diff --git a/ChangeLog.md b/ChangeLog.md
index 7d333d4a..bdf277a6 100644
--- a/ChangeLog.md
+++ b/ChangeLog.md
@@ -5,7 +5,7 @@ Starting with v1.31.6, this file will contain a record of major features and upd
## Upcoming
- Fixed Dockerfile builds breaking with AL2023 ([Link to PR](https://github.com/aws/graph-notebook/pull/466))
- Fixed `--store-to` option for several magics ([Link to PR](https://github.com/aws/graph-notebook/pull/463))
-- Fixed broken documentation links in Neptune-ML-04-Introduction-to-Edge-Classification-Gremlin ([Link to PR](https://github.com/aws/graph-notebook/pull/467))
+- Fixed broken documentation links in Neptune ML notebooks ([PR #1](https://github.com/aws/graph-notebook/pull/467)) ([PR #2](https://github.com/aws/graph-notebook/pull/468))
## Release 3.7.3 (March 14, 2023)
- Fixed detailed mode output for graph summary requests ([Link to PR](https://github.com/aws/graph-notebook/pull/461))
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-00-Getting-Started-with-Neptune-ML-Gremlin.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-00-Getting-Started-with-Neptune-ML-Gremlin.ipynb
index 62160b22..4ad57de5 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-00-Getting-Started-with-Neptune-ML-Gremlin.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-00-Getting-Started-with-Neptune-ML-Gremlin.ipynb
@@ -30,7 +30,7 @@
"\n",
"Graphs and graph data is all about using the values and connections within that data to provide novel insight. However one common issue with graph data is that it is frequently incomplete, meaning that it contains missing property values or connections. While incomplete data is not unique to graphs the connected nature how we want to use graph data makes these gaps even more impactful, usually lead to inefficent traversals and/or incorrect results. Neptune ML was released to help mitigate these issues by integrating Machine Learning (ML) models into real time graph traversals to predect/infer missing graph elements such as properties and connections. \n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. These models cover many common use cases such as:\n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. These models cover many common use cases such as:\n",
"\n",
"* [Identifying fradulent transactions](https://aws.amazon.com/blogs/machine-learning/detecting-fraud-in-heterogeneous-networks-using-amazon-sagemaker-and-deep-graph-library/)\n",
"* Predicting group membership in a social or identity network\n",
@@ -95,7 +95,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"## Loading data \n",
"\n",
@@ -214,7 +214,7 @@
"\n",
"In your use case the models will require configuration and training parameter adjustments to maximize the accuracy of the inferences they generate. These pretrained models use the default parameters to demonstrate the base accuracy of the models prior to tuning. Additional information on how to tune these models via the links below:\n",
"\n",
- "* [Training File Configuration](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-on-graphs-processing-training-config-file.html)\n",
+ "* [Training File Configuration](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-processing-training-config-file.html)\n",
"* [Tuning Hyperparameters](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-customizing-hyperparams.html)\n",
"* [Improving Model Performance](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-improve-model-performance.html)\n",
"\n",
@@ -713,4 +713,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
-}
+}
\ No newline at end of file
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-01-Introduction-to-Node-Classification-Gremlin.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-01-Introduction-to-Node-Classification-Gremlin.ipynb
index 64a9d378..68558cec 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-01-Introduction-to-Node-Classification-Gremlin.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-01-Introduction-to-Node-Classification-Gremlin.ipynb
@@ -20,7 +20,7 @@
"\n",
"Note: This notebook take approximately 1 hour to complete
\n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph.\n",
"\n",
"For this notebook we are going to show how to perform a common machine learning task known as **node classification**. Node classification is a common semi-supervised machine learning task where a model built using labeled nodes, ones where the property value exists, can predict the value (or class) of the nodes. Node classification is not unique to GNN based models (look at DeepWalk or node2vec) but the GNN based models in Neptune ML provide additional context to the predictions by combining the connectivity and features of the local neighborhood of a node to create a more predictive model.\n",
"\n",
@@ -87,7 +87,7 @@
{
"cell_type": "markdown",
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"# Load the data\n",
"The first step in building a Neptune ML model is to load data into the Neptune cluster. Loading data for Neptune ML follows the standard process of ingesting data into Amazon Neptune, for this example we'll be using the Bulk Loader. \n",
@@ -262,7 +262,7 @@
"\n",
"# Export the data and model configuration\n",
"\n",
- "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
+ "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
],
"metadata": {}
},
@@ -271,7 +271,7 @@
"source": [
"With our product knowledge graph loaded we are ready to export the data and configuration which will be used to train the ML model. \n",
"\n",
- "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required. \n",
+ "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required.\n",
"\n",
"Note: The configuration used in this notebook specifies only a minimal set of configuration options meaning that our model's predictions are not as accurate as they could be. The parameters included in this configuration are one of a couple of sets of options available to the end user to tune the model and optimize the accuracy of the resulting predictions.
\n",
"\n",
@@ -313,7 +313,7 @@
"\n",
"In our export example below we have specified that the `title` property of our `movie` should be exported and trained as a `text_word2vec` feature and that our `age` field should range from 0-100 and that data should be bucketed into 10 distinct groups. \n",
"\n",
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
"\n",
"Running the cell below we set the export configuration and run the export process. Neptune export is capable of automatically creating a clone of the cluster by setting `cloneCluster=True` which takes about 20 minutes to complete and will incur additional costs while the cloned cluster is running. Exporting from the existing cluster takes about 5 minutes but requires that the `neptune_query_timeout` parameter in the [parameter group](https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html) is set to a large enough value (>72000) to prevent timeout errors."
],
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-02-Introduction-to-Node-Regression-Gremlin.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-02-Introduction-to-Node-Regression-Gremlin.ipynb
index 75e42f75..a5d65598 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-02-Introduction-to-Node-Regression-Gremlin.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-02-Introduction-to-Node-Regression-Gremlin.ipynb
@@ -27,7 +27,7 @@
"\n",
"**Note:** This notebook take approximately 1 hour to complete\n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph.\n",
"\n",
"For this notebook we are going to show how to perform a common machine learning task known as **node regression**. Node regression is a common semi-supervised machine learning task where a model built using labeled nodes, ones where the property value exists, can predict the numerical value of propertues on a nodes. Node regression is not unique to GNN based models (look at DeepWalk or node2vec) but the GNN based models in Neptune ML provide additional context to the predictions by combining the connectivity and features of the local neighborhood of a node to create a more predictive model.\n",
"\n",
@@ -87,7 +87,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"# Load the data\n",
"The first step in building a Neptune ML model is to load data into the Neptune cluster. Loading data for Neptune ML follows the standard process of ingesting data into Amazon Neptune, for this example we'll be using the Bulk Loader. \n",
@@ -256,7 +256,7 @@
"source": [
"# Export the data and model configuration\n",
"\n",
- "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
+ "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
]
},
{
@@ -265,7 +265,7 @@
"source": [
"With our product knowledge graph loaded we are ready to export the data and configuration which will be used to train the ML model. \n",
"\n",
- "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required. \n",
+ "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required.\n",
"\n",
"Note: The configuration used in this notebook specifies only a minimal set of configuration options meaning that our model's predictions are not as accurate as they could be. The parameters included in this configuration are one of a couple of sets of options available to the end user to tune the model and optimize the accuracy of the resulting predictions.
\n",
"\n",
@@ -309,7 +309,7 @@
"\n",
"In our export example below we have specified that the `title` property of our `movie` should be exported and trained as a `text_word2vec` feature and that our `age` field should range from 0-100 and that data should be bucketed into 10 distinct groups. \n",
"\n",
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
"\n",
"Running the cell below we set the export configuration and run the export process. Neptune export is capable of automatically creating a clone of the cluster by setting `cloneCluster=True` which takes about 20 minutes to complete and will incur additional costs while the cloned cluster is running. Exporting from the existing cluster takes about 5 minutes but requires that the `neptune_query_timeout` parameter in the [parameter group](https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html) is set to a large enough value (>72000) to prevent timeout errors."
]
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-03-Introduction-to-Link-Prediction-Gremlin.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-03-Introduction-to-Link-Prediction-Gremlin.ipynb
index e77a06e9..2f429ed9 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-03-Introduction-to-Link-Prediction-Gremlin.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-03-Introduction-to-Link-Prediction-Gremlin.ipynb
@@ -26,7 +26,7 @@
"\n",
"Note: This notebook take approximately 1 hour to complete
\n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph.\n",
"\n",
"For this notebook we are going to show how to perform a common machine learning task known as **link prediction**. Link prediction is a unsupervised machine learning task where a model built using nodes and edges in the graph to predict whether an edge exists between two particular nodes. Link prediction is not unique to GNN based models (look at DeepWalk or node2vec) but the GNN based models in Neptune ML provide additional context to the predictions by combining the connectivity and features of the local neighborhood of a node to create a more predictive model.\n",
"\n",
@@ -85,7 +85,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"# Load the data\n",
"The first step in building a Neptune ML model is to load data into the Neptune cluster. Loading data for Neptune ML follows the standard process of ingesting data into Amazon Neptune, for this example we'll be using the Bulk Loader. \n",
@@ -253,7 +253,7 @@
"source": [
"# Export the data and model configuration\n",
"\n",
- "**Note:** Before exporting data ensure that Neptune Export has been configured as described here: [Neptune Export Service](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html#machine-learning-data-export-service-run-export)"
+ "**Note:** Before exporting data ensure that Neptune Export has been configured as described here: [Neptune Export Service](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-manual-setup.html#ml-manual-setup-export-svc)"
]
},
{
@@ -262,7 +262,7 @@
"source": [
"With our product knowledge graph loaded we are ready to export the data and configuration which will be used to train the ML model. \n",
"\n",
- "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required. \n",
+ "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required.\n",
"\n",
"Note: The configuration used in this notebook specifies only a minimal set of configuration options meaning that our model's predictions are not as accurate as they could be. The parameters included in this configuration are one of a couple of sets of options available to the end user to tune the model and optimize the accuracy of the resulting predictions.
\n",
"\n",
@@ -291,7 +291,7 @@
"\n",
"In our export example below we have specified that the `title` property of our `movie` should be exported and trained as a `text_word2vec` feature and that our `age` field should range from 0-100 and that data should be bucketed into 10 distinct groups. \n",
"\n",
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
"\n",
"Running the cell below we set the export configuration and run the export process. Neptune export is capable of automatically creating a clone of the cluster by setting `cloneCluster=True` which takes about 20 minutes to complete and will incur additional costs while the cloned cluster is running. Exporting from the existing cluster takes about 5 minutes but requires that the `neptune_query_timeout` parameter in the [parameter group](https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html) is set to a large enough value (>72000) to prevent timeout errors."
]
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-04-Introduction-to-Edge-Classification-Gremlin.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-04-Introduction-to-Edge-Classification-Gremlin.ipynb
index f4da8ef0..5f8b1548 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-04-Introduction-to-Edge-Classification-Gremlin.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-04-Introduction-to-Edge-Classification-Gremlin.ipynb
@@ -88,7 +88,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"# Load the data\n",
"The first step in building a Neptune ML model is to load data into the Neptune cluster. Loading data for Neptune ML follows the standard process of ingesting data into Amazon Neptune, for this example we'll be using the Bulk Loader. \n",
@@ -670,4 +670,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
-}
+}
\ No newline at end of file
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-05-Introduction-to-Edge-Regression-Gremlin.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-05-Introduction-to-Edge-Regression-Gremlin.ipynb
index 3dc05098..4c48a843 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-05-Introduction-to-Edge-Regression-Gremlin.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-05-Introduction-to-Edge-Regression-Gremlin.ipynb
@@ -26,7 +26,7 @@
"\n",
"Note: This notebook take approximately 1 hour to complete
\n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph.\n",
"\n",
"For this notebook we are going to show how to perform a common machine learning task known as **edge regression**. Edge regression is a common semi-supervised machine learning task where a model built using labeled edges, ones where the property value exists, can predict the numerical value of propertues on a nodes. Edge regression is not unique to GNN based models (look at DeepWalk or node2vec) but the GNN based models in Neptune ML provide additional context to the predictions by combining the connectivity and features of the local neighborhood of a node to create a more predictive model.\n",
"\n",
@@ -86,7 +86,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"# Load the data\n",
"The first step in building a Neptune ML model is to load data into the Neptune cluster. Loading data for Neptune ML follows the standard process of ingesting data into Amazon Neptune, for this example we'll be using the Bulk Loader. \n",
@@ -257,7 +257,7 @@
"source": [
"# Export the data and model configuration\n",
"\n",
- "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
+ "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
]
},
{
@@ -266,7 +266,7 @@
"source": [
"With our product knowledge graph loaded we are ready to export the data and configuration which will be used to train the ML model. \n",
"\n",
- "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required. \n",
+ "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required.\n",
"\n",
"Note: The configuration used in this notebook specifies only a minimal set of configuration options meaning that our model's predictions are not as accurate as they could be. The parameters included in this configuration are one of a couple of sets of options available to the end user to tune the model and optimize the accuracy of the resulting predictions.
\n",
"\n",
@@ -310,7 +310,7 @@
"\n",
"In our export example below we have specified that the `title` property of our `movie` should be exported and trained as a `text_word2vec` feature and that our `age` field should range from 0-100 and that data should be bucketed into 10 distinct groups. \n",
"\n",
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
"\n",
"Running the cell below we set the export configuration and run the export process. Neptune export is capable of automatically creating a clone of the cluster by setting `cloneCluster=True` which takes about 20 minutes to complete and will incur additional costs while the cloned cluster is running. Exporting from the existing cluster takes about 5 minutes but requires that the `neptune_query_timeout` parameter in the [parameter group](https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html) is set to a large enough value (>72000) to prevent timeout errors."
]
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-00-Getting-Started-with-Neptune-ML-SPARQL.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-00-Getting-Started-with-Neptune-ML-SPARQL.ipynb
index 626f9454..b1cf7b71 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-00-Getting-Started-with-Neptune-ML-SPARQL.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-00-Getting-Started-with-Neptune-ML-SPARQL.ipynb
@@ -30,7 +30,7 @@
"\n",
"Graphs and graph data is all about using the values and connections within that data to provide novel insight. However one common issue with graph data is that it is frequently incomplete, meaning that it contains missing property values or connections. While incomplete data is not unique to graphs the connected nature how we want to use graph data makes these gaps even more impactful, usually lead to inefficent traversals and/or incorrect results. Neptune ML was released to help mitigate these issues by integrating Machine Learning (ML) models into real time graph traversals to predect/infer missing graph elements such as properties and connections. \n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. These models cover many common use cases such as:\n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. These models cover many common use cases such as:\n",
"\n",
"* [Identifying fradulent transactions](https://aws.amazon.com/blogs/machine-learning/detecting-fraud-in-heterogeneous-networks-using-amazon-sagemaker-and-deep-graph-library/)\n",
"* Predicting group membership in a social or identity network\n",
@@ -214,7 +214,7 @@
"\n",
"In your use case the models will require configuration and training parameter adjustments to maximize the accuracy of the inferences they generate. These pretrained models use the default parameters to demonstrate the base accuracy of the models prior to tuning. Additional information on how to tune these models via the links below:\n",
"\n",
- "* [Training File Configuration](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-on-graphs-processing-training-config-file.html)\n",
+ "* [Training File Configuration](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-processing-training-config-file.html)\n",
"* [Tuning Hyperparameters](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-customizing-hyperparams.html)\n",
"* [Improving Model Performance](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-improve-model-performance.html)\n",
"\n",
@@ -785,4 +785,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
-}
+}
\ No newline at end of file
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-01-Introduction-to-Object-Classification-SPARQL.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-01-Introduction-to-Object-Classification-SPARQL.ipynb
index 0f554510..faa63072 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-01-Introduction-to-Object-Classification-SPARQL.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-01-Introduction-to-Object-Classification-SPARQL.ipynb
@@ -26,7 +26,7 @@
"\n",
"Note: This notebook take approximately 1 hour to complete
\n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph.\n",
"\n",
"For this notebook we are going to show how to perform a common machine learning task known as **node classification**. Node classification is a common semi-supervised machine learning task where a model built using existing nodes, which in RDF means existing string Literal values. Where the triple does not exist, we can predict the value of the Literal. Node classification is not unique to GNN based models (look at DeepWalk or node2vec) but the GNN based models in Neptune ML provide additional context to the predictions by combining the connectivity and features of the local neighborhood of a node to create a more predictive model.\n",
"\n",
@@ -87,7 +87,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"## Loading the data \n",
"\n",
@@ -147,7 +147,7 @@
"source": [
"# Export the data and model configuration\n",
"\n",
- "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
+ "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
]
},
{
@@ -156,7 +156,7 @@
"source": [
"With our product knowledge graph loaded we are ready to export the data and configuration which will be used to train the ML model. \n",
"\n",
- "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required. \n",
+ "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required.\n",
"\n",
"Note: The configuration used in this notebook specifies only a minimal set of configuration options meaning that our model's predictions are not as accurate as they could be. The parameters included in this configuration are one of a couple of sets of options available to the end user to tune the model and optimize the accuracy of the resulting predictions.
\n",
"\n",
@@ -198,7 +198,7 @@
"\n",
"In our export example below we have specified that the `title` property of our `movie` should be exported and trained as a `text_word2vec` feature and that our `age` field should range from 0-100 and that data should be bucketed into 10 distinct groups. \n",
"\n",
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
"\n",
"Running the cell below we set the export configuration and run the export process. Neptune export is capable of automatically creating a clone of the cluster by setting `cloneCluster=True` which takes about 20 minutes to complete and will incur additional costs while the cloned cluster is running. Exporting from the existing cluster takes about 5 minutes but requires that the `neptune_query_timeout` parameter in the [parameter group](https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html) is set to a large enough value (>72000) to prevent timeout errors."
]
@@ -742,4 +742,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
-}
+}
\ No newline at end of file
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-02-Introduction-to-Object-Regression-SPARQL.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-02-Introduction-to-Object-Regression-SPARQL.ipynb
index 3988fc6a..41f87a46 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-02-Introduction-to-Object-Regression-SPARQL.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-02-Introduction-to-Object-Regression-SPARQL.ipynb
@@ -27,12 +27,12 @@
"\n",
"**Note:** This notebook take approximately 1 hour to complete\n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a SPARQL query to predict the values of RDF Literals the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a SPARQL query to predict the values of RDF Literals the graph.\n",
"\n",
"For this notebook we are going to show how to perform a common machine learning task known as **regression**. \n",
"With regression we predict numerical values.\n",
"\n",
- "As this is an RDF graph, the value we are predicting is in the form of an typed . Regression is a common semi-supervised machine learning task where a model is built using existing values to predict missing values of the same and . Regression is not unique to GNN based models (look at DeepWalk or node2vec) but the GNN based models in Neptune ML provide additional context to the predictions by combining the connectivity and features of the local neighborhood of a node to create a more predictive model.\n",
+ "As this is an RDF graph, the value we are predicting is in the form of an <rdf:integer> typed <rdf:Literal>. Regression is a common semi-supervised machine learning task where a model is built using existing values to predict missing values of the same <rdf:subject> and <rdf:predicate>. Regression is not unique to GNN based models (look at DeepWalk or node2vec) but the GNN based models in Neptune ML provide additional context to the predictions by combining the connectivity and features of the local neighborhood of a node to create a more predictive model.\n",
"\n",
"Node regression is commonly used to solve many common buisness problems such as:\n",
"\n",
@@ -92,7 +92,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"# Load the data\n",
"The first step in building a Neptune ML model is to load data into the Neptune cluster. Loading data for Neptune ML follows the standard process of ingesting data into Amazon Neptune, for this example we'll be using the Bulk Loader. \n",
@@ -184,7 +184,7 @@
"source": [
"# Export the data and model configuration\n",
"\n",
- "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
+ "Note: Before exporting data ensure that Neptune Export has been configured as described here:
Neptune Export Service "
]
},
{
@@ -193,7 +193,7 @@
"source": [
"With our product knowledge graph loaded we are ready to export the data and configuration which will be used to train the ML model. \n",
"\n",
- "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required. \n",
+ "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required.\n",
"\n",
"Note: The configuration used in this notebook specifies only a minimal set of configuration options meaning that our model's predictions are not as accurate as they could be. The parameters included in this configuration are one of a couple of sets of options available to the end user to tune the model and optimize the accuracy of the resulting predictions.
\n",
"\n",
@@ -237,7 +237,7 @@
"\n",
"In our export example below we have specified that the `title` property of our `movie` should be exported and trained as a `text_word2vec` feature and that our `age` field should range from 0-100 and that data should be bucketed into 10 distinct groups. \n",
"\n",
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
"\n",
"Running the cell below we set the export configuration and run the export process. Neptune export is capable of automatically creating a clone of the cluster by setting `cloneCluster=True` which takes about 20 minutes to complete and will incur additional costs while the cloned cluster is running. Exporting from the existing cluster takes about 5 minutes but requires that the `neptune_query_timeout` parameter in the [parameter group](https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html) is set to a large enough value (>72000) to prevent timeout errors."
]
@@ -704,4 +704,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
-}
+}
\ No newline at end of file
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-03-Introduction-to-Link-Prediction-SPARQL.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-03-Introduction-to-Link-Prediction-SPARQL.ipynb
index 4636aa68..702fd000 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-03-Introduction-to-Link-Prediction-SPARQL.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-SPARQL/Neptune-ML-03-Introduction-to-Link-Prediction-SPARQL.ipynb
@@ -26,7 +26,7 @@
"\n",
"Note: This notebook take approximately 1 hour to complete
\n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a SPARQL query to predict statements in the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy to use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a SPARQL query to predict statements in the graph.\n",
"\n",
"For this notebook we are going to show how to perform a common machine learning task known as **link prediction**. Link prediction is a unsupervised machine learning task where a model built using existing RDF statements where the ``rdf:Statement`` is a relationship between two IRI's.\n",
"This model is then used to predict other statements.\n",
@@ -109,7 +109,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview).\n",
+ "If the check above did not say that this cluster is ready to run Neptune ML jobs then please check that the cluster meets all the pre-requisites defined [here](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html).\n",
"\n",
"# Load the data\n",
"The first step in building a Neptune ML model is to load data into the Neptune cluster. Loading data for Neptune ML follows the standard process of ingesting data into Amazon Neptune, for this example we'll be using the Bulk Loader. \n",
@@ -240,7 +240,7 @@
"source": [
"# Export the data and model configuration\n",
"\n",
- "**Note:** Before exporting data ensure that Neptune Export has been configured as described here: [Neptune Export Service](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html#machine-learning-data-export-service-run-export)"
+ "**Note:** Before exporting data ensure that Neptune Export has been configured as described here: [Neptune Export Service](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-manual-setup.html#ml-manual-setup-export-svc)"
]
},
{
@@ -249,7 +249,7 @@
"source": [
"With our product knowledge graph loaded we are ready to export the data and configuration which will be used to train the ML model. \n",
"\n",
- "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-service.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required. \n",
+ "The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object which specifies the type of machine learning model to build, in this example node classification, as well as any feature configurations required.\n",
"\n",
"Note: The configuration used in this notebook specifies only a minimal set of configuration options meaning that our model's predictions are not as accurate as they could be. The parameters included in this configuration are one of a couple of sets of options available to the end user to tune the model and optimize the accuracy of the resulting predictions.
\n",
"\n",
@@ -278,7 +278,7 @@
"\n",
"In our export example below we have specified that the `title` property of our `movie` should be exported and trained as a `text_word2vec` feature and that our `age` field should range from 0-100 and that data should be bucketed into 10 distinct groups. \n",
"\n",
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters \n",
"\n",
"Running the cell below we set the export configuration and run the export process. Neptune export is capable of automatically creating a clone of the cluster by setting `cloneCluster=True` which takes about 20 minutes to complete and will incur additional costs while the cloned cluster is running. Exporting from the existing cluster takes about 5 minutes but requires that the `neptune_query_timeout` parameter in the [parameter group](https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html) is set to a large enough value (>72000) to prevent timeout errors.\n",
"\n",
@@ -705,4 +705,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
-}
+}
\ No newline at end of file
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/01-People-Analytics/People-Analytics-using-Neptune-ML.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/01-People-Analytics/People-Analytics-using-Neptune-ML.ipynb
index 4dde9d83..89780c9d 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/01-People-Analytics/People-Analytics-using-Neptune-ML.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/01-People-Analytics/People-Analytics-using-Neptune-ML.ipynb
@@ -27,7 +27,7 @@
"\n",
"To demonstrate how you might accomplish this we are going to use a well-known dataset originally from IBM that contains HR attrition data. This dataset is available from [Kaggle](https://www.kaggle.com/pavansubhasht/ibm-hr-analytics-attrition-dataset) under the [Open Data Commons License]( https://opendatacommons.org/licenses/dbcl/1-0/).\n",
"\n",
- "To answer these questions, we will be using [Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) which is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy-to-use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph.\n",
+ "To answer these questions, we will be using [Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) which is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy-to-use mechanism to build/train/maintain these models and then use the predictive capabilities of these models within a Gremlin query to predict elements or property values in the graph.\n",
"\n",
"## Setup\n",
"\n",
@@ -561,7 +561,7 @@
"source": [
"Now that we have prepared our dataset, the first step in the process of building our Attrition Prediction model is to export the data and configuration. \n",
"\n",
- "**Note:** The details of how this is configured and the options available are covered in the [Neptune ML documentation](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export-parameters.html) as well as the [Neptune-ML-01-Introduction-to-Node-Classification-Gremlin notebook](https://github.com/aws/graph-notebook/blob/main/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-01-Introduction-to-Node-Classification-Gremlin.ipynb), so we will not cover the detail here.\n",
+ "**Note:** The details of how this is configured and the options available are covered in the [Neptune ML documentation](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html#machine-learning-params) as well as the [Neptune-ML-01-Introduction-to-Node-Classification-Gremlin notebook](https://github.com/aws/graph-notebook/blob/main/src/graph_notebook/notebooks/04-Machine-Learning/Neptune-ML-01-Introduction-to-Node-Classification-Gremlin.ipynb), so we will not cover the detail here.\n",
"\n",
"For our `Attrition` prediction model, we have specified our targets as the `Attrition` property of the `Employee` vertices as shown here:\n",
"```\n",
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/02-Job-Recommendation-Text-Encoding.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/02-Job-Recommendation-Text-Encoding.ipynb
index a03acbd2..ca335761 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/02-Job-Recommendation-Text-Encoding.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/02-Job-Recommendation-Text-Encoding.ipynb
@@ -30,7 +30,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of machine learning on graph structured data within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) to provide a simple mechanism for training and deploying these models. Since Neptune ML is integrated deeply with Amazon Neptune, predictions can be made directly through the Gremlin query language. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of machine learning on graph structured data within Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) to provide a simple mechanism for training and deploying these models. Since Neptune ML is integrated deeply with Amazon Neptune, predictions can be made directly through the Gremlin query language.\n",
"\n",
"Graph databases are useful for storing and navigating highly connected datasets across diverse industries such as financial services, advertising and product recommendations, cybersecurity, and more. The properties of these datasets also vary depending on the business domain. Let’s take a look at a graph that relates job applicants to job postings from [Kaggle](https://www.kaggle.com/competitions/job-recommendation/data). In this dataset, some properties are in numerical type, like the age or the count of job history for a user node. Other node or edge properties are in the form of raw text, like the self description property of a user node and the description or the requirement of a job node. These text properties in a graph can be stored in different languages, such as job posts from different countries. Also, the length of text properties can vary, with some being long and others short. Due to these requirements, customers need an efficient and effective way to encode text data at scale in a graph feature processing step.\n",
"\n",
@@ -345,7 +345,7 @@
"source": [
"### Step 2: Export Data: Choose the Right Text Encoders for Text Properties\n",
"\n",
- "**Note:** Before exporting data ensure that Neptune Export has been configured as described here: [Neptune Export Service](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). \n",
+ "**Note:** Before exporting data ensure that Neptune Export has been configured as described here: [Neptune Export Service](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-manual-setup.html#ml-manual-setup-export-svc).\n",
"\n",
"The export process is triggered by calling to the [Neptune Export service endpoint](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning-data-export.html). This call contains a configuration object, which specifies the link prediction task on the edge of `(\"user\", \"apply\", \"job\")` for this job recommendation problem. The configuration also specifies the desired feature processing to apply.\n",
"\n",
@@ -372,7 +372,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters "
+ "Important: The example below is an example of a minimal amount of the features of the model configuration parameters and will not create the most accurate model possible. Additional options are available for tuning this configuration to produce an optimal model are described here:
Neptune Export Process Parameters "
]
},
{
@@ -728,4 +728,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
-}
+}
\ No newline at end of file
diff --git a/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/03-Real-Time-Fraud-Detection-Using-Inductive-Inference.ipynb b/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/03-Real-Time-Fraud-Detection-Using-Inductive-Inference.ipynb
index 9beafff7..22bdb2d3 100644
--- a/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/03-Real-Time-Fraud-Detection-Using-Inductive-Inference.ipynb
+++ b/src/graph_notebook/notebooks/04-Machine-Learning/Sample-Applications/03-Real-Time-Fraud-Detection-Using-Inductive-Inference.ipynb
@@ -14,7 +14,7 @@
"\n",
"In this notebook, we are going to show how to perform real-time fraud detection using Neptune ML. \n",
"\n",
- "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html#machine-learning-overview) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models using the graph data stored in Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy-to-use mechanism to build, train, and maintain these models. Then, you can use the predictive capabilities of these models within a Gremlin or SPARQL query to predict elements or property values in the graph. \n",
+ "[Neptune ML](https://docs.aws.amazon.com/neptune/latest/userguide/machine-learning.html) is a feature of Amazon Neptune that enables users to automate the creation, management, and usage of Graph Neural Network (GNN) machine learning models using the graph data stored in Amazon Neptune. Neptune ML is built using [Amazon SageMaker](https://aws.amazon.com/sagemaker/) and [Deep Graph Library](https://www.dgl.ai/) and provides a simple and easy-to-use mechanism to build, train, and maintain these models. Then, you can use the predictive capabilities of these models within a Gremlin or SPARQL query to predict elements or property values in the graph.\n",
"\n",
"For this tutorial, we are going to see how we can build a real-time fraud detection solution using **real-time inductive inference** and a common graph machine learning task known as **node classification**. \n",
"\n",
@@ -66,7 +66,7 @@
"id": "3f359ffb",
"metadata": {},
"source": [
- "This notebook demonstrates how to use Neptune ML to perform fraud detection using the IEEE CIS dataset (https://www.kaggle.com/c/ieee-fraud-detection/data). The dataset contains Transaction and Identity tables, both having transaction ID as the primary key column. The tables have many anonymized columns, defining the relationships of the transaction records with cards, devices, products and other identifiers as shown by the figure below. "
+ "This notebook demonstrates how to use Neptune ML to perform fraud detection using the [IEEE CIS dataset](https://www.kaggle.com/c/ieee-fraud-detection/data). The dataset contains Transaction and Identity tables, both having transaction ID as the primary key column. The tables have many anonymized columns, defining the relationships of the transaction records with cards, devices, products and other identifiers as shown by the figure below."
]
},
{
@@ -1076,4 +1076,4 @@
},
"nbformat": 4,
"nbformat_minor": 5
-}
+}
\ No newline at end of file