jprdonnelly/MachineLearningNotebooks
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Merge pull request #360 from Azure/paledger/update-readme

Browse Source 
Update readme/cluster location from PM's instructions
This commit is contained in:
Paula Ledgerwood
2019-05-06 10:08:22 -07:00
committed by GitHub
parent 56b0664b6b 5328186d68
commit dcce6f227f
4 changed files with 31 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
how-to-use-azureml/deployment/accelerated-models/README.md
View File

@@ -13,8 +13,21 @@ To learn more about the azureml-accel-model classes, see the section [Model Clas
### Step 1: Create an Azure ML workspace
Follow [these instructions](https://docs.microsoft.com/en-us/azure/machine-learning/service/quickstart-create-workspace-with-python) to install the Azure ML SDK on your local machine, create an Azure ML workspace, and set up your notebook environment, which is required for the next step.
### Step 2: Install the Azure ML Accelerated Models SDK
### Step 2: Check your FPGA quota
Use the Azure CLI to check whether you have quota.
```shell
az vm list-usage --location "eastus" -o table
```
The other locations are ``southeastasia``, ``westeurope``, and ``westus2``.
Under the "Name" column, look for "Standard PBS Family vCPUs" and ensure you have at least 6 vCPUs under "CurrentValue."
If you do not have quota, then submit a request form [here](https://aka.ms/accelerateAI).
### Step 3: Install the Azure ML Accelerated Models SDK
Once you have set up your environment, install the Azure ML Accel Models SDK. This package requires tensorflow >= 1.6,<2.0 to be installed.
If you already have tensorflow >= 1.6,<2.0 installed in your development environment, you can install the SDK package using:
@@ -35,7 +48,7 @@ If your machine supports GPU (for example, on an [Azure DSVM](https://docs.micro
pip install azureml-accel-models[gpu]
```
### Step 3: Follow our notebooks
### Step 4: Follow our notebooks
The notebooks in this repo walk through the following scenarios:
* [Quickstart](accelerated-models-quickstart.ipynb), deploy and inference a ResNet50 model trained on ImageNet

8
how-to-use-azureml/deployment/accelerated-models/accelerated-models-object-detection.ipynb
View File

@@ -273,11 +273,12 @@
"from azureml.core.compute import AksCompute, ComputeTarget\n",
"\n",
"# Uses the specific FPGA enabled VM (sku: Standard_PB6s)\n",
"# Authentication is enabled by default, but for testing we specify False\n",
"# Standard_PB6s are available in: eastus, westus2, westeurope, southeastasia\n",
"prov_config = AksCompute.provisioning_configuration(vm_size = \"Standard_PB6s\",\n",
" agent_count = 1)\n",
" agent_count = 1, \n",
" location = \"eastus\")\n",
"\n",
"aks_name = 'my-aks-pb6-ssd-vgg'\n",
"aks_name = 'aks-pb6-obj'\n",
"# Create the cluster\n",
"aks_target = ComputeTarget.create(workspace = ws, \n",
" name = aks_name, \n",
@@ -318,6 +319,7 @@
"from azureml.core.webservice import Webservice, AksWebservice\n",
"\n",
"# Set the web service configuration (for creating a test service, we don't want autoscale enabled)\n",
"# Authentication is enabled by default, but for testing we specify False\n",
"aks_config = AksWebservice.deploy_configuration(autoscale_enabled=False,\n",
" num_replicas=1,\n",
" auth_enabled = False)\n",

6
how-to-use-azureml/deployment/accelerated-models/accelerated-models-quickstart.ipynb
View File

@@ -341,9 +341,10 @@
"from azureml.core.compute import AksCompute, ComputeTarget\n",
"\n",
"# Uses the specific FPGA enabled VM (sku: Standard_PB6s)\n",
"# Authentication is enabled by default, but for testing we specify False\n",
"# Standard_PB6s are available in: eastus, westus2, westeurope, southeastasia\n",
"prov_config = AksCompute.provisioning_configuration(vm_size = \"Standard_PB6s\",\n",
" agent_count = 1)\n",
" agent_count = 1, \n",
" location = \"eastus\")\n",
"\n",
"aks_name = 'my-aks-pb6'\n",
"# Create the cluster\n",
@@ -386,6 +387,7 @@
"from azureml.core.webservice import Webservice, AksWebservice\n",
"\n",
"#Set the web service configuration (for creating a test service, we don't want autoscale enabled)\n",
"# Authentication is enabled by default, but for testing we specify False\n",
"aks_config = AksWebservice.deploy_configuration(autoscale_enabled=False,\n",
" num_replicas=1,\n",
" auth_enabled = False)\n",

10
how-to-use-azureml/deployment/accelerated-models/accelerated-models-training.ipynb
View File

@@ -47,7 +47,7 @@
" * [Transfer Learning](#transfer-learning)\n",
" * [Transfer Learning with Custom Weights](#custom-weights)\n",
"* [Create Image](#create-image)\n",
"* [Deploy Model](#deploy-model)\n",
"* [Deploy Image](#deploy-image)\n",
"* [Test the service](#test-service)\n",
"* [Clean-up](#cleanup)\n",
"* [Appendix](#appendix)"
@@ -630,11 +630,12 @@
"from azureml.core.compute import AksCompute, ComputeTarget\n",
"\n",
"# Uses the specific FPGA enabled VM (sku: Standard_PB6s)\n",
"# Authentication is enabled by default, but for testing we specify False\n",
"# Standard_PB6s are available in: eastus, westus2, westeurope, southeastasia\n",
"prov_config = AksCompute.provisioning_configuration(vm_size = \"Standard_PB6s\",\n",
" agent_count = 1)\n",
" agent_count = 1,\n",
" location = \"eastus\")\n",
"\n",
"aks_name = 'my-aks-pb6-training'\n",
"aks_name = 'aks-pb6-tl'\n",
"# Create the cluster\n",
"aks_target = ComputeTarget.create(workspace = ws, \n",
" name = aks_name, \n",
@@ -675,6 +676,7 @@
"from azureml.core.webservice import Webservice, AksWebservice\n",
"\n",
"# Set the web service configuration (for creating a test service, we don't want autoscale enabled)\n",
"# Authentication is enabled by default, but for testing we specify False\n",
"aks_config = AksWebservice.deploy_configuration(autoscale_enabled=False,\n",
" num_replicas=1,\n",
" auth_enabled = False)\n",