butchland/fastai_xla_extensions issues and pull requests

#49 - ImportError: cannot import name 'inf' from 'torch'

Issue - State: open - Opened by pankaj-kvhld over 1 year ago

#48 - NameError: name 'xmp' is not defined

Issue - State: open - Opened by pankaj-kvhld over 1 year ago

#47 - Bump nokogiri from 1.13.6 to 1.14.3 in /docs

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#46 - Bump activesupport from 6.0.3.2 to 6.0.6.1 in /docs

Pull Request - State: open - Opened by dependabot[bot] almost 2 years ago
Labels: dependencies

#45 - Bump nokogiri from 1.13.6 to 1.13.9 in /docs

Pull Request - State: closed - Opened by dependabot[bot] about 2 years ago - 1 comment
Labels: dependencies

#44 - Please be compatible with the latest version of fastai, thank you!

Issue - State: open - Opened by alleniver over 2 years ago

#43 - Bump tzinfo from 1.2.7 to 1.2.10 in /docs

Pull Request - State: closed - Opened by dependabot[bot] over 2 years ago
Labels: dependencies

#42 - Bump nokogiri from 1.13.4 to 1.13.6 in /docs

Pull Request - State: closed - Opened by dependabot[bot] over 2 years ago
Labels: dependencies

#41 - Bump nokogiri from 1.12.5 to 1.13.4 in /docs

Pull Request - State: closed - Opened by dependabot[bot] over 2 years ago
Labels: dependencies

#40 - Bump nokogiri from 1.12.5 to 1.13.3 in /docs

Pull Request - State: closed - Opened by dependabot[bot] almost 3 years ago - 1 comment
Labels: dependencies

#39 - Bump nokogiri from 1.11.5 to 1.12.5 in /docs

Pull Request - State: closed - Opened by dependabot[bot] about 3 years ago
Labels: dependencies

#38 - Bump addressable from 2.7.0 to 2.8.0 in /docs

Pull Request - State: closed - Opened by dependabot[bot] over 3 years ago
Labels: dependencies

#37 - Is this repo still working? Can't run the colab notebooks provided in the doc

Issue - State: closed - Opened by coldfir3 over 3 years ago - 1 comment

#36 - Bump nokogiri from 1.10.10 to 1.11.5 in /docs

Pull Request - State: closed - Opened by dependabot[bot] over 3 years ago
Labels: dependencies

#35 - xla_fit_one_cycle fails with an error when using fastai version 2.3.1

Issue - State: open - Opened by butchland over 3 years ago

#34 - Bump rexml from 3.2.4 to 3.2.5 in /docs

Pull Request - State: closed - Opened by dependabot[bot] over 3 years ago
Labels: dependencies

#33 - xla_lr_find locks xm.xla_device in main process

Issue - State: closed - Opened by butchland over 3 years ago
Labels: bug

#28 - creating a model with concat_pool = True creates a model that can't run on multiple tpu cores

Issue - State: open - Opened by butchland almost 4 years ago
Labels: bug

#27 - using fastai data loaders is slower on multi core is slow if using large image sizes

Issue - State: open - Opened by butchland almost 4 years ago

#26 - add `utils.print_aten_ops`

Pull Request - State: closed - Opened by tyoc213 almost 4 years ago - 1 comment

#25 - fastai_xla_extensions on kaggle errors out

Issue - State: closed - Opened by butchland almost 4 years ago - 7 comments

#22 - export and load_learner (for loading for inference) is not working

Issue - State: closed - Opened by butchland almost 4 years ago - 1 comment

#18 - Fixing some issues been found testing locally XLA

Pull Request - State: closed - Opened by tyoc213 almost 4 years ago - 1 comment

#17 - Proxy to get attr

Pull Request - State: closed - Opened by tyoc213 about 4 years ago - 2 comments

#16 - Run on multiple TPUs

Pull Request - State: closed - Opened by tyoc213 about 4 years ago - 1 comment

#15 - default_device no longer sets to overridden default_device method

Issue - State: closed - Opened by butchland over 4 years ago - 1 comment

#14 - Setting pretrained = true causes the cnn_learner model to not train

Issue - State: closed - Opened by butchland over 4 years ago - 2 comments

#13 - Let user aquire device

Pull Request - State: closed - Opened by tyoc213 over 4 years ago - 1 comment

#12 - replace fastai2 with fastai on all that is not explore_nbs

Pull Request - State: closed - Opened by tyoc213 over 4 years ago - 1 comment

#11 - run batch transforms on CPU if dataloader is using a TPU

Issue - State: closed - Opened by butchland over 4 years ago - 1 comment
Labels: enhancement

#10 - Tracing

Pull Request - State: closed - Opened by tyoc213 over 4 years ago

#9 - Xla run debug

Pull Request - State: closed - Opened by tyoc213 over 4 years ago

#8 - Sending RNN test

Pull Request - State: closed - Opened by tyoc213 over 4 years ago - 2 comments

#7 - Add patch to LMLanguageLearner to fix problem on load_ignore_keys where state_dict error due to use of TPU

Issue - State: open - Opened by butchland over 4 years ago

#6 - Language model trainer is very slow or not finishing fit_one_cycle for 1 epoch

Issue - State: open - Opened by butchland over 4 years ago - 2 comments

#5 - batch transforms for vision are slow

Issue - State: open - Opened by butchland over 4 years ago - 4 comments
Labels: no stale

#4 - learner.lr_find returns wrong result

Issue - State: closed - Opened by butchland over 4 years ago - 2 comments

#3 - It seems that using the value of the line in lr_find gives good results on trainning?

Pull Request - State: closed - Opened by tyoc213 over 4 years ago

#2 - TODO

Issue - State: closed - Opened by tyoc213 over 4 years ago

#1 - Trying to pinpoint why we can't train a model

Pull Request - State: closed - Opened by tyoc213 over 4 years ago - 3 comments

GitHub / butchland/fastai_xla_extensions issues and pull requests