optimum-graphcore

optimum-graphcore copied to clipboard

Remove Run on Gradient links

1

Graphcore's contract with Paperspace has expired. This PR removes the "Run on Gradient" links in the examples.

JamesRandom

Bump scikit-learn from 0.24.2 to 1.0.1 in /examples/image-classification

1

Bumps [scikit-learn](https://github.com/scikit-learn/scikit-learn) from 0.24.2 to 1.0.1. Release notes Sourced from scikit-learn's releases. scikit-learn 1.0.1 We're happy to announce the 1.0.1 release with several bugfixes: You can see the changelog here:...

dependabot[bot]

Remove workflow deleting doc

1

docs now deleted automatically after 30 days https://github.com/huggingface/doc-builder/blob/main/.github/workflows/delete_old_pr_documentations.yml As done in optimum : https://github.com/huggingface/optimum/pull/1565 cc @regisss

echarlaix

Bump CI SDK

1

# What does this PR do? Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks...

katalinic-gc

Add `run_speech_recognition_seq2seq.py`

5

# What does this PR do? Adds `run_speech_recognition_seq2seq.py` for training/fine-tuning Seq2Seq speech recognition models, such as Whisper, on the IPU. ## Before submitting - [ ] This PR fixes a...

callumm-graphcore

BART: disable returning kv states since there exists an on device cache

1

# What does this PR do? Given that the kv cache is on device, there is no need to return `past_key_values`. However, this would require overriding the forward methods of...

kundaMwiza

What would it require to support LLaMA-based models as well?

StrangeTcy

Support simpler syntax for specifying pipeline splits

2

# What does this PR do? This PR is a WIP ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the...

payoto

Uses torch.fx to parallelize and transform pipelined models

1

# What does this PR do? - Provides a custom tracer, built upon `transformers.utils.fx.HFTracer` allowing the trace and transform the models we support in `optimum-graphcore` - A set of pipelining...

michaelbenayoun

Set LayerNorm's eps to a number that is larger than 6e-5

2

Currently, many LayerNorm's eps are smaller than 6.1e-5 (smallest fp16 value), which might cause underflow.

gejinchen