{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "Copyright (c) Microsoft Corporation. All rights reserved.\n", "\n", "Licensed under the MIT License." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "![Impressions](https://PixelServer20190423114238.azurewebsites.net/api/impressions/MachineLearningNotebooks/how-to-use-azureml/automated-machine-learning/forecasting-hierarchical-timeseries/auto-ml-forecasting-hierarchical-timeseries.png)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Many Models - Automated ML\n", "**_Generate many models time series forecasts with Automated Machine Learning_**\n", "\n", "---" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For this notebook we are using a synthetic dataset portraying sales data to predict the quantity of a vartiety of product SKUs across several states, stores, and product categories.\n", "\n", "**NOTE: There are limits on how many runs we can do in parallel per workspace, and we currently recommend to set the parallelism to maximum of 320 runs per experiment per workspace. If users want to have more parallelism and increase this limit they might encounter Too Many Requests errors (HTTP 429).**" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Prerequisites\n", "You'll need to create a compute Instance by following the instructions in the [EnvironmentSetup.md](../Setup_Resources/EnvironmentSetup.md)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 1.0 Set up workspace, datastore, experiment" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "gather": { "logged": 1613003526897 } }, "outputs": [], "source": [ "import azureml.core\n", "from azureml.core import Workspace, Datastore\n", "import pandas as pd\n", "\n", "# Set up your workspace\n", "ws = Workspace.from_config()\n", "ws.get_details()\n", "\n", "# Set up your datastores\n", "dstore = ws.get_default_datastore()\n", "\n", "output = {}\n", "output[\"SDK version\"] = azureml.core.VERSION\n", "output[\"Subscription ID\"] = ws.subscription_id\n", "output[\"Workspace\"] = ws.name\n", "output[\"Resource Group\"] = ws.resource_group\n", "output[\"Location\"] = ws.location\n", "output[\"Default datastore name\"] = dstore.name\n", "pd.set_option(\"display.max_colwidth\", -1)\n", "outputDf = pd.DataFrame(data=output, index=[\"\"])\n", "outputDf.T" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Choose an experiment" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "gather": { "logged": 1613003540729 } }, "outputs": [], "source": [ "from azureml.core import Experiment\n", "\n", "experiment = Experiment(ws, \"automl-many-models\")\n", "\n", "print(\"Experiment name: \" + experiment.name)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 2.0 Data\n", "\n", "This notebook uses simulated orange juice sales data to walk you through the process of training many models on Azure Machine Learning using Automated ML. \n", "\n", "The time series data used in this example was simulated based on the University of Chicago's Dominick's Finer Foods dataset which featured two years of sales of 3 different orange juice brands for individual stores. The full simulated dataset includes 3,991 stores with 3 orange juice brands each thus allowing 11,973 models to be trained to showcase the power of the many models pattern.\n", "\n", " \n", "In this notebook, two datasets will be created: one with all 11,973 files and one with only 10 files that can be used to quickly test and debug. For each dataset, you'll be walked through the process of:\n", "\n", "1. Registering the blob container as a Datastore to the Workspace\n", "2. Registering a tabular dataset to the Workspace" ] }, { "cell_type": "markdown", "metadata": { "nteract": { "transient": { "deleting": false } } }, "source": [ "### 2.1 Data Preparation\n", "The OJ data is available in the public blob container. The data is split to be used for training and for inferencing. For the current dataset, the data was split on time column ('WeekStarting') before and after '1992-5-28' .\n", "\n", "The container has\n", "
    \n", "
  1. 'oj-data-tabular' and 'oj-inference-tabular' folders that contains training and inference data respectively for the 11,973 models.
    2. \n", "
  2. It also has 'oj-data-small-tabular' and 'oj-inference-small-tabular' folders that has training and inference data for 10 models.
    3. \n", "
\n", "\n", "To create the [TabularDataset](https://docs.microsoft.com/en-us/python/api/azureml-core/azureml.data.tabular_dataset.tabulardataset?view=azure-ml-py) needed for the ParallelRunStep, you first need to register the blob container to the workspace." ] }, { "cell_type": "markdown", "metadata": { "nteract": { "transient": { "deleting": false } } }, "source": [ " To use your own data, put your own data in a blobstore folder. As shown it can be one file or multiple files. We can then register datastore using that blob as shown below.\n", " \n", "

How sample data in blob store looks like

\n", "\n", "['oj-data-tabular'](https://ms.portal.azure.com/#blade/Microsoft_Azure_Storage/ContainerMenuBlade/overview/storageAccountId/%2Fsubscriptions%2F102a16c3-37d3-48a8-9237-4c9b1e8e80e0%2FresourceGroups%2FAutoMLSampleNotebooksData%2Fproviders%2FMicrosoft.Storage%2FstorageAccounts%2Fautomlsamplenotebookdata/path/automl-sample-notebook-data/etag/%220x8D84EAA65DE50B7%22/defaultEncryptionScope/%24account-encryption-key/denyEncryptionScopeOverride//defaultId//publicAccessVal/Container)\n", "![image-4.png](mm-1.png)\n", "\n", "['oj-inference-tabular'](https://ms.portal.azure.com/#blade/Microsoft_Azure_Storage/ContainerMenuBlade/overview/storageAccountId/%2Fsubscriptions%2F102a16c3-37d3-48a8-9237-4c9b1e8e80e0%2FresourceGroups%2FAutoMLSampleNotebooksData%2Fproviders%2FMicrosoft.Storage%2FstorageAccounts%2Fautomlsamplenotebookdata/path/automl-sample-notebook-data/etag/%220x8D84EAA65DE50B7%22/defaultEncryptionScope/%24account-encryption-key/denyEncryptionScopeOverride//defaultId//publicAccessVal/Container)\n", "![image-3.png](mm-2.png)\n", "\n", "['oj-data-small-tabular'](https://ms.portal.azure.com/#blade/Microsoft_Azure_Storage/ContainerMenuBlade/overview/storageAccountId/%2Fsubscriptions%2F102a16c3-37d3-48a8-9237-4c9b1e8e80e0%2FresourceGroups%2FAutoMLSampleNotebooksData%2Fproviders%2FMicrosoft.Storage%2FstorageAccounts%2Fautomlsamplenotebookdata/path/automl-sample-notebook-data/etag/%220x8D84EAA65DE50B7%22/defaultEncryptionScope/%24account-encryption-key/denyEncryptionScopeOverride//defaultId//publicAccessVal/Container)\n", "\n", "![image-5.png](mm-3.png)\n", "\n", "['oj-inference-small-tabular'](https://ms.portal.azure.com/#blade/Microsoft_Azure_Storage/ContainerMenuBlade/overview/storageAccountId/%2Fsubscriptions%2F102a16c3-37d3-48a8-9237-4c9b1e8e80e0%2FresourceGroups%2FAutoMLSampleNotebooksData%2Fproviders%2FMicrosoft.Storage%2FstorageAccounts%2Fautomlsamplenotebookdata/path/automl-sample-notebook-data/etag/%220x8D84EAA65DE50B7%22/defaultEncryptionScope/%24account-encryption-key/denyEncryptionScopeOverride//defaultId//publicAccessVal/Container)\n", "![image-6.png](mm-4.png)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### 2.2 Register the blob container as DataStore\n", "\n", "A Datastore is a place where data can be stored that is then made accessible to a compute either by means of mounting or copying the data to the compute target.\n", "\n", "Please refer to [Datastore](https://docs.microsoft.com/en-us/python/api/azureml-core/azureml.core.datastore(class)?view=azure-ml-py) documentation on how to access data from Datastore.\n", "\n", "In this next step, we will be registering blob storage as datastore to the Workspace." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from azureml.core import Datastore\n", "\n", "# Please change the following to point to your own blob container and pass in account_key\n", "blob_datastore_name = \"automl_many_models\"\n", "container_name = \"automl-sample-notebook-data\"\n", "account_name = \"automlsamplenotebookdata\"\n", "\n", "oj_datastore = Datastore.register_azure_blob_container(\n", " workspace=ws,\n", " datastore_name=blob_datastore_name,\n", " container_name=container_name,\n", " account_name=account_name,\n", " create_if_not_exists=True,\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### 2.3 Using tabular datasets \n", "\n", "Now that the datastore is available from the Workspace, [TabularDataset](https://docs.microsoft.com/en-us/python/api/azureml-core/azureml.data.tabular_dataset.tabulardataset?view=azure-ml-py) can be created. Datasets in Azure Machine Learning are references to specific data in a Datastore. We are using TabularDataset, so that users who have their data which can be in one or many files (*.parquet or *.csv) and have not split up data according to group columns needed for training, can do so using out of box support for 'partiion_by' feature of TabularDataset shown in section 5.0 below." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "gather": { "logged": 1613007017296 } }, "outputs": [], "source": [ "from azureml.core import Dataset\n", "\n", "ds_name_small = \"oj-data-small-tabular\"\n", "input_ds_small = Dataset.Tabular.from_delimited_files(\n", " path=oj_datastore.path(ds_name_small + \"/\"), validate=False\n", ")\n", "\n", "inference_name_small = \"oj-inference-small-tabular\"\n", "inference_ds_small = Dataset.Tabular.from_delimited_files(\n", " path=oj_datastore.path(inference_name_small + \"/\"), validate=False\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 3.0 Build the training pipeline\n", "Now that the dataset, WorkSpace, and datastore are set up, we can put together a pipeline for training.\n", "\n", "> Note that if you have an AzureML Data Scientist role, you will not have permission to create compute resources. Talk to your workspace or IT admin to create the compute targets described in this section, if they do not already exist." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Choose a compute target\n", "\n", "You will need to create a [compute target](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-set-up-training-targets#amlcompute) for your AutoML run. In this tutorial, you create AmlCompute as your training compute resource.\n", "\n", "\\*\\*Creation of AmlCompute takes approximately 5 minutes.**\n", "\n", "If the AmlCompute with that name is already in your workspace this code will skip the creation process. As with other Azure services, there are limits on certain resources (e.g. AmlCompute) associated with the Azure Machine Learning service. Please read this [article](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-manage-quotas) on the default limits and how to request more quota." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "gather": { "logged": 1613007037308 } }, "outputs": [], "source": [ "from azureml.core.compute import ComputeTarget, AmlCompute\n", "\n", "# Name your cluster\n", "compute_name = \"mm-compute\"\n", "\n", "\n", "if compute_name in ws.compute_targets:\n", " compute_target = ws.compute_targets[compute_name]\n", " if compute_target and type(compute_target) is AmlCompute:\n", " print(\"Found compute target: \" + compute_name)\n", "else:\n", " print(\"Creating a new compute target...\")\n", " provisioning_config = AmlCompute.provisioning_configuration(\n", " vm_size=\"STANDARD_D16S_V3\", max_nodes=20\n", " )\n", " # Create the compute target\n", " compute_target = ComputeTarget.create(ws, compute_name, provisioning_config)\n", "\n", " # Can poll for a minimum number of nodes and for a specific timeout.\n", " # If no min node count is provided it will use the scale settings for the cluster\n", " compute_target.wait_for_completion(\n", " show_output=True, min_node_count=None, timeout_in_minutes=20\n", " )\n", "\n", " # For a more detailed view of current cluster status, use the 'status' property\n", " print(compute_target.status.serialize())" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Set up training parameters\n", "\n", "This dictionary defines the AutoML and many models settings. For this forecasting task we need to define several settings including the name of the time column, the maximum forecast horizon, and the partition column name definition.\n", "\n", "| Property | Description|\n", "| :--------------- | :------------------- |\n", "| **task** | forecasting |\n", "| **primary_metric** | This is the metric that you want to optimize.
Forecasting supports the following primary metrics
spearman_correlation
normalized_root_mean_squared_error
r2_score
normalized_mean_absolute_error |\n", "| **blocked_models** | Blocked models won't be used by AutoML. |\n", "| **iteration_timeout_minutes** | Maximum amount of time in minutes that the model can train. This is optional but provides customers with greater control on exit criteria. |\n", "| **iterations** | Number of models to train. This is optional but provides customers with greater control on exit criteria. |\n", "| **experiment_timeout_hours** | Maximum amount of time in hours that the experiment can take before it terminates. This is optional but provides customers with greater control on exit criteria. |\n", "| **label_column_name** | The name of the label column. |\n", "| **forecast_horizon** | The forecast horizon is how many periods forward you would like to forecast. This integer horizon is in units of the timeseries frequency (e.g. daily, weekly). Periods are inferred from your data. |\n", "| **n_cross_validations** | Number of cross validation splits. Rolling Origin Validation is used to split time-series in a temporally consistent way. |\n", "| **enable_early_stopping** | Flag to enable early termination if the score is not improving in the short term. |\n", "| **time_column_name** | The name of your time column. |\n", "| **enable_engineered_explanations** | Engineered feature explanations will be downloaded if enable_engineered_explanations flag is set to True. By default it is set to False to save storage space. |\n", "| **time_series_id_column_name** | The column names used to uniquely identify timeseries in data that has multiple rows with the same timestamp. |\n", "| **track_child_runs** | Flag to disable tracking of child runs. Only best run is tracked if the flag is set to False (this includes the model and metrics of the run). |\n", "| **pipeline_fetch_max_batch_size** | Determines how many pipelines (training algorithms) to fetch at a time for training, this helps reduce throttling when training at large scale. |\n", "| **partition_column_names** | The names of columns used to group your models. For timeseries, the groups must not split up individual time-series. That is, each group must contain one or more whole time-series. |" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "gather": { "logged": 1613007061544 } }, "outputs": [], "source": [ "from azureml.train.automl.runtime._many_models.many_models_parameters import (\n", " ManyModelsTrainParameters,\n", ")\n", "\n", "partition_column_names = [\"Store\", \"Brand\"]\n", "automl_settings = {\n", " \"task\": \"forecasting\",\n", " \"primary_metric\": \"normalized_root_mean_squared_error\",\n", " \"iteration_timeout_minutes\": 10, # This needs to be changed based on the dataset. We ask customer to explore how long training is taking before settings this value\n", " \"iterations\": 15,\n", " \"experiment_timeout_hours\": 0.25,\n", " \"label_column_name\": \"Quantity\",\n", " \"n_cross_validations\": 3,\n", " \"time_column_name\": \"WeekStarting\",\n", " \"drop_column_names\": \"Revenue\",\n", " \"max_horizon\": 6,\n", " \"grain_column_names\": partition_column_names,\n", " \"track_child_runs\": False,\n", "}\n", "\n", "mm_paramters = ManyModelsTrainParameters(\n", " automl_settings=automl_settings, partition_column_names=partition_column_names\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Set up many models pipeline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Parallel run step is leveraged to train multiple models at once. To configure the ParallelRunConfig you will need to determine the appropriate number of workers and nodes for your use case. The process_count_per_node is based off the number of cores of the compute VM. The node_count will determine the number of master nodes to use, increasing the node count will speed up the training process.\n", "\n", "| Property | Description|\n", "| :--------------- | :------------------- |\n", "| **experiment** | The experiment used for training. |\n", "| **train_data** | The file dataset to be used as input to the training run. |\n", "| **node_count** | The number of compute nodes to be used for running the user script. We recommend to start with 3 and increase the node_count if the training time is taking too long. |\n", "| **process_count_per_node** | Process count per node, we recommend 2:1 ratio for number of cores: number of processes per node. eg. If node has 16 cores then configure 8 or less process count per node or optimal performance. |\n", "| **train_pipeline_parameters** | The set of configuration parameters defined in the previous section. |\n", "\n", "Calling this method will create a new aggregated dataset which is generated dynamically on pipeline execution." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from azureml.contrib.automl.pipeline.steps import AutoMLPipelineBuilder\n", "\n", "\n", "training_pipeline_steps = AutoMLPipelineBuilder.get_many_models_train_steps(\n", " experiment=experiment,\n", " train_data=input_ds_small,\n", " compute_target=compute_target,\n", " node_count=2,\n", " process_count_per_node=8,\n", " run_invocation_timeout=920,\n", " train_pipeline_parameters=mm_paramters,\n", ")" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from azureml.pipeline.core import Pipeline\n", "\n", "training_pipeline = Pipeline(ws, steps=training_pipeline_steps)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Submit the pipeline to run\n", "Next we submit our pipeline to run. The whole training pipeline takes about 40m using a STANDARD_D16S_V3 VM with our current ParallelRunConfig setting." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "training_run = experiment.submit(training_pipeline)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "training_run.wait_for_completion(show_output=False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Check the run status, if training_run is in completed state, continue to forecasting. If training_run is in another state, check the portal for failures." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 5.0 Publish and schedule the train pipeline (Optional)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 5.1 Publish the pipeline\n", "\n", "Once you have a pipeline you're happy with, you can publish a pipeline so you can call it programmatically later on. See this [tutorial](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-create-your-first-pipeline#publish-a-pipeline) for additional information on publishing and calling pipelines." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# published_pipeline = training_pipeline.publish(name = 'automl_train_many_models',\n", "# description = 'train many models',\n", "# version = '1',\n", "# continue_on_step_failure = False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 7.2 Schedule the pipeline\n", "You can also [schedule the pipeline](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-schedule-pipelines) to run on a time-based or change-based schedule. This could be used to automatically retrain models every month or based on another trigger such as data drift." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# from azureml.pipeline.core import Schedule, ScheduleRecurrence\n", "\n", "# training_pipeline_id = published_pipeline.id\n", "\n", "# recurrence = ScheduleRecurrence(frequency=\"Month\", interval=1, start_time=\"2020-01-01T09:00:00\")\n", "# recurring_schedule = Schedule.create(ws, name=\"automl_training_recurring_schedule\",\n", "# description=\"Schedule Training Pipeline to run on the first day of every month\",\n", "# pipeline_id=training_pipeline_id,\n", "# experiment_name=experiment.name,\n", "# recurrence=recurrence)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 6.0 Forecasting" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Set up output dataset for inference data\n", "Output of inference can be represented as [OutputFileDatasetConfig](https://docs.microsoft.com/en-us/python/api/azureml-core/azureml.data.output_dataset_config.outputdatasetconfig?view=azure-ml-py) object and OutputFileDatasetConfig can be registered as a dataset. " ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from azureml.data import OutputFileDatasetConfig\n", "\n", "output_inference_data_ds = OutputFileDatasetConfig(\n", " name=\"many_models_inference_output\", destination=(dstore, \"oj/inference_data/\")\n", ").register_on_complete(name=\"oj_inference_data_ds\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For many models we need to provide the ManyModelsInferenceParameters object.\n", "\n", "#### ManyModelsInferenceParameters arguments\n", "| Property | Description|\n", "| :--------------- | :------------------- |\n", "| **partition_column_names** | List of column names that identifies groups. |\n", "| **target_column_name** | \\[Optional] Column name only if the inference dataset has the target. |\n", "| **time_column_name** | \\[Optional] Column name only if it is timeseries. |\n", "| **many_models_run_id** | \\[Optional] Many models run id where models were trained. |\n", "\n", "#### get_many_models_batch_inference_steps arguments\n", "| Property | Description|\n", "| :--------------- | :------------------- |\n", "| **experiment** | The experiment used for inference run. |\n", "| **inference_data** | The data to use for inferencing. It should be the same schema as used for training.\n", "| **compute_target** | The compute target that runs the inference pipeline.|\n", "| **node_count** | The number of compute nodes to be used for running the user script. We recommend to start with the number of cores per node (varies by compute sku). |\n", "| **process_count_per_node** | The number of processes per node.\n", "| **train_run_id** | \\[Optional\\] The run id of the hierarchy training, by default it is the latest successful training many model run in the experiment. |\n", "| **train_experiment_name** | \\[Optional\\] The train experiment that contains the train pipeline. This one is only needed when the train pipeline is not in the same experiement as the inference pipeline. |\n", "| **process_count_per_node** | \\[Optional\\] The number of processes per node, by default it's 4. |" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from azureml.contrib.automl.pipeline.steps import AutoMLPipelineBuilder\n", "from azureml.train.automl.runtime._many_models.many_models_parameters import (\n", " ManyModelsInferenceParameters,\n", ")\n", "\n", "mm_parameters = ManyModelsInferenceParameters(\n", " partition_column_names=[\"Store\", \"Brand\"],\n", " time_column_name=\"WeekStarting\",\n", " target_column_name=\"Quantity\",\n", ")\n", "\n", "inference_steps = AutoMLPipelineBuilder.get_many_models_batch_inference_steps(\n", " experiment=experiment,\n", " inference_data=inference_ds_small,\n", " node_count=2,\n", " process_count_per_node=8,\n", " compute_target=compute_target,\n", " run_invocation_timeout=300,\n", " output_datastore=output_inference_data_ds,\n", " train_run_id=training_run.id,\n", " train_experiment_name=training_run.experiment.name,\n", " inference_pipeline_parameters=mm_parameters,\n", ")" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from azureml.pipeline.core import Pipeline\n", "\n", "inference_pipeline = Pipeline(ws, steps=inference_steps)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "inference_run = experiment.submit(inference_pipeline)\n", "inference_run.wait_for_completion(show_output=False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Retrieve results\n", "\n", "The forecasting pipeline forecasts the orange juice quantity for a Store by Brand. The pipeline returns one file with the predictions for each store and outputs the result to the forecasting_output Blob container. The details of the blob container is listed in 'forecasting_output.txt' under Outputs+logs. \n", "\n", "The following code snippet:\n", "1. Downloads the contents of the output folder that is passed in the parallel run step \n", "2. Reads the parallel_run_step.txt file that has the predictions as pandas dataframe and \n", "3. Displays the top 10 rows of the predictions" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from azureml.contrib.automl.pipeline.steps.utilities import get_output_from_mm_pipeline\n", "\n", "forecasting_results_name = \"forecasting_results\"\n", "forecasting_output_name = \"many_models_inference_output\"\n", "forecast_file = get_output_from_mm_pipeline(\n", " inference_run, forecasting_results_name, forecasting_output_name\n", ")\n", "df = pd.read_csv(forecast_file, delimiter=\" \", header=None)\n", "df.columns = [\n", " \"Week Starting\",\n", " \"Store\",\n", " \"Brand\",\n", " \"Quantity\",\n", " \"Advert\",\n", " \"Price\",\n", " \"Revenue\",\n", " \"Predicted\",\n", "]\n", "print(\n", " \"Prediction has \", df.shape[0], \" rows. Here the first 10 rows are being displayed.\"\n", ")\n", "df.head(10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 7.0 Publish and schedule the inference pipeline (Optional)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 7.1 Publish the pipeline\n", "\n", "Once you have a pipeline you're happy with, you can publish a pipeline so you can call it programmatically later on. See this [tutorial](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-create-your-first-pipeline#publish-a-pipeline) for additional information on publishing and calling pipelines." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# published_pipeline_inf = inference_pipeline.publish(name = 'automl_forecast_many_models',\n", "# description = 'forecast many models',\n", "# version = '1',\n", "# continue_on_step_failure = False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 7.2 Schedule the pipeline\n", "You can also [schedule the pipeline](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-schedule-pipelines) to run on a time-based or change-based schedule. This could be used to automatically retrain or forecast models every month or based on another trigger such as data drift." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# from azureml.pipeline.core import Schedule, ScheduleRecurrence\n", "\n", "# forecasting_pipeline_id = published_pipeline.id\n", "\n", "# recurrence = ScheduleRecurrence(frequency=\"Month\", interval=1, start_time=\"2020-01-01T09:00:00\")\n", "# recurring_schedule = Schedule.create(ws, name=\"automl_forecasting_recurring_schedule\",\n", "# description=\"Schedule Forecasting Pipeline to run on the first day of every week\",\n", "# pipeline_id=forecasting_pipeline_id,\n", "# experiment_name=experiment.name,\n", "# recurrence=recurrence)" ] } ], "metadata": { "authors": [ { "name": "jialiu" } ], "categories": [ "how-to-use-azureml", "automated-machine-learning" ], "kernelspec": { "display_name": "Python 3.6", "language": "python", "name": "python36" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.8" } }, "nbformat": 4, "nbformat_minor": 4 }