Update docs for aml and nasbenchmark #2647
Conversation
|
Please revert the changes to Chinese docs. |
| **Step 2.** Run docker. | ||
|
|
||
| For NAS-Bench-101, | ||
| To avoid storage and legal issues, we do not provide any prepared databases. Please follow the following steps. |
There was a problem hiding this comment.
legal issues -> legality issues.
There was a problem hiding this comment.
will be fixed in new PR.
| @@ -7,40 +7,25 @@ | |||
| Example Usages <BenchmarksExample> | |||
| ``` | |||
|
|
|||
| ## Introduction | |||
| To imporve the reproducibility of NAS algorithms as well as reducing computing resource requirements, researchers proposed a series of NAS benchmarks such as [NAS-Bench-101](https://arxiv.org/abs/1902.09635), [NAS-Bench-201](https://arxiv.org/abs/2001.00326), [NDS](https://arxiv.org/abs/1905.13214), etc. NNI provides a query interface for users to acquire these benchmarks. Within just a few lines of codes, researcher are able to evaluate their NAS algorithms easily and fairly by utilizing these benchmarks. | |||
There was a problem hiding this comment.
it seems that "and" should not be in front of etc. Refer to https://www.sohu.com/a/149363473_814105.
| "metadata": {}, | ||
| "source": [ | ||
| "Use the following architecture as an example:<br>\n", | ||
| "" |
There was a problem hiding this comment.
will be fixed in new PR.
| "cell_type": "markdown", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "An architecture of NAS-Bench-101 could be trained more than once. Each element of the returned generator is a dict which contains one of the training results of this architecture including train/valid/test accuracy, training time, number of epochs, etc. The results of NAS-Bench-201 and NDS follow similar formats." |
There was a problem hiding this comment.
this architecture => this trial config (architecture + hyper-parameters)
There was a problem hiding this comment.
will be fixed in new PR.
| "metadata": {}, | ||
| "source": [ | ||
| "Use the following architecture as an example:<br>\n", | ||
| "" |
There was a problem hiding this comment.
will be fixed in new PR.
| @@ -5,12 +5,22 @@ NNI supports running an experiment on [AML](https://azure.microsoft.com/en-us/se | |||
| ## Setup environment | |||
| Step 1. Install NNI, follow the install guide [here](../Tutorial/QuickStart.md). | |||
|
|
|||
| Step 2. Create AML account, follow the document [here](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-manage-workspace-cli). | |||
| Step 2. Create an Azure account using this [link](https://azure.microsoft.com/en-us/free/services/machine-learning/). If you already have an Azure account, skip this step. | |||
There was a problem hiding this comment.
I think it's an AzureML subscription rather than an Azure account?
There was a problem hiding this comment.
This line will guide users to create an azure account with a subscription. Subscription seems to attached to user's account instead of AML. I will fix this in the new PR.
|  | ||
|
|
||
| Step 6. Create an AML cluster as the computeTarget. | ||
|  |
There was a problem hiding this comment.
Why is this photo half blank?
There was a problem hiding this comment.
Used wrong version. Will be fixed in new PR.
|
|
||
| ```bash | ||
| docker run -v ${HOME}/.nni/nasbenchmark:/outputs -v /path/to/your/nni:/nni python:3.7 /bin/bash /nni/examples/nas/benchmarks/nds.sh | ||
| git clone https://github.com/microsoft/nni.git |
There was a problem hiding this comment.
I think we should tell users to refer to Quickstart to clone NNI. They could clone the wrong version of NNI if they just follow this command.
There was a problem hiding this comment.
Sure. Will be fixed in new PR.
| Run the following commands to start the example experiment: | ||
| ``` | ||
| git clone https://github.com/microsoft/nni.git |
There was a problem hiding this comment.
Will be fixed in new PR.
|
This PR will be closed since it will be replaced by #2651 |
Update docs for aml and nasbenchmark