ppai/land-cover: simplify tests and pin python version (#8501)

* rebase

* remove constraints file

* pin python version

* add copyright

* add python version

Author: davidcavazos

people-and-planet-ai/conftest.py

@@ -65,7 +65,7 @@ def unique_name(test_name: str, unique_id: str) -> str:
 
 
 @pytest.fixture(scope="session")
-def bucket_name(test_name: str, location: str, unique_id: str) -> str:
+def bucket_name(test_name: str, location: str, unique_id: str) -> Iterable[str]:
     # Override for local testing.
     if "GOOGLE_CLOUD_BUCKET" in os.environ:
         bucket_name = os.environ["GOOGLE_CLOUD_BUCKET"]
@@ -192,7 +192,6 @@ def aiplatform_cleanup(model_name: str, location: str, versions: list[str]) -> N
     )
 
 
-
 def run_notebook(
     ipynb_file: str,
     prelude: str = "",
@@ -203,7 +202,8 @@ def run_notebook(
     skip_shell_commands: bool = False,
     until_end: bool = False,
 ) -> None:
-    import nbclient
+    from nbclient.client import NotebookClient
+    from nbclient.exceptions import CellExecutionError
     import nbformat
 
     def notebook_filter_section(
@@ -283,10 +283,10 @@ def notebook_filter_section(
 
     # Run the notebook.
     error = ""
-    client = nbclient.NotebookClient(nb)
+    client = NotebookClient(nb)
     try:
         client.execute()
-    except nbclient.exceptions.CellExecutionError as e:
+    except CellExecutionError as e:
         # Remove colors and other escape characters to make it easier to read in the logs.
         #   https://stackoverflow.com/a/33925425
         error = re.sub(r"(\x9B|\x1B\[)[0-?]*[ -\/]*[@-~]", "", str(e))
@@ -299,22 +299,24 @@ def notebook_filter_section(
 
 def run_notebook_parallel(
     ipynb_file: str,
-    prelude: str,
-    sections: list[str],
-    variables: dict[str, dict] = {},
-    replace: dict[str, dict[str, str]] = {},
-    skip_shell_commands: list[str] = [],
+    sections: dict[str, dict],
+    prelude: str = "",
+    variables: dict = {},
+    replace: dict[str, str] = {},
+    skip_shell_commands: bool = False,
 ) -> None:
     args = [
         {
             "ipynb_file": ipynb_file,
             "section": section,
-            "prelude": prelude,
-            "variables": variables.get(section, {}),
-            "replace": replace.get(section, {}),
-            "skip_shell_commands": section in skip_shell_commands,
+            "prelude": params.get("prelude", prelude),
+            "variables": {**variables, **params.get("variables", {})},
+            "replace": {**replace, **params.get("replace", {})},
+            "skip_shell_commands": params.get(
+                "skip_shell_commands", skip_shell_commands
+            ),
         }
-        for section in sections
+        for section, params in sections.items()
     ]
     with multiprocessing.Pool(len(args)) as pool:
         pool.map(_run_notebook_section, args)

people-and-planet-ai/land-cover-classification/README.md

@@ -1,6 +1,10 @@
 # 🌍 Land cover classification -- _image segmentation_
 
-[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/land-cover-classification/README.ipynb)
+## [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/land-cover-classification/README.ipynb) 🌍 TensorFlow with Earth Engine introduction
+
+## [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/land-cover-classification/cloud-tensorflow.ipynb) ☁️ Scaling TensorFlow with Cloud
+
+## [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/land-cover-classification/land-cover-change.ipynb) 🗺️ Visualizing land cover change
 
 > [Watch the video in YouTube<br> ![thumbnail](http://img.youtube.com/vi/zImQf91ffFo/0.jpg)](https://youtu.be/zImQf91ffFo)
 

people-and-planet-ai/land-cover-classification/cloud-tensorflow.ipynb

@@ -162,8 +162,7 @@
       "source": [
         "repo_url = \"https://raw.githubusercontent.com/GoogleCloudPlatform/python-docs-samples/main/people-and-planet-ai/land-cover-classification\"\n",
         "\n",
-        "!wget --quiet -O requirements.txt {repo_url}/requirements.txt\n",
-        "!wget --quiet -O constraints.txt {repo_url}/constraints.txt"
+        "!wget --quiet {repo_url}/requirements.txt"
       ]
     },
     {
@@ -172,7 +171,7 @@
         "id": "rPWUYDKCwibV"
       },
       "source": [
-        "> 💡 For more information about the `requirements.txt` and `constraints.txt` files, see the [`pip` user guide](https://pip.pypa.io/en/stable/user_guide/)."
+        "> 💡 For more information about the `requirements.txt` files, see the [`pip` user guide](https://pip.pypa.io/en/stable/user_guide/)."
       ]
     },
     {
@@ -195,7 +194,7 @@
       "outputs": [],
       "source": [
         "# Install the dependencies.\n",
-        "!pip install --quiet -r requirements.txt -c constraints.txt\n",
+        "!pip install --quiet -r requirements.txt\n",
         "\n",
         "# Restart the runtime by ending the process.\n",
         "exit()"

people-and-planet-ai/land-cover-classification/constraints.txt

@@ -14,6 +14,7 @@
 
 from __future__ import annotations
 
+from collections.abc import Iterable
 import tempfile
 import textwrap
 
@@ -61,15 +62,15 @@ def model_path(bucket_name: str) -> str:
 
 
 @pytest.fixture(scope="session")
-def cloud_run_service_name(unique_name: str, location: str) -> str:
+def cloud_run_service_name(unique_name: str, location: str) -> Iterable[str]:
     # The notebook itself creates the service.
     service_name = unique_name
     yield service_name
     conftest.cloud_run_cleanup(service_name, location)
 
 
 @pytest.fixture(scope="session")
-def aiplatform_model_name(unique_name: str, location: str) -> str:
+def aiplatform_model_name(unique_name: str, location: str) -> Iterable[str]:
     # The notebook itself creates the service.
     model_name = unique_name.replace("-", "_")
     yield model_name
@@ -164,60 +165,62 @@ def test_land_cover_tensorflow(
                 ee_init()
                 """
             ),
-            sections=[
-                "# 📚 Understand the data",
-                "# 🗄 Create the dataset",
-                "# 🧠 Train the model",
-                "## 💻 Local predictions",
-                "## ☁️ Cloud Run predictions",
-                "## 🧺 Dataflow batch prediction",
-                "## 🌍 Earth Engine with AI Platform",
-            ],
-            variables={
+            sections={
+                "# 📚 Understand the data": {},
                 "# 🗄 Create the dataset": {
-                    "points_per_class": 1,
-                    "--data-path": f"gs://{bucket_name}/dataflow-data",
+                    "variables": {
+                        "points_per_class": 1,
+                        "--data-path": f"gs://{bucket_name}/dataflow-data",
+                    },
+                    "replace": {
+                        '--runner="DataflowRunner"': " ".join(
+                            [
+                                '--runner="DataflowRunner"',
+                                f"--job_name={unique_name}-dataset",
+                                "--max-requests=1",
+                            ]
+                        )
+                    },
                 },
                 "# 🧠 Train the model": {
-                    "display_name": unique_name,
-                    "data_path": data_path,
-                    "epochs": 1,
+                    "variables": {
+                        "display_name": unique_name,
+                        "data_path": data_path,
+                        "epochs": 1,
+                    },
                 },
                 "## 💻 Local predictions": {
-                    "model_path": model_path,
+                    "variables": {
+                        "model_path": model_path,
+                    },
                 },
                 "## ☁️ Cloud Run predictions": {
-                    "service_name": cloud_run_service_name,
-                    "model_path": model_path,
-                    "identity_token": identity_token,
+                    "variables": {
+                        "service_name": cloud_run_service_name,
+                        "model_path": model_path,
+                        "identity_token": identity_token,
+                    },
                 },
                 "## 🧺 Dataflow batch prediction": {
-                    "model_path": model_path,
+                    "variables": {
+                        "model_path": model_path,
+                    },
+                    "replace": {
+                        '--runner="DataflowRunner"': " ".join(
+                            [
+                                '--runner="DataflowRunner"',
+                                f"--job_name={unique_name}-predict",
+                                "--max-requests=1",
+                                f"--locations-file={locations_file.name}",
+                            ]
+                        )
+                    },
                 },
                 "## 🌍 Earth Engine with AI Platform": {
-                    "model_name": aiplatform_model_name,
-                    "model_path": model_path,
-                },
-            },
-            replace={
-                "# 🗄 Create the dataset": {
-                    '--runner="DataflowRunner"': " ".join(
-                        [
-                            '--runner="DataflowRunner"',
-                            f"--job_name={unique_name}-dataset",
-                            "--max-requests=1",
-                        ]
-                    )
-                },
-                "## 🧺 Dataflow batch prediction": {
-                    '--runner="DataflowRunner"': " ".join(
-                        [
-                            '--runner="DataflowRunner"',
-                            f"--job_name={unique_name}-predict",
-                            "--max-requests=1",
-                            f"--locations-file={locations_file.name}",
-                        ]
-                    )
+                    "variables": {
+                        "model_name": aiplatform_model_name,
+                        "model_path": model_path,
+                    },
                 },
             },
         )

people-and-planet-ai/land-cover-classification/e2e_test.py

@@ -22,7 +22,8 @@
 
 TEST_CONFIG_OVERRIDE = {
     # You can opt out from the test for specific Python versions.
-    # ℹ️ Testing only in Python 3.9 since it's the latest supported version for Dataflow.
+    # ℹ️ Test only in Python 3.9 since that's what Dataflow currently supports:
+    #   https://cloud.google.com/dataflow/docs/support/beam-runtime-support
     "ignored_versions": ["2.7", "3.6", "3.7", "3.8", "3.10"],
     # Old samples are opted out of enforcing Python type hints
     # All new samples should feature them
@@ -39,7 +40,5 @@
     "pip_version_override": None,
     # A dictionary you want to inject into your test. Don't put any
     # secrets here. These values will override predefined values.
-    "envs": {
-        "PYTEST_ADDOPTS": "-n=16",  # not CPU-bound, it's bound by Cloud requests.
-    },
+    "envs": {},
 }

people-and-planet-ai/land-cover-classification/noxfile_config.py

@@ -107,11 +107,11 @@ def run_tensorflow(
     import tensorflow as tf
 
     class LandCoverModel(ModelHandler[np.ndarray, np.ndarray, tf.keras.Model]):
-        def load_model(self) -> tf.keras.Model:  # noqa: ANN101
+        def load_model(self) -> tf.keras.Model:
             return tf.keras.models.load_model(model_path)
 
         def run_inference(
-            self,  # noqa: ANN101
+            self,
             batch: Sequence[np.ndarray],
             model: tf.keras.Model,
             inference_args: Optional[dict] = None,

people-and-planet-ai/land-cover-classification/predict_batch.py

@@ -1,7 +1,7 @@
 # Requirements to run tests.
-apache-beam[interactive]==2.41.0
+apache-beam[interactive]==2.42.0
 importnb==2022.10.24
-ipykernel==6.16.0
+ipykernel==6.17.1
 nbclient==0.7.0
-pytest-xdist==2.5.0
-pytest==7.1.2
+pytest-xdist==3.0.2
+pytest==7.2.0

people-and-planet-ai/land-cover-classification/requirements-test.txt

@@ -1,8 +1,8 @@
 # Requirements to run the notebooks.
-apache-beam[gcp]==2.41.0
-earthengine-api==0.1.325
+apache-beam[gcp]==2.42.0
+earthengine-api==0.1.331
 folium==0.13.0
-google-cloud-aiplatform==1.17.1
-imageio==2.22.0
+google-cloud-aiplatform==1.18.3
+imageio==2.22.4
 plotly==5.11.0
 tensorflow==2.10.0

people-and-planet-ai/land-cover-classification/requirements.txt

@@ -0,0 +1 @@
+3.10.x

people-and-planet-ai/land-cover-classification/serving/.python-version

@@ -1,5 +1,5 @@
 # Requirements for the prediction web service.
 Flask==2.2.2
-earthengine-api==0.1.325
+earthengine-api==0.1.331
 gunicorn==20.1.0
 tensorflow==2.10.0

people-and-planet-ai/land-cover-classification/serving/requirements.txt

@@ -20,8 +20,8 @@
     url="https://github.com/GoogleCloudPlatform/python-docs-samples/tree/main/people-and-planet-ai/land-cover-classification",
     packages=["serving"],
     install_requires=[
-        "apache-beam[gcp]==2.41.0",
-        "earthengine-api==0.1.325",
+        "apache-beam[gcp]==2.42.0",
+        "earthengine-api==0.1.331",
         "tensorflow==2.10.0",
     ],
 )