fixed links to zenml docs (#28)

zenml-io · Mar 27, 2023 · 64a2b2d · 64a2b2d
1 parent 6cd2985
commit 64a2b2d
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 4 deletions.
diff --git a/1-1_Pipelines.ipynb b/1-1_Pipelines.ipynb
@@ -68,10 +68,11 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "As an ML practitioner, you are probably familiar with building ML models using Scikit-learn, PyTorch, TensorFlow, or similar. An **[ML Pipeline](https://docs.zenml.io/developer-guide/steps-and-pipelines)** is simply an extension, including other steps you would typically do before or after building a model, like data acquisition, preprocessing, model deployment, or monitoring. The ML pipeline essentially defines a step-by-step procedure of your work as an ML practitioner. Defining ML pipelines explicitly in code is great because:\n",
+    "As an ML practitioner, you are probably familiar with building ML models using Scikit-learn, PyTorch, TensorFlow, or similar. An **[ML Pipeline](https://docs.zenml.io/starter-guide/pipelines)** is simply an extension, including other steps you would typically do before or after building a model, like data acquisition, preprocessing, model deployment, or monitoring. The ML pipeline essentially defines a step-by-step procedure of your work as an ML practitioner. Defining ML pipelines explicitly in code is great because:\n",
     "- We can easily rerun all of our work, not just the model, eliminating bugs and making our models easier to reproduce.\n",
     "- Data and models can be versioned and tracked, so we can see at a glance which dataset a model was trained on and how it compares to other models.\n",
     "- If the entire pipeline is coded up, we can automate many operational tasks, like retraining and redeploying models when the underlying problem or data changes or rolling out new and improved models with CI/CD workflows.\n",
@@ -80,11 +81,12 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "## ZenML Setup\n",
-    "Throughout this series, we will define our ML pipelines using [ZenML](https://github.com/zenml-io/zenml/). ZenML is an excellent tool for this task, as it is straightforward and intuitive to use and has [integrations](https://docs.zenml.io/mlops-stacks/integrations) with most of the advanced MLOps tools we will want to use later. Make sure you have ZenML installed (via `pip install zenml`). Next, let's run some commands to make sure you start with a fresh ML stack."
+    "Throughout this series, we will define our ML pipelines using [ZenML](https://github.com/zenml-io/zenml/). ZenML is an excellent tool for this task, as it is straightforward and intuitive to use and has [integrations](https://docs.zenml.io/component-gallery/integrations) with most of the advanced MLOps tools we will want to use later. Make sure you have ZenML installed (via `pip install zenml`). Next, let's run some commands to make sure you start with a fresh ML stack."
    ]
   },
   {
@@ -142,14 +144,15 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "In practice, your ML workflows will, of course, be much more complicated than that. You might have complex preprocessing that you do not want to redo every time you train a model, you will need to compare the performance of different models, deploy them in a production setting, and much more. Here ML pipelines come into play, allowing us to define our workflows in modular steps that we can then mix and match.\n",
     "\n",
     "![Digits Pipeline](_assets/1-1/digits_pipeline.png)\n",
     "\n",
-    "We can identify three distinct steps in our example: data loading, model training, and model evaluation. Let us now define each of them as a ZenML **[Pipeline Step](https://docs.zenml.io/developer-guide/steps-and-pipelines#step)** simply by moving each step to its own function and decorating them with ZenML's `@step` [Python decorator](https://realpython.com/primer-on-python-decorators/)."
+    "We can identify three distinct steps in our example: data loading, model training, and model evaluation. Let us now define each of them as a ZenML **[Pipeline Step](https://docs.zenml.io/starter-guide/pipelines#step)** simply by moving each step to its own function and decorating them with ZenML's `@step` [Python decorator](https://realpython.com/primer-on-python-decorators/)."
    ]
   },
   {

diff --git a/1-2_Artifact_Lineage.ipynb b/1-2_Artifact_Lineage.ipynb
@@ -308,17 +308,19 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "## Artifact Storage\n",
     "\n",
     "You might now wonder how our ML pipelines can keep track of which artifacts changed and which did not. This requires several additional MLOps components that you would typically have to set up and configure yourself. Luckily, ZenML automatically set this up for us.\n",
     "\n",
-    "Under the hood, all the artifacts in our ML pipeline are automatically stored in an [Artifact Store](https://docs.zenml.io/component-gallery/artifact-stores). By default, this is simply a place in your local file system, but we could also configure ZenML to store this data in a cloud bucket like [Amazon S3](https://docs.zenml.io/component-gallery/artifact-stores/amazon-s3) or any other place instead. We will see this in more detail when we migrate our MLOps stack to the cloud in a later chapter."
+    "Under the hood, all the artifacts in our ML pipeline are automatically stored in an [Artifact Store](https://docs.zenml.io/component-gallery/artifact-stores). By default, this is simply a place in your local file system, but we could also configure ZenML to store this data in a cloud bucket like [Amazon S3](https://docs.zenml.io/component-gallery/artifact-stores/s3) or any other place instead. We will see this in more detail when we migrate our MLOps stack to the cloud in a later chapter."
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [