butchland fastai_xla_extensions issues

butchland / fastai_xla_extensions

A Python package to allow fastai to run on TPUs using Pytorch-XLA

https://butchland.github.io/fastai_xla_extensions

Apache License 2.0

36 stars 7 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

ImportError: cannot import name 'inf' from 'torch'

#49 pankaj-kvhld opened 1 year ago
0
NameError: name 'xmp' is not defined

#48 pankaj-kvhld opened 1 year ago
0
Bump nokogiri from 1.13.6 to 1.14.3 in /docs

#47 dependabot[bot] opened 1 year ago
0
Bump activesupport from 6.0.3.2 to 6.0.6.1 in /docs

#46 dependabot[bot] opened 1 year ago
0
Bump nokogiri from 1.13.6 to 1.13.9 in /docs

#45 dependabot[bot] closed 1 year ago
1
Please be compatible with the latest version of fastai, thank you!

#44 alleniver opened 2 years ago
0
Bump tzinfo from 1.2.7 to 1.2.10 in /docs

#43 dependabot[bot] closed 2 years ago
0
Bump nokogiri from 1.13.4 to 1.13.6 in /docs

#42 dependabot[bot] closed 2 years ago
0
Bump nokogiri from 1.12.5 to 1.13.4 in /docs

#41 dependabot[bot] closed 2 years ago
0
Bump nokogiri from 1.12.5 to 1.13.3 in /docs

#40 dependabot[bot] closed 2 years ago
1
Bump nokogiri from 1.11.5 to 1.12.5 in /docs

#39 dependabot[bot] closed 2 years ago
0
Bump addressable from 2.7.0 to 2.8.0 in /docs

#38 dependabot[bot] closed 3 years ago
0
Is this repo still working? Can't run the colab notebooks provided in the doc

#37 coldfir3 closed 3 years ago
1
Bump nokogiri from 1.10.10 to 1.11.5 in /docs

#36 dependabot[bot] closed 3 years ago
0
xla_fit_one_cycle fails with an error when using fastai version 2.3.1

#35 butchland opened 3 years ago
0
Bump rexml from 3.2.4 to 3.2.5 in /docs

#34 dependabot[bot] closed 3 years ago
0
xla_lr_find locks xm.xla_device in main process

#33 butchland closed 3 years ago
0
creating a model with concat_pool = True creates a model that can't run on multiple tpu cores

#28 butchland opened 3 years ago
0
using fastai data loaders is slower on multi core is slow if using large image sizes

#27 butchland opened 3 years ago
0
add `utils.print_aten_ops`

#26 tyoc213 closed 3 years ago
1
fastai_xla_extensions on kaggle errors out

#25 butchland closed 3 years ago
7
export and load_learner (for loading for inference) is not working

#22 butchland closed 3 years ago
1
Fixing some issues been found testing locally XLA

#18 tyoc213 closed 3 years ago
1
Proxy to get attr

#17 tyoc213 closed 3 years ago
2
Run on multiple TPUs

#16 tyoc213 closed 4 years ago
1
default_device no longer sets to overridden default_device method

#15 butchland closed 3 years ago
1
Setting pretrained = true causes the cnn_learner model to not train

#14 butchland closed 3 years ago
2
Let user aquire device

#13 tyoc213 closed 4 years ago
1
replace fastai2 with fastai on all that is not explore_nbs

#12 tyoc213 closed 4 years ago
1
run batch transforms on CPU if dataloader is using a TPU

#11 butchland closed 3 years ago
1
Tracing

#10 tyoc213 closed 4 years ago
0
Xla run debug

#9 tyoc213 closed 4 years ago
0
Sending RNN test

#8 tyoc213 closed 4 years ago
2
Add patch to LMLanguageLearner to fix problem on load_ignore_keys where state_dict error due to use of TPU

#7 butchland opened 4 years ago
0
Language model trainer is very slow or not finishing fit_one_cycle for 1 epoch

#6 butchland opened 4 years ago
2
batch transforms for vision are slow

#5 butchland opened 4 years ago
4
learner.lr_find returns wrong result

#4 butchland closed 4 years ago
2
It seems that using the value of the line in lr_find gives good results on trainning?

#3 tyoc213 closed 4 years ago
0
TODO

#2 tyoc213 closed 4 years ago
0
Trying to pinpoint why we can't train a model

#1 tyoc213 closed 4 years ago
3