huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-09 19:34:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

revert build

Browse Source
This commit is contained in:
OlivierDehaene 2023-04-11 11:08:20 +02:00
parent a1a6b5cc20
commit c2beaa279e
2 changed files with 9 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
.github/workflows/build.yaml vendored
View File

@ -62,16 +62,23 @@ jobs:
id: meta id: meta
uses: docker/metadata-action@v4.3.0 uses: docker/metadata-action@v4.3.0
with: with:
flavor: |
latest=auto
images: | images: |
ghcr.io/huggingface/text-generation-inference
registry.internal.huggingface.tech/api-inference/community/text-generation-inference registry.internal.huggingface.tech/api-inference/community/text-generation-inference
db4c2190dd824d1f950f5d1555fbadf0.azurecr.io/text-generation-inference
tags: | tags: |
type=raw,value=llama-sha-${{ env.GITHUB_SHA_SHORT }} type=semver,pattern={{version}}
type=semver,pattern={{major}}.{{minor}}
type=raw,value=latest,enable=${{ github.ref == format('refs/heads/{0}', github.event.repository.default_branch) }}
type=raw,value=sha-${{ env.GITHUB_SHA_SHORT }}
- name: Build and push Docker image - name: Build and push Docker image
uses: docker/build-push-action@v2 uses: docker/build-push-action@v2
with: with:
context: . context: .
file: Dockerfile file: Dockerfile
push: true push: ${{ github.event_name != 'pull_request' }}
platforms: 'linux/amd64' platforms: 'linux/amd64'
tags: ${{ steps.meta.outputs.tags }} tags: ${{ steps.meta.outputs.tags }}
labels: ${{ steps.meta.outputs.labels }} labels: ${{ steps.meta.outputs.labels }}

2
server/text_generation_server/models/custom_modeling/flash_llama_modeling.py
View File

@ -233,8 +233,6 @@ class PositionRotaryEmbedding(RotaryEmbedding):
): ):
self._seq_len_cached = seqlen self._seq_len_cached = seqlen
t = torch.arange(seqlen, device=device, dtype=self.inv_freq.dtype) t = torch.arange(seqlen, device=device, dtype=self.inv_freq.dtype)
# Don't do einsum, it converts fp32 to fp16
# freqs = torch.einsum("i,j->ij", t, self.inv_freq)
freqs = torch.outer(t, self.inv_freq.to(device=t.device)) freqs = torch.outer(t, self.inv_freq.to(device=t.device))
self._cos_cached = torch.cos(freqs).to(dtype) self._cos_cached = torch.cos(freqs).to(dtype)
self._sin_cached = torch.sin(freqs).to(dtype) self._sin_cached = torch.sin(freqs).to(dtype)