. Please don’t tag Jeremy and Rachel unless other people in the forums cannot help you, see Etiquette for Posting to Forums .
Regarding your question, I think the from_func helper function would be useful. Let us know what you learn!
lion137
October 24, 2018, 9:59pm
814
General remark, isn’t it better to, in notebook, import fastai in that way:
import fastai as fs
Than I can type: fs.
, hit the tab and it give me suggestion?
As you prefer. If you import like suggested in the course (from fastai import *
and any application like from fastai.vision import *
) you can hit tab at any point in the notebook and get the same suggestions.
2 Likes
Michal_w
(Michal Wawrzyniuk)
October 24, 2018, 11:11pm
816
One of the way to use fast.ai is git clone fastai repository v1 and add symlink to folder with your notebook pointing to fastai folder
Rvbens
(Rubén Chaves)
October 25, 2018, 12:36am
818
I having this problem with fastai 1.0.13. It work fine with fastai 1.0.6
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-25-07536b5eb84c> in <module>
----> 1 path = untar_data(URLs.MNIST_SAMPLE); path
~/anaconda3/envs/fastai/lib/python3.6/site-packages/fastai/datasets.py in untar_data(url, fname, dest, data)
83 def untar_data(url:str, fname:PathOrStr=None, dest:PathOrStr=None, data=True):
84 "Download `url` if doesn't exist to `fname` and un-tgz to folder `dest`"
---> 85 dest = Path(ifnone(dest, _url2path(url)))
86 if not dest.exists():
87 fname = download_data(url, fname=fname)
~/anaconda3/envs/fastai/lib/python3.6/site-packages/fastai/datasets.py in _url2path(url, data)
57 def _expand_path(fpath): return Path(fpath).expanduser()
58 def _url2name(url): return url.split('/')[-1]
---> 59 def _url2path(url, data=True): return datapath4file(f'{_url2name(url)}') if data else modelpath4file(f'{_url2name(url)}')
60 def _url2tgz(url): return datapath4file(f'{_url2name(url)}.tgz')
61
~/anaconda3/envs/fastai/lib/python3.6/site-packages/fastai/datasets.py in datapath4file(filename)
70 local_path = URLs.LOCAL_PATH/'data'/filename
71 if local_path.exists() or local_path.with_suffix('.tgz').exists(): return local_path
---> 72 else: return _expand_path(Config.get_key('data_path'))/filename
73
74 def download_data(url:str, fname:PathOrStr=None):
~/anaconda3/envs/fastai/lib/python3.6/site-packages/fastai/datasets.py in get_key(cls, key)
36
37 @classmethod
---> 38 def get_key(cls, key): return cls.get().get(key, cls.DEFAULT_CONFIG.get(key,None))
39
40 @classmethod
AttributeError: 'NoneType' object has no attribute 'get'
jeremy
(Jeremy Howard)
October 25, 2018, 1:01am
819
You should use the developer install from the readme instead of this approach - all the same benefits, but more convenient.
4 Likes
Michal_w
(Michal Wawrzyniuk)
October 25, 2018, 1:17am
820
Saving learn instance ???
learn.save(‘stage-1’)
For
Total time: 01:13
epoch train loss valid loss error_rate
1 1.174627 0.330782 0.101796 (00:18)
2 0.498293 0.266701 0.084498 (00:18)
3 0.232486 0.211435 0.068530 (00:18)
4 0.307904 0.225669 0.077844 (00:18)
Which weights are saved ?
All?
Last epoch weights?
The best weights?
Thanks
Michal