forked from graphcore/tutorials
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrelease-commit-log-3.0.txt
73 lines (73 loc) · 3.77 KB
/
release-commit-log-3.0.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Your branch is up to date with 'origin/sdk-release-2.6'.
Already up to date.
Your branch is up to date with 'origin/sdk-release-3.0'.
Already up to date.
=================================================================
d0e286f7 Update links to point at SDK 3.0.0
197eed5e Running set tool for python notebooks
3f293dbb Running sst for release 3.0 tensorflow1
3822df3d Running sst tool for tensorflow2
34a7937a Suggesting users to raise GitHub issues rather than ZenDesk tickets
13274250 Removes `torch.no_grad` from tutorials. Closes T69914
7c5ad64d Regenerating set for release 3.0
382503bc Convert popvision README.rst to markdown
4173f84c Small documentation fixes to the VSCodeSetup.md for popxl tutorials
b964e9b3 Add PopTorch and PopART to trademark linter
7e9576b4 Update Jupyter guide for IPUs
71881c52 Point PopXL Addons to the newer release
a2011d02 Convert Poplar Tutorial `rst`->`md` using `pandoc`
5d905623 update gcl library name in makefile
0b2e2920 Pegged every link I think we can [CP 2.6]
8b57b2e5 Add VS Code guide
67a196b3 Prevent link parser from detecting C++ lambdas
43a7be9e pytorch vit model: Changing image input type to half
6c971160 Version the URLs to examples [CP 2.6]
b70b18d3 Updating headings and copyright
fddda241 Remove implicit dependency of PyTorch test on PopART custom op
4b4daa93 Remove deprecated ipus_per_replica parameter from popdist.tensorflow.set_ipu_config()
906c6409 Remove horovod from TF2 PopDist
b587fe11 Revert: Remove horovod from TF2 PopDist
805c7b18 Remove horovod from TF2 PopDist
ac6c19b7 Allow lines beginning with Copyright to pass the lint test
5d709f15 Change in PrintTensor format
e8aad174 Improving headers in pipelining tutorial
befe2600 Ignore MacOS filesystem special files
083bd900 Improving headings in READMEs
e1a70381 Improving popxl headings
0f5e8384 Add pointer to packedBERT training code in tutorials packedBERT README
efb2f548 Improving headers in tensorflow2 tensorboard
34a418f6 Removing poptorch specific fp16 behaviour as new release acts the same native pytorch
e5e4e6f5 Improving headers in tensorflow1/pipelining
a130962d Implemented test_urls internal link test
10678764 Changing variable names
9bd4c4ba Upgraded TF1 tests + deleted SubProcessChecker class
7a8b73ea I think numpy 1.19.0 is a high enough version to silence the github warnings.
27e9fa36 Upgraded pytorch tests
4b937add Removed use of SubProcessChecker `kernel_benchmarks/tensorflow`
2ddad36d [HIGH] Fix the failing test test_simple_sharding.py. Tacked on a few other bits of tidying
8cc2ae57 Upgraded TF2 tests
75c100a8 Added popxl link to README
cb71059e Upgraded all poplar tests
3e391bc1 POPXL tutorial 3: update images
ae727b08 Updated `.format(` to f-strings and brought 2 files closer together.
6f50c763 Test python files don't contain "SubProcessChecker" or "unittest"
58f094ae Test python files don't contain .format(
fbfa649b T66682: Make pytorch/basics tutorial deterministic + test the outputs
aed6504b Adding -q flag to pip install of requirements.txt
ad3bdb9f Add User Guide For Reading PVTI Files
a674ae38 Add internal link testing as a linter
95288f03 Mention RTS in PopXL tuto 5 name
2db7c058 Pin click==8.0.4 in black
72abbfd4 Adding vit model to README list
ebaa40e5 Support `md` in copyright linter (+small fixes)
e8b0fd1a Roll back black to support blacken-docs
79bdf535 Changing black docs version to v1.12.0
60634328 Run all linters
b8ec6580 Small tweaks to pre-commit config
1740a2b2 Move linting to pre-commit
2440eedd Update IPU Estimator feature example README to signpost IPULoggingTensorHook
a048967d Fully parallelised the URL test
829cd45d Upgrade HOROVOD to recomended version
ef8afe4e [HIGH] Upgrading all files to f-string style as a starting point
e38605ae Missed one instance of 'PopXL-addons'
e9856c58 Upgraded test - pytorch/finetuning_bert [Don't Cherry Pick for 2.6]