Upgrade to Synapse 1.14

.github/workflows/fast_tests.yml

@@ -24,10 +24,10 @@ jobs:
   start-runner:
     name: Start self-hosted EC2 runner
     needs: authorize
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
     env:
       AWS_REGION: us-east-1
-      EC2_AMI_ID: ami-0a82d7d7ad5d25f56
+      EC2_AMI_ID: ami-0a2179742e502fdfe
       EC2_INSTANCE_TYPE: dl1.24xlarge
       EC2_SUBNET_ID: subnet-b7533b96
       EC2_SECURITY_GROUP: sg-08af7938042271373
@@ -77,7 +77,7 @@ jobs:
           ref: ${{ github.event.pull_request.merge_commit_sha }}
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -89,7 +89,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/fast_tests.sh
   diffusers:
     name: Run tests for optimum.habana.diffusers
@@ -113,7 +113,7 @@ jobs:
           ref: ${{ github.event.pull_request.merge_commit_sha }}
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -125,7 +125,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/fast_tests_diffusers.sh
   stop-runner:
     name: Stop self-hosted EC2 runner
@@ -134,7 +134,7 @@ jobs:
       - start-runner # required to get output from the start-runner job
       - transformers # required to wait for the tests to be finished
       - diffusers # required to wait for the tests to be finished
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
     env:
       AWS_REGION: us-east-1
     if: ${{ always() }} # required to stop the runner even if the error happened in the previous jobs

.github/workflows/slow_tests.yml

@@ -12,10 +12,10 @@ concurrency:
 jobs:
   start-runner:
     name: Start self-hosted EC2 runner
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
     env:
       AWS_REGION: us-west-2
-      EC2_AMI_ID: ami-01b277257cd28a061
+      EC2_AMI_ID: ami-0961e95b539f72c46
       EC2_INSTANCE_TYPE: dl1.24xlarge
       EC2_SUBNET_ID: subnet-452c913d
       EC2_SECURITY_GROUP: sg-0894f4f70dd6bd778
@@ -55,7 +55,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -67,7 +67,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/example_diff_tests.sh
   stable-diffusion:
     name: Test Stable Diffusion
@@ -83,7 +83,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -95,7 +95,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_diffusers.sh
   deepspeed:
     name: Test DeepSpeed models
@@ -112,7 +112,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -124,7 +124,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_deepspeed.sh
   multi-card:
     name: Test multi-card models
@@ -141,7 +141,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -153,7 +153,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_8x.sh
   single-card:
     name: Test single-card models
@@ -171,7 +171,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -183,7 +183,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_1x.sh
   albert-xxl-single-card:
     name: Test single-card ALBERT XXL
@@ -204,7 +204,7 @@ jobs:
       - name: Pull image
         if: github.event.schedule == '0 21 * * 6'
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run test
         if: github.event.schedule == '0 21 * * 6'
         run: |
@@ -217,7 +217,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/albert_xxl_1x.sh
       - name: Warning
         if: github.event.schedule != '0 21 * * 6'
@@ -240,7 +240,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -252,7 +252,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             make slow_tests_text_generation_example TOKEN=${{ secrets.TEXT_GENERATION_CI_HUB_TOKEN }}
   stop-runner:
     name: Stop self-hosted EC2 runner
@@ -264,7 +264,7 @@ jobs:
       - single-card
       - albert-xxl-single-card
       - text-generation
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
     env:
       AWS_REGION: us-west-2
     if: ${{ always() }} # required to stop the runner even if the error happened in the previous jobs

.github/workflows/slow_tests_gaudi2.yml

@@ -17,7 +17,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -30,7 +30,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_diffusers.sh
   deepspeed:
     name: Test DeepSpeed models
@@ -43,7 +43,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -56,7 +56,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_deepspeed.sh
   multi-card:
     name: Test multi-card models
@@ -69,7 +69,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -82,7 +82,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_8x.sh
   single-card:
     name: Test single-card models
@@ -96,7 +96,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -110,7 +110,7 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             /bin/bash tests/ci/slow_tests_1x.sh
   text-generation:
     name: Test text-generation example
@@ -125,7 +125,7 @@ jobs:
         uses: actions/checkout@v2
       - name: Pull image
         run: |
-            docker pull vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+            docker pull vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
       - name: Run tests
         run: |
             docker run \
@@ -138,5 +138,5 @@ jobs:
             --cap-add=sys_nice \
             --net=host \
             --ipc=host \
-            vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest \
+            vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest \
             make slow_tests_text_generation_example TOKEN=${{ secrets.TEXT_GENERATION_CI_HUB_TOKEN }}

Makefile

@@ -49,7 +49,7 @@ slow_tests_8x: test_installs
 
 # Run DeepSpeed non-regression tests
 slow_tests_deepspeed: test_installs
-	python -m pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.13.0
+	python -m pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.14.0
 	python -m pytest tests/test_examples.py -v -s -k "deepspeed"
 
 slow_tests_diffusers: test_installs
@@ -58,7 +58,7 @@ slow_tests_diffusers: test_installs
 
 # Run text-generation non-regression tests
 slow_tests_text_generation_example: test_installs
-	python -m pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.13.0
+	python -m pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.14.0
 	python -m pytest tests/test_text_generation_example.py tests/test_encoder_decoder_text_summarization.py -v -s --token $(TOKEN)
 
 # Check if examples are up to date with the Transformers library

README.md

@@ -42,7 +42,7 @@ The `--upgrade-strategy eager` option is needed to ensure `optimum-habana` is up
 
 > To use DeepSpeed on HPUs, you also need to run the following command:
 >```bash
->pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.13.0
+>pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.14.0
 >```
 
 Optimum Habana is a fast-moving project, and you may want to install it from source:
@@ -211,7 +211,7 @@ Please refer to Habana Gaudi's official [installation guide](https://docs.habana
 
 > Tests should be run in a Docker container based on Habana Docker images.
 >
-> The current version has been validated for SynapseAI 1.13.
+> The current version has been validated for SynapseAI 1.14.
 
 
 ## Development

docs/Dockerfile

@@ -1,4 +1,4 @@
-FROM vault.habana.ai/gaudi-docker/1.13.0/ubuntu20.04/habanalabs/pytorch-installer-2.1.0:latest
+FROM vault.habana.ai/gaudi-docker/1.14.0/ubuntu22.04/habanalabs/pytorch-installer-2.1.1:latest
 
 ARG commit_sha
 ARG clone_url
@@ -7,13 +7,7 @@ ARG clone_url
 RUN apt-get update && apt-get install -y \
     software-properties-common \
     npm
-
-# Need node to build doc HTML. Taken from https://stackoverflow.com/a/67491580
-RUN apt-get update && apt-get install -y \
-    software-properties-common \
-    npm
-RUN npm install npm@9.8.1 -g && \
-    npm install n -g && \
+RUN npm install n -g && \
     n latest
 
 RUN git clone $clone_url optimum-habana && cd optimum-habana && git checkout $commit_sha

docs/source/installation.mdx

@@ -23,6 +23,6 @@ python -m pip install --upgrade-strategy eager optimum[habana]
 To use DeepSpeed on HPUs, you also need to run the following command:
 
 ```bash
-python -m pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.13.0
+python -m pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.14.0
 ```
 

docs/source/usage_guides/deepspeed.mdx

@@ -31,7 +31,7 @@ You can find more information about DeepSpeed Gaudi integration [here](https://d
 To use DeepSpeed on Gaudi, you need to install Optimum Habana and [Habana's DeepSpeed fork](https://github.com/HabanaAI/DeepSpeed) with:
 ```bash
 pip install optimum[habana]
-pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.13.0
+pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.14.0
 ```
 
 
@@ -78,7 +78,7 @@ It is strongly advised to read [this section](https://huggingface.co/docs/transf
 
 </Tip>
 
-Other examples of configurations for HPUs are proposed [here](https://github.com/HabanaAI/Model-References/tree/1.13.0/PyTorch/nlp/DeepSpeedExamples/deepspeed-bert/scripts) by Habana.
+Other examples of configurations for HPUs are proposed [here](https://github.com/HabanaAI/Model-References/tree/1.14.0/PyTorch/nlp/DeepSpeedExamples/deepspeed-bert/scripts) by Habana.
 
 The [Transformers documentation](https://huggingface.co/docs/transformers/main_classes/deepspeed#configuration) explains how to write a configuration from scratch very well.
 A more complete description of all configuration possibilities is available [here](https://www.deepspeed.ai/docs/config-json/).

examples/audio-classification/README.md

@@ -99,7 +99,7 @@ On 8 HPUs, this script should run in ~12 minutes and yield an accuracy of **80.4
 
 > You need to install DeepSpeed with:
 > ```bash
-> pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.13.0
+> pip install git+https://github.com/HabanaAI/DeepSpeed.git@1.14.0
 > ```
 
 DeepSpeed can be used with almost the same command as for a multi-card run:

examples/gaudi_spawn.py

@@ -85,7 +85,7 @@ def main():
         if not is_deepspeed_available():
             raise ImportError(
                 "--use_deepspeed requires deepspeed: `pip install"
-                " git+https://github.com/HabanaAI/DeepSpeed.git@1.13.0`."
+                " git+https://github.com/HabanaAI/DeepSpeed.git@1.14.0`."
             )
 
     # Patch sys.argv