From 763b507c3052b1214856d5a30cb6bae7de180a2b Mon Sep 17 00:00:00 2001
From: Jingya HUANG <44135271+JingyaHuang@users.noreply.github.com>
Date: Mon, 2 Sep 2024 14:42:28 +0200
Subject: [PATCH] Add workflow for caching stable diffusion models (#685)

* workflow for sd models

* rename and schedule weekly

* rename the workflows
---
 ...ence_cache.yml => inference_cache_llm.yml} |  2 +-
 .../inference_cache_stable_diffusion.yml      | 52 +++++++++++++++++++
 2 files changed, 53 insertions(+), 1 deletion(-)
 rename .github/workflows/{inference_cache.yml => inference_cache_llm.yml} (97%)
 create mode 100644 .github/workflows/inference_cache_stable_diffusion.yml

diff --git a/.github/workflows/inference_cache.yml b/.github/workflows/inference_cache_llm.yml
similarity index 97%
rename from .github/workflows/inference_cache.yml
rename to .github/workflows/inference_cache_llm.yml
index 51d65909f..9ead54d9a 100644
--- a/.github/workflows/inference_cache.yml
+++ b/.github/workflows/inference_cache_llm.yml
@@ -1,4 +1,4 @@
-name: Optimum neuron inference cache builder
+name: Optimum neuron LLM inference cache builder
 
 on:
   workflow_dispatch:
diff --git a/.github/workflows/inference_cache_stable_diffusion.yml b/.github/workflows/inference_cache_stable_diffusion.yml
new file mode 100644
index 000000000..6dd811d47
--- /dev/null
+++ b/.github/workflows/inference_cache_stable_diffusion.yml
@@ -0,0 +1,52 @@
+name: Optimum neuron SD inference cache builder
+
+on:
+  workflow_dispatch:
+  schedule:
+    # Schedule the workflow to run every Saturday at midnight UTC
+    - cron: '0 0 * * 6'
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
+
+jobs:
+  cache:
+    name: Create optimum-neuron inference cache
+    runs-on:
+      group: aws-inf2-8xlarge
+    env:
+      AWS_REGION: us-east-1
+    strategy:
+      fail-fast: false
+      matrix:
+        config: [stable-diffusion]
+    steps:
+      - name: Install Neuron runtime
+        run: |
+          . /etc/os-release
+          sudo tee /etc/apt/sources.list.d/neuron.list > /dev/null <<EOF
+          deb https://apt.repos.neuron.amazonaws.com ${VERSION_CODENAME} main
+          EOF
+          wget -qO - https://apt.repos.neuron.amazonaws.com/GPG-PUB-KEY-AMAZON-AWS-NEURON.PUB | sudo apt-key add -
+          sudo apt-get update -y
+          sudo apt-get install aws-neuronx-tools=2.18.3.0 aws-neuronx-runtime-lib=2.21.41.0-fb1705f5f aws-neuronx-collectives=2.21.46.0-69b77134b  -y
+          export PATH=/opt/aws/neuron/bin:$PATH
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Install python and create venv
+        run: |
+          sudo apt install python3-venv python3-dev -y
+          python3 -m venv aws_neuron_venv_pytorch
+          source aws_neuron_venv_pytorch/bin/activate
+          python -m pip install -U pip
+          python -m pip config set global.extra-index-url https://pip.repos.neuron.amazonaws.com
+      - name: Install optimum neuron
+        run: |
+          source aws_neuron_venv_pytorch/bin/activate
+          python -m pip install .[neuronx]
+      - name: Create cache for ${{matrix.config}} models
+        run: |
+          source aws_neuron_venv_pytorch/bin/activate
+          config_prefix_url=https://huggingface.co/aws-neuron/optimum-neuron-cache/raw/main/inference-cache-config
+          HF_TOKEN=${{secrets.HF_TOKEN_OPTIMUM_NEURON_CACHE}} \
+            python tools/auto_fill_inference_cache.py --config_file ${config_prefix_url}/${{matrix.config}}.json