003-Running-inferences-with-extra-pipeline-steps¶

📥 Download 003-Running-inferences-with-extra-pipeline-steps and try it out

Introduction¶

This notebook is meant for someone with minimal knowledge, to be able to meaningfully use the most important functions of the Geospatial SDK.

For more information about the Geospatial Studio see the docs page: Geospatial Studio Docs

For more information about the Geospatial Studio SDK and all the functions available through it, see the SDK docs page: Geospatial Studio SDK Docs

In [1]:

# Install extra requirements
! pip install boto3

# Install extra requirements ! pip install boto3

Requirement already satisfied: boto3 in /Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/.toolkit/lib/python3.12/site-packages (1.40.21)
Requirement already satisfied: botocore<1.41.0,>=1.40.21 in /Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/.toolkit/lib/python3.12/site-packages (from boto3) (1.40.40)
Requirement already satisfied: jmespath<2.0.0,>=0.7.1 in /Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/.toolkit/lib/python3.12/site-packages (from boto3) (1.0.1)
Requirement already satisfied: s3transfer<0.14.0,>=0.13.0 in /Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/.toolkit/lib/python3.12/site-packages (from boto3) (0.13.1)
Requirement already satisfied: python-dateutil<3.0.0,>=2.1 in /Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/.toolkit/lib/python3.12/site-packages (from botocore<1.41.0,>=1.40.21->boto3) (2.9.0.post0)
Requirement already satisfied: urllib3!=2.2.0,<3,>=1.25.4 in /Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/.toolkit/lib/python3.12/site-packages (from botocore<1.41.0,>=1.40.21->boto3) (2.5.0)
Requirement already satisfied: six>=1.5 in /Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/.toolkit/lib/python3.12/site-packages (from python-dateutil<3.0.0,>=2.1->botocore<1.41.0,>=1.40.21->boto3) (1.17.0)

[notice] A new release of pip is available: 24.3.1 -> 25.3
[notice] To update, run: pip install --upgrade pip

In [2]:

%load_ext autoreload
%autoreload 2

%load_ext autoreload %autoreload 2

In [2]:

# import the required packages

import urllib3
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)

from geostudio import Client
from geostudio import gswidgets

# import the required packages import urllib3 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning) from geostudio import Client from geostudio import gswidgets

Connecting to the platform¶

First, we set up the connection to the platform backend. To do this we need the base url for the studio UI and an API key.

To get an API Key:

1. Go to the Geospatial Studio UI page and navigate to the Manage your API keys link.
2. This should pop-up a window where you can generate, access and delete your api keys. NB: every user is limited to a maximum of two activate api keys at any one time.

Store the API key and geostudio ui base url in a credentials file locally, for example in /User/bob/.geostudio_config_file. You can do this by:

echo "GEOSTUDIO_API_KEY=<paste_api_key_here>" > .geostudio_config_file
echo "BASE_STUDIO_UI_URL=<paste_ui_base_url_here>" >> .geostudio_config_file

Copy and paste the file path to this credentials file in call below.

In [3]:

#############################################################
# Initialize Geostudio client using a geostudio config file
#############################################################
gfm_client = Client(geostudio_config_file=".geostudio_config_file")

############################################################# # Initialize Geostudio client using a geostudio config file ############################################################# gfm_client = Client(geostudio_config_file=".geostudio_config_file")

Using api key and base urls from geostudio config file
Using api key and base urls from geostudio config file
Using api key and base urls from geostudio config file

Running inference¶

This notebook is used to run inference with extra custom post-processing steps to be applied after model prediction.

In [5]:

# Replace with your tune_id

tune_id = "geotune-4kyeuubgnuk8kjdlnjdrsl"

# Replace with your tune_id tune_id = "geotune-4kyeuubgnuk8kjdlnjdrsl"

In [6]:

# define payload
request_prebuilt_steps_payload = {
    "description": "prebuilt steps inference",
    "location": "austin",
    "fine_tuning_id": tune_id,  # replace this with the ID of your tune
    "model_display_name": "geofm-sandbox-models",  # Keep as is
    "spatial_domain": {  # provide downloadable link to the file(s) to run inference on.
        "urls": [
            "https://ibm.box.com/shared/static/kpvuc75q0j29rvf3w2qiamgczuvuymmd.tif"
        ]
    },
    "temporal_domain": ["2025-11-24"],
    # Provide this
    "geoserver_push": [
        # Keep these 2 layers as is.
        {
            "z_index": 0,
            "workspace": "geofm",
            "layer_name": "input_rgb",
            "file_suffix": "",
            "display_name": "Input image (RGB)",
            "filepath_key": "model_input_original_image_rgb",
            "geoserver_style": {
                "rgb": [
                    {
                        "label": "RedChannel",
                        "channel": 1,
                        "maxValue": 255,
                        "minValue": 0,
                    },
                    {
                        "label": "GreenChannel",
                        "channel": 2,
                        "maxValue": 255,
                        "minValue": 0,
                    },
                    {
                        "label": "BlueChannel",
                        "channel": 3,
                        "maxValue": 255,
                        "minValue": 0,
                    },
                ]
            },
            "visible_by_default": "True",
        },
        {
            "z_index": 1,
            "workspace": "geofm",
            "layer_name": "pred",
            "file_suffix": "",
            "display_name": "Model prediction",
            "filepath_key": "model_output_image",
            "geoserver_style": {
                "segmentation": [
                    {
                        "color": "#7d7247",
                        "label": "no-buildings",
                        "opacity": 0,
                        "quantity": "0",
                    },
                    {
                        "color": "#390c8c",
                        "label": "buildings",
                        "opacity": 1,
                        "quantity": "1",
                    },
                ]
            },
            "visible_by_default": "True",
        },
        # Add all layers to push created by your step(s).
        {
            "z_index": 1,
            "workspace": "geofm",
            "layer_name": "im2poly",  # Name of the registered step
            "file_suffix": "gpkg",  # Suffix of the generated file
            "display_name": "im2poly_regularize_image",  # Display name of the Step
            "filepath_key": "model_output_im2poly_regularize_image",  # Should be <model_output_{step_registered_name}_image>
            "geoserver_style": {},  # for shape files, keep this empty
            "visible_by_default": "True",
        },
    ],
    "post_processing": {
        "cloud_masking": "False",
        "regularization": "False",
        # Defining the extra post-processing steps
        "regularization_custom": [
            {
                "name": "im2poly_regularize",  # step_registered_name
                "params": {  # params of the function
                    "geoserver_suffix_extension": "gpkg",  # Expected
                    "params": {
                        "simplify_tolerance": 2,
                        "parallel_threshold": 2.0,
                        "allow_45_degree": "True",
                    },
                },
            }
        ],
    },
}

# define payload request_prebuilt_steps_payload = { "description": "prebuilt steps inference", "location": "austin", "fine_tuning_id": tune_id, # replace this with the ID of your tune "model_display_name": "geofm-sandbox-models", # Keep as is "spatial_domain": { # provide downloadable link to the file(s) to run inference on. "urls": [ "https://ibm.box.com/shared/static/kpvuc75q0j29rvf3w2qiamgczuvuymmd.tif" ] }, "temporal_domain": ["2025-11-24"], # Provide this "geoserver_push": [ # Keep these 2 layers as is. { "z_index": 0, "workspace": "geofm", "layer_name": "input_rgb", "file_suffix": "", "display_name": "Input image (RGB)", "filepath_key": "model_input_original_image_rgb", "geoserver_style": { "rgb": [ { "label": "RedChannel", "channel": 1, "maxValue": 255, "minValue": 0, }, { "label": "GreenChannel", "channel": 2, "maxValue": 255, "minValue": 0, }, { "label": "BlueChannel", "channel": 3, "maxValue": 255, "minValue": 0, }, ] }, "visible_by_default": "True", }, { "z_index": 1, "workspace": "geofm", "layer_name": "pred", "file_suffix": "", "display_name": "Model prediction", "filepath_key": "model_output_image", "geoserver_style": { "segmentation": [ { "color": "#7d7247", "label": "no-buildings", "opacity": 0, "quantity": "0", }, { "color": "#390c8c", "label": "buildings", "opacity": 1, "quantity": "1", }, ] }, "visible_by_default": "True", }, # Add all layers to push created by your step(s). { "z_index": 1, "workspace": "geofm", "layer_name": "im2poly", # Name of the registered step "file_suffix": "gpkg", # Suffix of the generated file "display_name": "im2poly_regularize_image", # Display name of the Step "filepath_key": "model_output_im2poly_regularize_image", # Should be "geoserver_style": {}, # for shape files, keep this empty "visible_by_default": "True", }, ], "post_processing": { "cloud_masking": "False", "regularization": "False", # Defining the extra post-processing steps "regularization_custom": [ { "name": "im2poly_regularize", # step_registered_name "params": { # params of the function "geoserver_suffix_extension": "gpkg", # Expected "params": { "simplify_tolerance": 2, "parallel_threshold": 2.0, "allow_45_degree": "True", }, }, } ], }, }

In [7]:

inference_response = gfm_client.try_out_tune(tune_id=tune_id, data=request_prebuilt_steps_payload)
inference_response

inference_response = gfm_client.try_out_tune(tune_id=tune_id, data=request_prebuilt_steps_payload) inference_response

Out[7]:

{'spatial_domain': {'bbox': [],
  'polygons': [],
  'tiles': [],
  'urls': ['https://ibm.box.com/shared/static/kpvuc75q0j29rvf3w2qiamgczuvuymmd.tif']},
 'temporal_domain': ['2025-11-24'],
 'fine_tuning_id': 'geotune-4kyeuubgnuk8kjdlnjdrsl',
 'maxcc': 100,
 'model_display_name': 'geofm-sandbox-models',
 'description': 'prebuilt steps inference',
 'location': 'austin',
 'geoserver_layers': None,
 'demo': None,
 'model_id': '6eeed629-5206-4c1a-8188-58ad754fd235',
 'inference_output': None,
 'id': '84e8b661-d568-4314-8b31-1827352215c0',
 'active': True,
 'created_by': 'Catherine.Wanjiru@ibm.com',
 'created_at': '2025-12-04T09:29:17.137150Z',
 'updated_at': '2025-12-04T09:29:17.457513Z',
 'status': 'PENDING',
 'tasks_count_total': 1,
 'tasks_count_success': 0,
 'tasks_count_failed': 0,
 'tasks_count_stopped': 0,
 'tasks_count_waiting': 1}

In [8]:

# define payload
request_download_scripts_payload = {
    "description": "download scripts",
    "location": "austin",
    "fine_tuning_id": tune_id,  # replace this with the ID of your tune
    "model_display_name": "geofm-sandbox-models",  # Keep as is
    "spatial_domain": {  # provide downloadable link to the file(s) to run inference on.
        "urls": [
            "https://ibm.box.com/shared/static/kpvuc75q0j29rvf3w2qiamgczuvuymmd.tif"
        ]
    },
    "temporal_domain": ["2025-11-24"],
    # Provide this
    "geoserver_push": [
        # Keep these 2 layers as is.
        {
            "z_index": 0,
            "workspace": "geofm",
            "layer_name": "input_rgb",
            "file_suffix": "",
            "display_name": "Input image (RGB)",
            "filepath_key": "model_input_original_image_rgb",
            "geoserver_style": {
                "rgb": [
                    {
                        "label": "RedChannel",
                        "channel": 1,
                        "maxValue": 255,
                        "minValue": 0,
                    },
                    {
                        "label": "GreenChannel",
                        "channel": 2,
                        "maxValue": 255,
                        "minValue": 0,
                    },
                    {
                        "label": "BlueChannel",
                        "channel": 3,
                        "maxValue": 255,
                        "minValue": 0,
                    },
                ]
            },
            "visible_by_default": "True",
        },
        {
            "z_index": 1,
            "workspace": "geofm",
            "layer_name": "pred",
            "file_suffix": "",
            "display_name": "Model prediction",
            "filepath_key": "model_output_image",
            "geoserver_style": {
                "segmentation": [
                    {
                        "color": "#7d7247",
                        "label": "no-buildings",
                        "opacity": 0,
                        "quantity": "0",
                    },
                    {
                        "color": "#390c8c",
                        "label": "buildings",
                        "opacity": 1,
                        "quantity": "1",
                    },
                ]
            },
            "visible_by_default": "True",
        },
        # Add all layers to push created by your step(s).
        {
            "z_index": 1,
            "workspace": "geofm",
            "layer_name": "im2poly",  # Name of the registered step
            "file_suffix": "gpkg",  # Suffix of the generated file
            "display_name": "im2poly_regularize_image",  # Display name of the Step
            "filepath_key": "model_output_im2poly_regularize_image",  # Should be <model_output_{step_registered_name}_image>
            "geoserver_style": {},  # for shape files, keep this empty
            "visible_by_default": "True",
        },
    ],
    "post_processing": {
        "cloud_masking": "False",
        "regularization": "False",
        "download_scripts": {
            "activated": "True",
            "verify_ssl": "False",
            "plugins_list": [
                {
                    "url": "https://s3.us-east.cloud-object-storage.appdomain.cloud/geospatial-studio-example-data/example_post_processing_scripts/user_masking_testing.py?response-content-type=application%2Foctet-stream&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=1c6317aac0054b9890797f09f217b54e%2F20251203%2Fus-east%2Fs3%2Faws4_request&X-Amz-Date=20251203T103901Z&X-Amz-Expires=259200&X-Amz-SignedHeaders=host&X-Amz-Signature=ca85cef43562cf6bc88d20a8f7463a311a0cdde8d53288ef552acc38b86a3849",
                    "filename": "user_masking_testing.py",
                },
                {
                    "url": "https://s3.us-east.cloud-object-storage.appdomain.cloud/geospatial-studio-example-data/example_post_processing_scripts/cloud_masking_testing.py?response-content-type=application%2Foctet-stream&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=1c6317aac0054b9890797f09f217b54e%2F20251203%2Fus-east%2Fs3%2Faws4_request&X-Amz-Date=20251203T104802Z&X-Amz-Expires=259200&X-Amz-SignedHeaders=host&X-Amz-Signature=4f702cb5424c1a8438acc5a0865142e95c58240ca9d122cc322016d1878ba1b8",
                    "filename": "cloud_masking_testing.py",
                },
            ],
        },
        # Defining the extra post-processing steps
        "regularization_custom": [
            {
                "name": "im2poly_regularize",  # step_registered_name
                "params": {  # params of the function
                    "geoserver_suffix_extension": "gpkg",  # Expected
                    "params": {
                        "simplify_tolerance": 2,
                        "parallel_threshold": 2.0,
                        "allow_45_degree": "True",
                    },
                },
            }
        ],
    },
}

# define payload request_download_scripts_payload = { "description": "download scripts", "location": "austin", "fine_tuning_id": tune_id, # replace this with the ID of your tune "model_display_name": "geofm-sandbox-models", # Keep as is "spatial_domain": { # provide downloadable link to the file(s) to run inference on. "urls": [ "https://ibm.box.com/shared/static/kpvuc75q0j29rvf3w2qiamgczuvuymmd.tif" ] }, "temporal_domain": ["2025-11-24"], # Provide this "geoserver_push": [ # Keep these 2 layers as is. { "z_index": 0, "workspace": "geofm", "layer_name": "input_rgb", "file_suffix": "", "display_name": "Input image (RGB)", "filepath_key": "model_input_original_image_rgb", "geoserver_style": { "rgb": [ { "label": "RedChannel", "channel": 1, "maxValue": 255, "minValue": 0, }, { "label": "GreenChannel", "channel": 2, "maxValue": 255, "minValue": 0, }, { "label": "BlueChannel", "channel": 3, "maxValue": 255, "minValue": 0, }, ] }, "visible_by_default": "True", }, { "z_index": 1, "workspace": "geofm", "layer_name": "pred", "file_suffix": "", "display_name": "Model prediction", "filepath_key": "model_output_image", "geoserver_style": { "segmentation": [ { "color": "#7d7247", "label": "no-buildings", "opacity": 0, "quantity": "0", }, { "color": "#390c8c", "label": "buildings", "opacity": 1, "quantity": "1", }, ] }, "visible_by_default": "True", }, # Add all layers to push created by your step(s). { "z_index": 1, "workspace": "geofm", "layer_name": "im2poly", # Name of the registered step "file_suffix": "gpkg", # Suffix of the generated file "display_name": "im2poly_regularize_image", # Display name of the Step "filepath_key": "model_output_im2poly_regularize_image", # Should be "geoserver_style": {}, # for shape files, keep this empty "visible_by_default": "True", }, ], "post_processing": { "cloud_masking": "False", "regularization": "False", "download_scripts": { "activated": "True", "verify_ssl": "False", "plugins_list": [ { "url": "https://s3.us-east.cloud-object-storage.appdomain.cloud/geospatial-studio-example-data/example_post_processing_scripts/user_masking_testing.py?response-content-type=application%2Foctet-stream&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=1c6317aac0054b9890797f09f217b54e%2F20251203%2Fus-east%2Fs3%2Faws4_request&X-Amz-Date=20251203T103901Z&X-Amz-Expires=259200&X-Amz-SignedHeaders=host&X-Amz-Signature=ca85cef43562cf6bc88d20a8f7463a311a0cdde8d53288ef552acc38b86a3849", "filename": "user_masking_testing.py", }, { "url": "https://s3.us-east.cloud-object-storage.appdomain.cloud/geospatial-studio-example-data/example_post_processing_scripts/cloud_masking_testing.py?response-content-type=application%2Foctet-stream&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=1c6317aac0054b9890797f09f217b54e%2F20251203%2Fus-east%2Fs3%2Faws4_request&X-Amz-Date=20251203T104802Z&X-Amz-Expires=259200&X-Amz-SignedHeaders=host&X-Amz-Signature=4f702cb5424c1a8438acc5a0865142e95c58240ca9d122cc322016d1878ba1b8", "filename": "cloud_masking_testing.py", }, ], }, # Defining the extra post-processing steps "regularization_custom": [ { "name": "im2poly_regularize", # step_registered_name "params": { # params of the function "geoserver_suffix_extension": "gpkg", # Expected "params": { "simplify_tolerance": 2, "parallel_threshold": 2.0, "allow_45_degree": "True", }, }, } ], }, }

In [9]:

inference_response = gfm_client.try_out_tune(tune_id=tune_id, data=request_download_scripts_payload)
inference_response

inference_response = gfm_client.try_out_tune(tune_id=tune_id, data=request_download_scripts_payload) inference_response

Out[9]:

{'spatial_domain': {'bbox': [],
  'polygons': [],
  'tiles': [],
  'urls': ['https://ibm.box.com/shared/static/kpvuc75q0j29rvf3w2qiamgczuvuymmd.tif']},
 'temporal_domain': ['2025-11-24'],
 'fine_tuning_id': 'geotune-4kyeuubgnuk8kjdlnjdrsl',
 'maxcc': 100,
 'model_display_name': 'geofm-sandbox-models',
 'description': 'download scripts',
 'location': 'austin',
 'geoserver_layers': None,
 'demo': None,
 'model_id': '6eeed629-5206-4c1a-8188-58ad754fd235',
 'inference_output': None,
 'id': 'ee82608d-39cf-4101-b39f-653579b1b314',
 'active': True,
 'created_by': 'Catherine.Wanjiru@ibm.com',
 'created_at': '2025-12-04T09:29:19.158507Z',
 'updated_at': '2025-12-04T09:29:19.495407Z',
 'status': 'PENDING',
 'tasks_count_total': 1,
 'tasks_count_success': 0,
 'tasks_count_failed': 0,
 'tasks_count_stopped': 0,
 'tasks_count_waiting': 1}

Monitor inference status and progress¶

After submitting the request, we can poll the inference service to check the progress as well as get the output details once its complete (this could take a few minutes depending on the request size and the current service load).

In [10]:

# Poll inference status
r = gfm_client.poll_inference_until_finished(inference_id=inference_response['id'])

# Poll inference status r = gfm_client.poll_inference_until_finished(inference_id=inference_response['id'])

COMPLETED - 289 seconds

In [11]:

response = gfm_client.get_inference(inference_id=inference_response['id'])
response

response = gfm_client.get_inference(inference_id=inference_response['id']) response

Out[11]:

{'spatial_domain': {'bbox': [],
  'polygons': [],
  'tiles': [],
  'urls': ['https://ibm.box.com/shared/static/kpvuc75q0j29rvf3w2qiamgczuvuymmd.tif']},
 'temporal_domain': ['2025-11-24'],
 'fine_tuning_id': 'geotune-4kyeuubgnuk8kjdlnjdrsl',
 'maxcc': 100,
 'model_display_name': 'geofm-sandbox-models',
 'description': 'download scripts',
 'location': 'austin',
 'geoserver_layers': {'predicted_layers': [{'uri': 'geofm:ee82608d-39cf-4101-b39f-653579b1b314-input_rgb',
    'display_name': 'Input image (RGB)',
    'sld_body': '<?xml version="1.0" encoding="UTF-8"?>\n    <StyledLayerDescriptor xmlns="http://www.opengis.net/sld" xmlns:ogc="http://www.opengis.net/ogc" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.opengis.net/sld\n    http://schemas.opengis.net/sld/1.0.0/StyledLayerDescriptor.xsd" version="1.0.0">\n        <NamedLayer>\n            <Name>geofm:ee82608d-39cf-4101-b39f-653579b1b314-input_rgb</Name>\n            <UserStyle>\n                <Title>geofm:ee82608d-39cf-4101-b39f-653579b1b314-input_rgb</Title>\n                <FeatureTypeStyle>\n                    <Rule>\n                        <RasterSymbolizer>\n                            <ChannelSelection>\n                                <RedChannel>\n        <SourceChannelName>1</SourceChannelName>\n        <ContrastEnhancement>\n        <Normalize>\n            <VendorOption name="algorithm">StretchToMinimumMaximum</VendorOption>\n            <VendorOption name="minValue">0</VendorOption>\n            <VendorOption name="maxValue">255</VendorOption>\n        </Normalize>\n        </ContrastEnhancement>\n    </RedChannel><GreenChannel>\n        <SourceChannelName>2</SourceChannelName>\n        <ContrastEnhancement>\n        <Normalize>\n            <VendorOption name="algorithm">StretchToMinimumMaximum</VendorOption>\n            <VendorOption name="minValue">0</VendorOption>\n            <VendorOption name="maxValue">255</VendorOption>\n        </Normalize>\n        </ContrastEnhancement>\n    </GreenChannel><BlueChannel>\n        <SourceChannelName>3</SourceChannelName>\n        <ContrastEnhancement>\n        <Normalize>\n            <VendorOption name="algorithm">StretchToMinimumMaximum</VendorOption>\n            <VendorOption name="minValue">0</VendorOption>\n            <VendorOption name="maxValue">255</VendorOption>\n        </Normalize>\n        </ContrastEnhancement>\n    </BlueChannel>            \n                            </ChannelSelection>\n                        </RasterSymbolizer>\n                    </Rule>\n                </FeatureTypeStyle>\n            </UserStyle>\n        </NamedLayer>\n    </StyledLayerDescriptor>\n    ',
    'z_index': 0,
    'visible_by_default': 'True'},
   {'uri': 'geofm:ee82608d-39cf-4101-b39f-653579b1b314-pred',
    'display_name': 'Model prediction',
    'sld_body': '<?xml version="1.0" encoding="UTF-8"?>\n    <StyledLayerDescriptor xmlns="http://www.opengis.net/sld" xmlns:ogc="http://www.opengis.net/ogc" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.opengis.net/sld\n    http://schemas.opengis.net/sld/1.0.0/StyledLayerDescriptor.xsd" version="1.0.0">\n    <NamedLayer>\n        <Name>geofm:ee82608d-39cf-4101-b39f-653579b1b314-pred</Name>\n        <UserStyle>\n        <Title>geofm:ee82608d-39cf-4101-b39f-653579b1b314-pred</Title>\n        <IsDefault>1</IsDefault>\n        <FeatureTypeStyle>\n            <Rule>\n            <RasterSymbolizer>\n                <Opacity>1.0</Opacity>\n                <ColorMap>\n                    <ColorMapEntry color="#7d7247"  opacity="0" quantity="0" label="no-buildings" /><ColorMapEntry color="#390c8c"  opacity="1" quantity="1" label="buildings" />\n                </ColorMap>\n            </RasterSymbolizer>\n            </Rule>\n        </FeatureTypeStyle>\n        </UserStyle>\n    </NamedLayer>\n    </StyledLayerDescriptor>',
    'z_index': 1,
    'visible_by_default': 'True'},
   {'uri': 'geofm:ee82608d-39cf-4101-b39f-653579b1b314-im2poly',
    'display_name': 'im2poly_regularize_image',
    'sld_body': None,
    'z_index': 1,
    'visible_by_default': 'True'}],
  'bbox_pred': [[16.302191847671576,
    48.224583386883076,
    16.322389984936816,
    48.2380774323371],
   [16.302188413198312,
    48.224583386883076,
    16.32239341941008,
    48.2380774323371],
   [16.30217764259808,
    48.22456748526141,
    16.32240522602398,
    48.23809283932311]]},
 'demo': None,
 'model_id': '6eeed629-5206-4c1a-8188-58ad754fd235',
 'inference_output': {'output_url': 'None'},
 'id': 'ee82608d-39cf-4101-b39f-653579b1b314',
 'active': True,
 'created_by': 'Catherine.Wanjiru@ibm.com',
 'created_at': '2025-12-04T09:29:19.158507Z',
 'updated_at': '2025-12-04T09:34:08.441348Z',
 'status': 'COMPLETED',
 'tasks_count_total': 5,
 'tasks_count_success': 5,
 'tasks_count_failed': 0,
 'tasks_count_stopped': 0,
 'tasks_count_waiting': 0}

You can view the finished inference in the UI now

Accessing inference outputs¶

Once an inference run is completed, the inputs and outputs of each task within an inference are packaged up into a zip file which is uploaded to a url you can use to download the files.

To access the inference task files:

1. Get the inference tasks list
2. Identify the specific inference task you want to view
3. Download task output files

In [12]:

# Get the inference tasks list
inf_tasks_res = gfm_client.get_inference_tasks(response["id"])
inf_tasks_res

# Get the inference tasks list inf_tasks_res = gfm_client.get_inference_tasks(response["id"]) inf_tasks_res

Out[12]:

{'inference_id': 'ee82608d-39cf-4101-b39f-653579b1b314',
 'status': 'COMPLETED',
 'tasks': [{'inference_id': 'ee82608d-39cf-4101-b39f-653579b1b314',
   'task_id': 'ee82608d-39cf-4101-b39f-653579b1b314-task_0',
   'inference_folder': '/data/ee82608d-39cf-4101-b39f-653579b1b314',
   'status': 'FINISHED',
   'pipeline_steps': [{'status': 'FINISHED',
     'end_time': '2025-12-04T09:31:18',
     'process_id': 'url-connector',
     'start_time': '2025-12-04T09:30:31',
     'step_number': 0},
    {'status': 'FINISHED',
     'end_time': '2025-12-04T09:32:17',
     'process_id': 'terratorch-inference',
     'start_time': '2025-12-04T09:31:31',
     'step_number': 1},
    {'status': 'FINISHED',
     'end_time': '2025-12-04T09:32:55',
     'process_id': 'postprocess-generic',
     'start_time': '2025-12-04T09:32:20',
     'step_number': 2},
    {'status': 'FINISHED',
     'end_time': '2025-12-04T09:34:07',
     'process_id': 'push-to-geoserver',
     'start_time': '2025-12-04T09:33:00',
     'step_number': 3}]},
  {'inference_id': 'ee82608d-39cf-4101-b39f-653579b1b314',
   'task_id': 'ee82608d-39cf-4101-b39f-653579b1b314-task_planning',
   'inference_folder': '/data/ee82608d-39cf-4101-b39f-653579b1b314',
   'status': 'FINISHED',
   'pipeline_steps': [{'status': 'FINISHED',
     'end_time': '2025-12-04T09:29:48',
     'process_id': 'inference-planner',
     'start_time': '2025-12-04T09:29:41',
     'step_number': 0}]}]}

In [13]:

df = gfm_client.inference_task_status_df(response["id"])


display(df.style.map(gswidgets.color_inference_tasks_by_status))

df = gfm_client.inference_task_status_df(response["id"]) display(df.style.map(gswidgets.color_inference_tasks_by_status))

0

/Users/catherinewanjiru/projects/climate/geospatial-studio-toolkit/geospatial-studio-sdk/geostudio/backends/v2/ginference/client.py:262: FutureWarning:

ChainedAssignmentError: behaviour will change in pandas 3.0!
You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.
A typical example is when you are setting values in a column of a DataFrame, like:

df["col"][row_indexer] = value

Use `df.loc[row_indexer, "col"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

	task_id	terratorch-inference	url-connector	postprocess-generic	push-to-geoserver
0	ee82608d-39cf-4101-b39f-653579b1b314-task_0	FINISHED	FINISHED	FINISHED	FINISHED

In [14]:

gswidgets.view_inference_process_timeline(gfm_client, inference_id = response["id"])

gswidgets.view_inference_process_timeline(gfm_client, inference_id = response["id"])

Accessing inference outputs¶

Once an inference run is completed, the inputs and outputs of each task within an inference are packaged up into a zip file which is uploaded to a url you can use to download the files.

To access the inference task files:

1. Get the inference tasks list
2. Identify the specific inference task you want to view
3. Download task output files

In [15]:

# Get the inference tasks list

inf_tasks_res = gfm_client.get_inference_tasks(response["id"])
inf_tasks_res

# Get the inference tasks list inf_tasks_res = gfm_client.get_inference_tasks(response["id"]) inf_tasks_res

Out[15]:

{'inference_id': 'ee82608d-39cf-4101-b39f-653579b1b314',
 'status': 'COMPLETED',
 'tasks': [{'inference_id': 'ee82608d-39cf-4101-b39f-653579b1b314',
   'task_id': 'ee82608d-39cf-4101-b39f-653579b1b314-task_0',
   'inference_folder': '/data/ee82608d-39cf-4101-b39f-653579b1b314',
   'status': 'FINISHED',
   'pipeline_steps': [{'status': 'FINISHED',
     'end_time': '2025-12-04T09:31:18',
     'process_id': 'url-connector',
     'start_time': '2025-12-04T09:30:31',
     'step_number': 0},
    {'status': 'FINISHED',
     'end_time': '2025-12-04T09:32:17',
     'process_id': 'terratorch-inference',
     'start_time': '2025-12-04T09:31:31',
     'step_number': 1},
    {'status': 'FINISHED',
     'end_time': '2025-12-04T09:32:55',
     'process_id': 'postprocess-generic',
     'start_time': '2025-12-04T09:32:20',
     'step_number': 2},
    {'status': 'FINISHED',
     'end_time': '2025-12-04T09:34:07',
     'process_id': 'push-to-geoserver',
     'start_time': '2025-12-04T09:33:00',
     'step_number': 3}]},
  {'inference_id': 'ee82608d-39cf-4101-b39f-653579b1b314',
   'task_id': 'ee82608d-39cf-4101-b39f-653579b1b314-task_planning',
   'inference_folder': '/data/ee82608d-39cf-4101-b39f-653579b1b314',
   'status': 'FINISHED',
   'pipeline_steps': [{'status': 'FINISHED',
     'end_time': '2025-12-04T09:29:48',
     'process_id': 'inference-planner',
     'start_time': '2025-12-04T09:29:41',
     'step_number': 0}]}]}

Next, Identify the task you want to view from the response above, ensure status of the task is FINISHED and set selected_task variable below to the task number at the end of the task id string. For example, if task_id is "6d1149fa-302d-4612-82dd-5879fc06081d-task_0", selected_task would be 0

In [16]:

# Select a task to view

selected_task = 0 
selected_task_id = f"{inf_tasks_res['inference_id']}-task_{selected_task}"

# Select a task to view selected_task = 0 selected_task_id = f"{inf_tasks_res['inference_id']}-task_{selected_task}"

In [17]:

selected_task_id

selected_task_id

Out[17]:

'ee82608d-39cf-4101-b39f-653579b1b314-task_0'

In [ ]:

# Download task output files

gswidgets.fileDownloaderTasks(client=gfm_client, task_id=selected_task_id,just_tifs=False)

# Download task output files gswidgets.fileDownloaderTasks(client=gfm_client, task_id=selected_task_id,just_tifs=False)

Just tiffs: False

HTML(value='<h1>Inference Task output downloader</h1> </p>Select the files and the download path and hit downl…

SelectMultiple(description='Files:', layout=Layout(width='1000px'), options=('ee82608d-39cf-4101-b39f-653579b1…

Text(value='./', description='Dl path:')

Button(description='Download', icon='check', layout=Layout(height='auto', width='800px'), style=ButtonStyle(),…

Output()