abstract CDN URL

Author: holmbergius

.claude/settings.local.json

@@ -0,0 +1,10 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(du -sh:*)",
+      "Bash(du:*)"
+    ],
+    "deny": [],
+    "ask": []
+  }
+}

docs/onnx.rst

@@ -7,6 +7,17 @@ All of the machine learning models are hosted on GitHub as LFS files.  The two m
 however need those files downloaded to the local machine prior to running inference.  These models are
 hosted on a separate CDN for convenient access and can be fetched by running the following functions:
 
+.. note::
+
+   The model download URL can be configured via the ``SCOUTBOT_MODEL_URL`` environment variable.
+   This allows organizations to host models on their own infrastructure. For example:
+
+   .. code-block:: bash
+
+      export SCOUTBOT_MODEL_URL="https://your-cdn.example.com/models"
+
+   Similarly, test data URLs can be configured via ``SCOUTBOT_DATA_URL``.
+
    - :meth:`scoutbot.wic.fetch`
    - :meth:`scoutbot.loc.fetch`
 

scoutbot/__init__.py

@@ -49,6 +49,7 @@
         # nms_thresh=agg_nms_thresh,  # Optional override of config
     )
 '''
+import os
 import cv2
 from os.path import exists
 
@@ -60,6 +61,14 @@
 log = utils.init_logging()
 QUIET = not utils.VERBOSE
 
+# Base URLs for downloading models and data (can be overridden via environment variables)
+MODEL_BASE_URL = os.getenv(
+    'SCOUTBOT_MODEL_URL', 'https://wildbookiarepository.azureedge.net/models'
+).rstrip('/')
+DATA_BASE_URL = os.getenv(
+    'SCOUTBOT_DATA_URL', 'https://wildbookiarepository.azureedge.net/data'
+).rstrip('/')
+
 
 from scoutbot import agg, loc, tile, wic, tile_batched  # NOQA
 from scoutbot.loc import CONFIGS as LOC_CONFIGS # NOQA
@@ -491,7 +500,7 @@ def example():
     )
 
     img_filepath = pooch.retrieve(
-        url=f'https://wildbookiarepository.azureedge.net/data/{TEST_IMAGE}',
+        url=f'{DATA_BASE_URL}/{TEST_IMAGE}',
         known_hash=TEST_IMAGE_HASH,
         progressbar=True,
     )

scoutbot/loc/__init__.py

@@ -21,7 +21,7 @@
 import tqdm
 import utool as ut
 
-from scoutbot import QUIET, log
+from scoutbot import MODEL_BASE_URL, QUIET, log
 from scoutbot.loc.transforms import (
     Compose,
     GetBoundingBoxes,
@@ -172,7 +172,7 @@ def fetch(pull=False, config=DEFAULT_CONFIG):
         onnx_model = model_path
     else:
         onnx_model = pooch.retrieve(
-            url=f'https://wildbookiarepository.azureedge.net/models/{model_name}',
+            url=f'{MODEL_BASE_URL}/{model_name}',
             known_hash=model_hash,
             progressbar=not QUIET,
         )

scoutbot/wic/__init__.py

@@ -18,7 +18,7 @@
 import torch
 import tqdm
 
-from scoutbot import QUIET, log
+from scoutbot import MODEL_BASE_URL, QUIET, log
 from scoutbot.wic.dataloader import (  # NOQA
     BATCH_SIZE,
     INPUT_SIZE,
@@ -82,7 +82,7 @@ def fetch(pull=False, config=DEFAULT_CONFIG):
         onnx_model = model_path
     else:
         onnx_model = pooch.retrieve(
-            url=f'https://wildbookiarepository.azureedge.net/models/{model_name}',
+            url=f'{MODEL_BASE_URL}/{model_name}',
             known_hash=model_hash,
             progressbar=not QUIET,
         )