Update documentation to add supported species, and update the two Gra…

…dio apps
WildMeOrg · Oct 4, 2022 · 471503c · 471503c
1 parent 6af0f64
commit 471503c
Show file tree

Hide file tree

Showing 10 changed files with 232 additions and 66 deletions.
diff --git a/README.rst b/README.rst
@@ -34,13 +34,13 @@ To then add GPU acceleration, you need to replace `onnxruntime` with `onnxruntim
 How to Run
 ----------
 
-You can run the tile-base Gradio demo with:
+You can run the tile-based Gradio demo with:
 
 .. code-block:: console
 
    (.venv) $ python app.py
 
-or, you can run the image-base Gradio demo with:
+or, you can run the image-based Gradio demo with:
 
 .. code-block:: console
 

diff --git a/app.py b/app.py
@@ -7,13 +7,26 @@
 
 from scoutbot import loc, wic
 
+PHASE1 = [
+    'Phase 1',
+    int(wic.CONFIGS['phase1']['thresh'] * 100),
+    int(loc.CONFIGS['phase1']['thresh'] * 100),
+    int(loc.CONFIGS['phase1']['nms'] * 100),
+]
+MVP = [
+    'MVP',
+    int(wic.CONFIGS['mvp']['thresh'] * 100),
+    int(loc.CONFIGS['mvp']['thresh'] * 100),
+    int(loc.CONFIGS['mvp']['nms'] * 100),
+]
+
 
 def predict(filepath, config, wic_thresh, loc_thresh, nms_thresh):
     start = time.time()
 
-    if config == 'MVP':
+    if config == MVP[0]:
         config = 'mvp'
-    elif config == 'Phase 1':
+    elif config == PHASE1[0]:
         config = 'phase1'
     else:
         raise ValueError()
@@ -22,8 +35,6 @@ def predict(filepath, config, wic_thresh, loc_thresh, nms_thresh):
     loc_thresh /= 100.0
     nms_thresh /= 100.0
 
-    nms_thresh = 1.0 - nms_thresh
-
     # Load data
     img = cv2.imread(filepath)
     img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
@@ -80,12 +91,12 @@ def predict(filepath, config, wic_thresh, loc_thresh, nms_thresh):
         gr.Radio(
             label='Model Configuration',
             type='value',
-            choices=['Phase 1', 'MVP'],
-            value='MVP',
+            choices=[PHASE1[0], MVP[0]],
+            value=MVP[0],
         ),
-        gr.Slider(label='WIC Confidence Threshold', value=7),
-        gr.Slider(label='Localizer Confidence Threshold', value=14),
-        gr.Slider(label='Localizer NMS Threshold', value=80),
+        gr.Slider(label='WIC Confidence Threshold', value=MVP[1]),
+        gr.Slider(label='Localizer Confidence Threshold', value=MVP[2]),
+        gr.Slider(label='Localizer NMS Threshold', value=MVP[3]),
     ],
     outputs=[
         gr.Image(type='numpy'),
@@ -94,16 +105,28 @@ def predict(filepath, config, wic_thresh, loc_thresh, nms_thresh):
         gr.Textbox(label='Predicted Localizer Detections', interactive=False),
     ],
     examples=[
-        ['examples/07a4b8db-f31c-261d-4580-e9402768fd45.true.jpg', 'MVP', 7, 14, 80],
-        ['examples/15e815d9-5aad-fa53-d1ed-33429020e15e.true.jpg', 'MVP', 7, 14, 80],
-        ['examples/1bb79811-3149-7a60-2d88-613dc3eeb261.true.jpg', 'MVP', 7, 14, 80],
-        ['examples/1e8372e4-357d-26e6-d7fd-0e0ae402463a.true.jpg', 'MVP', 7, 14, 80],
-        ['examples/201bc65e-d64e-80d3-2610-5865a22d04b4.false.jpg', 'MVP', 7, 14, 80],
-        ['examples/3affd8b6-9722-f2d5-9171-639615b4c38f.true.jpg', 'MVP', 7, 14, 80],
-        ['examples/4aedb818-f2f4-e462-8b75-5c8e34a01a59.false.jpg', 'MVP', 7, 14, 80],
-        ['examples/474bc2b6-dc51-c1b5-4612-efe810bbe091.true.jpg', 'MVP', 7, 14, 80],
-        ['examples/c3014107-3464-60b5-e04a-e4bfafdf8809.false.jpg', 'MVP', 7, 14, 80],
-        ['examples/f835ce33-292a-9116-794e-f8859b5956ec.true.jpg', 'MVP', 7, 14, 80],
+        # Phase 1
+        ['examples/07a4b8db-f31c-261d-4580-e9402768fd45.true.jpg'] + PHASE1,
+        ['examples/15e815d9-5aad-fa53-d1ed-33429020e15e.true.jpg'] + PHASE1,
+        ['examples/1bb79811-3149-7a60-2d88-613dc3eeb261.true.jpg'] + PHASE1,
+        ['examples/1e8372e4-357d-26e6-d7fd-0e0ae402463a.true.jpg'] + PHASE1,
+        ['examples/201bc65e-d64e-80d3-2610-5865a22d04b4.false.jpg'] + PHASE1,
+        ['examples/3affd8b6-9722-f2d5-9171-639615b4c38f.true.jpg'] + PHASE1,
+        ['examples/4aedb818-f2f4-e462-8b75-5c8e34a01a59.false.jpg'] + PHASE1,
+        ['examples/474bc2b6-dc51-c1b5-4612-efe810bbe091.true.jpg'] + PHASE1,
+        ['examples/c3014107-3464-60b5-e04a-e4bfafdf8809.false.jpg'] + PHASE1,
+        ['examples/f835ce33-292a-9116-794e-f8859b5956ec.true.jpg'] + PHASE1,
+        # MVP
+        ['examples/07a4b8db-f31c-261d-4580-e9402768fd45.true.jpg'] + MVP,
+        ['examples/15e815d9-5aad-fa53-d1ed-33429020e15e.true.jpg'] + MVP,
+        ['examples/1bb79811-3149-7a60-2d88-613dc3eeb261.true.jpg'] + MVP,
+        ['examples/1e8372e4-357d-26e6-d7fd-0e0ae402463a.true.jpg'] + MVP,
+        ['examples/201bc65e-d64e-80d3-2610-5865a22d04b4.false.jpg'] + MVP,
+        ['examples/3affd8b6-9722-f2d5-9171-639615b4c38f.true.jpg'] + MVP,
+        ['examples/4aedb818-f2f4-e462-8b75-5c8e34a01a59.false.jpg'] + MVP,
+        ['examples/474bc2b6-dc51-c1b5-4612-efe810bbe091.true.jpg'] + MVP,
+        ['examples/c3014107-3464-60b5-e04a-e4bfafdf8809.false.jpg'] + MVP,
+        ['examples/f835ce33-292a-9116-794e-f8859b5956ec.true.jpg'] + MVP,
     ],
     cache_examples=True,
     allow_flagging='never',

diff --git a/app2.py b/app2.py
@@ -6,10 +6,28 @@
 import numpy as np
 
 import scoutbot
+from scoutbot import agg, loc, wic
+
+PHASE1 = [
+    'Phase 1',
+    int(wic.CONFIGS['phase1']['thresh'] * 100),
+    int(loc.CONFIGS['phase1']['thresh'] * 100),
+    int(loc.CONFIGS['phase1']['nms'] * 100),
+    int(agg.CONFIGS['phase1']['thresh'] * 100),
+    int(agg.CONFIGS['phase1']['nms'] * 100),
+]
+MVP = [
+    'MVP',
+    int(wic.CONFIGS['mvp']['thresh'] * 100),
+    int(loc.CONFIGS['mvp']['thresh'] * 100),
+    int(loc.CONFIGS['mvp']['nms'] * 100),
+    int(agg.CONFIGS['mvp']['thresh'] * 100),
+    int(agg.CONFIGS['mvp']['nms'] * 100),
+]
 
 
 def predict(
-    filepath, config, wic_thresh, loc_thresh, agg_thresh, loc_nms_thresh, agg_nms_thresh
+    filepath, config, wic_thresh, loc_thresh, loc_nms_thresh, agg_thresh, agg_nms_thresh
 ):
     start = time.time()
 
@@ -26,9 +44,6 @@ def predict(
     agg_thresh /= 100.0
     agg_nms_thresh /= 100.0
 
-    loc_nms_thresh = 1.0 - loc_nms_thresh
-    agg_nms_thresh = 1.0 - agg_nms_thresh
-
     # Load data
     img = cv2.imread(filepath)
     img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
@@ -81,14 +96,14 @@ def predict(
         gr.Radio(
             label='Model Configuration',
             type='value',
-            choices=['Phase 1', 'MVP'],
-            value='MVP',
+            choices=[PHASE1[0], MVP[0]],
+            value=MVP[0],
         ),
-        gr.Slider(label='WIC Confidence Threshold', value=7),
-        gr.Slider(label='Localizer Confidence Threshold', value=14),
-        gr.Slider(label='Aggregation Confidence Threshold', value=51),
-        gr.Slider(label='Localizer NMS Threshold', value=80),
-        gr.Slider(label='Aggregation NMS Threshold', value=80),
+        gr.Slider(label='WIC Confidence Threshold', value=MVP[1]),
+        gr.Slider(label='Localizer Confidence Threshold', value=MVP[2]),
+        gr.Slider(label='Localizer NMS Threshold', value=MVP[3]),
+        gr.Slider(label='Aggregation Confidence Threshold', value=MVP[4]),
+        gr.Slider(label='Aggregation NMS Threshold', value=MVP[5]),
     ],
     outputs=[
         gr.Image(type='numpy'),
@@ -97,16 +112,28 @@ def predict(
         gr.Textbox(label='Predicted Detections', interactive=False),
     ],
     examples=[
-        ['examples/0d4e4df2-7b69-91b1-1985-c8421f2f3253.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/18cef191-74ed-2b5e-55a5-f58bd3d483ff.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/1be4d40a-6fd0-42ce-da6c-294e45781f41.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/1d3c85e9-ee24-f290-e7e1-6e338f2eaebb.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/3e043302-af1c-75a7-4057-3a2f25c123bf.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/43ecc08d-502a-7a51-9d68-3e40a76439a2.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/479058af-e774-e6aa-a2b0-9a42dd6ff8b1.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/7c910b87-ae3a-f580-d431-03cd89793803.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/8fa04489-cd94-7d8f-7e2e-5f0fe2f7ae76.jpg', 'MVP', 7, 14, 51, 80, 80],
-        ['examples/bb7b4345-b98a-c727-4c94-6090f0aa4355.jpg', 'MVP', 7, 14, 51, 80, 80],
+        # Phase 1
+        ['examples/0d4e4df2-7b69-91b1-1985-c8421f2f3253.jpg'] + PHASE1,
+        ['examples/18cef191-74ed-2b5e-55a5-f58bd3d483ff.jpg'] + PHASE1,
+        ['examples/1be4d40a-6fd0-42ce-da6c-294e45781f41.jpg'] + PHASE1,
+        ['examples/1d3c85e9-ee24-f290-e7e1-6e338f2eaebb.jpg'] + PHASE1,
+        ['examples/3e043302-af1c-75a7-4057-3a2f25c123bf.jpg'] + PHASE1,
+        ['examples/43ecc08d-502a-7a51-9d68-3e40a76439a2.jpg'] + PHASE1,
+        ['examples/479058af-e774-e6aa-a2b0-9a42dd6ff8b1.jpg'] + PHASE1,
+        ['examples/7c910b87-ae3a-f580-d431-03cd89793803.jpg'] + PHASE1,
+        ['examples/8fa04489-cd94-7d8f-7e2e-5f0fe2f7ae76.jpg'] + PHASE1,
+        ['examples/bb7b4345-b98a-c727-4c94-6090f0aa4355.jpg'] + PHASE1,
+        # MVP
+        ['examples/0d4e4df2-7b69-91b1-1985-c8421f2f3253.jpg'] + MVP,
+        ['examples/18cef191-74ed-2b5e-55a5-f58bd3d483ff.jpg'] + MVP,
+        ['examples/1be4d40a-6fd0-42ce-da6c-294e45781f41.jpg'] + MVP,
+        ['examples/1d3c85e9-ee24-f290-e7e1-6e338f2eaebb.jpg'] + MVP,
+        ['examples/3e043302-af1c-75a7-4057-3a2f25c123bf.jpg'] + MVP,
+        ['examples/43ecc08d-502a-7a51-9d68-3e40a76439a2.jpg'] + MVP,
+        ['examples/479058af-e774-e6aa-a2b0-9a42dd6ff8b1.jpg'] + MVP,
+        ['examples/7c910b87-ae3a-f580-d431-03cd89793803.jpg'] + MVP,
+        ['examples/8fa04489-cd94-7d8f-7e2e-5f0fe2f7ae76.jpg'] + MVP,
+        ['examples/bb7b4345-b98a-c727-4c94-6090f0aa4355.jpg'] + MVP,
     ],
     cache_examples=True,
     allow_flagging='never',

diff --git a/docs/_static/theme.css b/docs/_static/theme.css
@@ -1,3 +1,88 @@
 .wy-nav-content {
     max-width: 900px !important;
 }
+
+.black {
+    color: black;
+}
+
+.gray {
+    color: gray;
+}
+
+.grey {
+    color: gray;
+}
+
+.silver {
+    color: silver;
+}
+
+.white {
+    color: white;
+}
+
+.maroon {
+    color: maroon;
+}
+
+.red {
+    color: red;
+}
+
+.magenta {
+    color: magenta;
+}
+
+.fuchsia {
+    color: fuchsia;
+}
+
+.pink {
+    color: pink;
+}
+
+.orange {
+    color: orange;
+}
+
+.yellow {
+    color: yellow;
+}
+
+.lime {
+    color: lime;
+}
+
+.green {
+    color: green;
+    font-weight: bold;
+}
+
+.olive {
+    color: olive;
+}
+
+.teal {
+    color: teal;
+}
+
+.cyan {
+    color: cyan;
+}
+
+.aqua {
+    color: aqua;
+}
+
+.blue {
+    color: blue;
+}
+
+.navy {
+    color: navy;
+}
+
+.purple {
+    color: purple;
+}
diff --git a/docs/colors.rst b/docs/colors.rst
@@ -0,0 +1,25 @@
+.. Color profiles for Sphinx.
+
+.. role:: black
+.. role:: gray
+.. role:: grey
+.. role:: silver
+.. role:: white
+.. role:: maroon
+.. role:: red
+.. role:: magenta
+.. role:: fuchsia
+.. role:: pink
+.. role:: orange
+.. role:: yellow
+.. role:: lime
+.. role:: green
+.. role:: olive
+.. role:: teal
+.. role:: cyan
+.. role:: aqua
+.. role:: blue
+.. role:: navy
+.. role:: purple
+
+.. (c) Lilian Besson, 2011-2016, https://bitbucket.org/lbesson/web-sphinx/
diff --git a/docs/onnx.rst b/docs/onnx.rst
@@ -1,3 +1,5 @@
+.. include:: colors.rst
+
 CDN Model Download (ONNX)
 -------------------------
 
@@ -39,15 +41,15 @@ are not clean and are mapped, for convience, when the final detection labels are
 supported species for each model:
 
    - Phase 1: ``phase1``
-      - `elephant_savanna`
+      - :green:`elephant_savanna`
       - - mapped to: `elephant`
 
    - MVP: ``mvp``
-      - `buffalo`
+      - :green:`buffalo`
       - `camel`
       - `canoe`
       - `car`
-      - `cow`
+      - :green:`cow`
       - `crocodile`
       - `dead_animalwhite_bones`
       - - mapped to: `white_bones`
@@ -56,7 +58,7 @@ supported species for each model:
       - `eland`
       - `elecarcass_old`
       - - mapped to: `white_bones`
-      - `elephant`
+      - :green:`elephant`
       - `gazelle_gr`
       - - mapped to: `gazelle_grants`
       - `gazelle_grants`
@@ -65,25 +67,27 @@ supported species for each model:
       - `gazelle_thomsons`
       - `gerenuk`
       - `giant_forest_hog`
-      - `giraffe`
+      - :green:`giraffe`
       - `goat`
-      - `hartebeest`
-      - `hippo`
+      - :green:`hartebeest`
+      - :green:`hippo`
       - `impala`
-      - `kob`
+      - :green:`kob`
       - `kudu`
       - `motorcycle`
       - `oribi`
-      - `oryx`
+      - :green:`oryx`
       - `ostrich`
       - `roof_grass`
       - `roof_mabati`
       - `sheep`
       - `test`
-      - `topi`
+      - :green:`topi`
       - `vehicle`
-      - `warthog`
-      - `waterbuck`
+      - :green:`warthog`
+      - :green:`waterbuck`
       - `white_bones`
       - `wildebeest`
-      - `zebra`
+      - :green:`zebra`
+
+All species above that are highlighted in green have an Average Precision (AP) of at least 50%.  The other species are supported in a preliminary sense and should not be heavily relied on.