added EMD
Browse files- app.py +42 -16
- image_utils.py +6 -8
app.py
CHANGED
@@ -30,16 +30,16 @@ selected_xai_tool = None
|
|
30 |
folder_to_name = {}
|
31 |
# class_descriptions = {}
|
32 |
classifier_predictions = {}
|
33 |
-
selected_dataset = "
|
34 |
|
35 |
root_visualization_dir = "./visualizations/"
|
36 |
-
viz_url = "https://static.taesiri.com/xai/CUB-
|
37 |
viz_archivefile = "CUB-Final.zip"
|
38 |
|
39 |
demonstration_url = "https://static.taesiri.com/xai/CUB-Demonstrations.zip"
|
40 |
demonst_zipfile = "demonstrations.zip"
|
41 |
|
42 |
-
picklefile_url = "https://static.taesiri.com/xai/Task1-CUB-
|
43 |
prediction_root = "./predictions/"
|
44 |
prediction_pickle = f"{prediction_root}predictions.pickle"
|
45 |
|
@@ -84,22 +84,48 @@ session_state = SessionState.get(
|
|
84 |
|
85 |
def resmaple_queries():
|
86 |
if session_state.first_run == 1:
|
87 |
-
|
88 |
-
|
89 |
-
)
|
90 |
-
|
91 |
-
|
92 |
-
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
93 |
|
94 |
correct_samples = list(
|
95 |
-
np.random.choice(
|
|
|
|
|
96 |
)
|
97 |
wrong_samples = list(
|
98 |
-
np.random.choice(
|
|
|
|
|
99 |
)
|
100 |
|
101 |
all_images = correct_samples + wrong_samples
|
102 |
random.shuffle(all_images)
|
|
|
103 |
session_state.queries = all_images
|
104 |
session_state.first_run = -1
|
105 |
# RESET INTERACTIONS
|
@@ -109,7 +135,7 @@ def resmaple_queries():
|
|
109 |
|
110 |
def render_experiment(query):
|
111 |
current_query = session_state.queries[query]
|
112 |
-
query_id = os.path.basename(current_query)
|
113 |
|
114 |
predicted_wnid = classifier_predictions[query_id][f"{CLASSIFIER_TAG}-predictions"]
|
115 |
prediction_confidence = classifier_predictions[query_id][
|
@@ -350,8 +376,8 @@ def main():
|
|
350 |
"Unselected",
|
351 |
"NOXAI",
|
352 |
"KNN",
|
353 |
-
|
354 |
-
|
355 |
"CHM Nearest Neighbors",
|
356 |
"CHM Correspondence",
|
357 |
]
|
@@ -380,11 +406,11 @@ def main():
|
|
380 |
st.markdown(f"## SELECTED METHOD ``{session_state.XAI_tool}``")
|
381 |
|
382 |
if session_state.XAI_tool == "NOXAI":
|
383 |
-
CLASSIFIER_TAG = "
|
384 |
selected_xai_tool = None
|
385 |
elif session_state.XAI_tool == "KNN":
|
386 |
selected_xai_tool = load_knn_nns
|
387 |
-
CLASSIFIER_TAG = "
|
388 |
elif session_state.XAI_tool == "CHM Nearest Neighbors":
|
389 |
selected_xai_tool = load_chm_nns
|
390 |
CLASSIFIER_TAG = "CHM"
|
|
|
30 |
folder_to_name = {}
|
31 |
# class_descriptions = {}
|
32 |
classifier_predictions = {}
|
33 |
+
selected_dataset = "CUB-iNAt-Unified"
|
34 |
|
35 |
root_visualization_dir = "./visualizations/"
|
36 |
+
viz_url = "https://static.taesiri.com/xai/CUB-iNAt-Unified.zip"
|
37 |
viz_archivefile = "CUB-Final.zip"
|
38 |
|
39 |
demonstration_url = "https://static.taesiri.com/xai/CUB-Demonstrations.zip"
|
40 |
demonst_zipfile = "demonstrations.zip"
|
41 |
|
42 |
+
picklefile_url = "https://static.taesiri.com/xai/Task1-CUB-ALL.pickle"
|
43 |
prediction_root = "./predictions/"
|
44 |
prediction_pickle = f"{prediction_root}predictions.pickle"
|
45 |
|
|
|
84 |
|
85 |
def resmaple_queries():
|
86 |
if session_state.first_run == 1:
|
87 |
+
|
88 |
+
# EMD_Corrent = [k for k, v in classifier_predictions.items() if v["EMD-Output"]]
|
89 |
+
# EMD_Wrong = [k for k, v in classifier_predictions.items() if not v["EMD-Output"]]
|
90 |
+
# KNN_Corrent = [k for k, v in classifier_predictions.items() if v["KNN-Output"]]
|
91 |
+
# KNN_Wrong = [k for k, v in classifier_predictions.items() if not v["KNN-Output"]]
|
92 |
+
# CHM_Corrent = [k for k, v in classifier_predictions.items() if v["CHM-Output"]]
|
93 |
+
# CHM_Wrong = [k for k, v in classifier_predictions.items() if not v["CHM-Output"]]
|
94 |
+
|
95 |
+
Corret_predictions_idx = [
|
96 |
+
k
|
97 |
+
for k, v in classifier_predictions.items()
|
98 |
+
if v[f"{CLASSIFIER_TAG}-Output"]
|
99 |
+
]
|
100 |
+
Wrong_predictions_idx = [
|
101 |
+
k
|
102 |
+
for k, v in classifier_predictions.items()
|
103 |
+
if not v[f"{CLASSIFIER_TAG}-Output"]
|
104 |
+
]
|
105 |
+
|
106 |
+
correct_classified_plots = [
|
107 |
+
f"{root_visualization_dir}{selected_dataset}/cub-inat-{x}.jpeg"
|
108 |
+
for x in Corret_predictions_idx
|
109 |
+
]
|
110 |
+
wrong_classified_plots = [
|
111 |
+
f"{root_visualization_dir}{selected_dataset}/cub-inat-{x}.jpeg"
|
112 |
+
for x in Wrong_predictions_idx
|
113 |
+
]
|
114 |
|
115 |
correct_samples = list(
|
116 |
+
np.random.choice(
|
117 |
+
a=correct_classified_plots, size=NUMBER_OF_TRIALS // 2, replace=False
|
118 |
+
)
|
119 |
)
|
120 |
wrong_samples = list(
|
121 |
+
np.random.choice(
|
122 |
+
a=wrong_classified_plots, size=NUMBER_OF_TRIALS // 2, replace=False
|
123 |
+
)
|
124 |
)
|
125 |
|
126 |
all_images = correct_samples + wrong_samples
|
127 |
random.shuffle(all_images)
|
128 |
+
|
129 |
session_state.queries = all_images
|
130 |
session_state.first_run = -1
|
131 |
# RESET INTERACTIONS
|
|
|
135 |
|
136 |
def render_experiment(query):
|
137 |
current_query = session_state.queries[query]
|
138 |
+
query_id = int(os.path.basename(current_query).split("-")[2].split(".")[0])
|
139 |
|
140 |
predicted_wnid = classifier_predictions[query_id][f"{CLASSIFIER_TAG}-predictions"]
|
141 |
prediction_confidence = classifier_predictions[query_id][
|
|
|
376 |
"Unselected",
|
377 |
"NOXAI",
|
378 |
"KNN",
|
379 |
+
"EMD Nearest Neighbors",
|
380 |
+
"EMD Correspondence",
|
381 |
"CHM Nearest Neighbors",
|
382 |
"CHM Correspondence",
|
383 |
]
|
|
|
406 |
st.markdown(f"## SELECTED METHOD ``{session_state.XAI_tool}``")
|
407 |
|
408 |
if session_state.XAI_tool == "NOXAI":
|
409 |
+
CLASSIFIER_TAG = "KNN"
|
410 |
selected_xai_tool = None
|
411 |
elif session_state.XAI_tool == "KNN":
|
412 |
selected_xai_tool = load_knn_nns
|
413 |
+
CLASSIFIER_TAG = "KNN"
|
414 |
elif session_state.XAI_tool == "CHM Nearest Neighbors":
|
415 |
selected_xai_tool = load_chm_nns
|
416 |
CLASSIFIER_TAG = "CHM"
|
image_utils.py
CHANGED
@@ -26,7 +26,7 @@ def load_query(image_path):
|
|
26 |
|
27 |
# Crop the center of the image
|
28 |
cropped_image = image.crop(
|
29 |
-
(left +
|
30 |
).resize((300, 300))
|
31 |
|
32 |
return cropped_image
|
@@ -47,7 +47,7 @@ def load_chm_nns(image_path):
|
|
47 |
bottom = (height + new_height) / 2
|
48 |
|
49 |
# Crop the center of the image
|
50 |
-
cropped_image = image.crop((left +
|
51 |
return cropped_image
|
52 |
|
53 |
|
@@ -65,7 +65,7 @@ def load_chm_corrs(image_path):
|
|
65 |
bottom = (height + new_height) / 2
|
66 |
|
67 |
# Crop the center of the image
|
68 |
-
cropped_image = image.crop((left +
|
69 |
return cropped_image
|
70 |
|
71 |
|
@@ -86,7 +86,7 @@ def load_knn_nns(image_path):
|
|
86 |
bottom = (height + new_height) / 2
|
87 |
|
88 |
# Crop the center of the image
|
89 |
-
cropped_image = image.crop((left +
|
90 |
return cropped_image
|
91 |
|
92 |
|
@@ -107,9 +107,7 @@ def load_emd_nns(image_path):
|
|
107 |
bottom = (height + new_height) / 2
|
108 |
|
109 |
# Crop the center of the image
|
110 |
-
cropped_image = image.crop(
|
111 |
-
(left + 10, top + 2075, right - 420, bottom - (925 + 25 + 10))
|
112 |
-
)
|
113 |
return cropped_image
|
114 |
|
115 |
|
@@ -127,7 +125,7 @@ def load_emd_corrs(image_path):
|
|
127 |
bottom = (height + new_height) / 2
|
128 |
|
129 |
# Crop the center of the image
|
130 |
-
cropped_image = image.crop((left +
|
131 |
return cropped_image
|
132 |
|
133 |
|
|
|
26 |
|
27 |
# Crop the center of the image
|
28 |
cropped_image = image.crop(
|
29 |
+
(left + 5, top + 40, right - 2125, bottom - (2805))
|
30 |
).resize((300, 300))
|
31 |
|
32 |
return cropped_image
|
|
|
47 |
bottom = (height + new_height) / 2
|
48 |
|
49 |
# Crop the center of the image
|
50 |
+
cropped_image = image.crop((left + 525, top + 2830, right - 0, bottom - (10)))
|
51 |
return cropped_image
|
52 |
|
53 |
|
|
|
65 |
bottom = (height + new_height) / 2
|
66 |
|
67 |
# Crop the center of the image
|
68 |
+
cropped_image = image.crop((left + 15, top + 1835, right - 45, bottom - 445))
|
69 |
return cropped_image
|
70 |
|
71 |
|
|
|
86 |
bottom = (height + new_height) / 2
|
87 |
|
88 |
# Crop the center of the image
|
89 |
+
cropped_image = image.crop((left + 525, top + 40, right - 10, bottom - (2805)))
|
90 |
return cropped_image
|
91 |
|
92 |
|
|
|
107 |
bottom = (height + new_height) / 2
|
108 |
|
109 |
# Crop the center of the image
|
110 |
+
cropped_image = image.crop((left + 525, top + 480, right - 5, bottom - (2365)))
|
|
|
|
|
111 |
return cropped_image
|
112 |
|
113 |
|
|
|
125 |
bottom = (height + new_height) / 2
|
126 |
|
127 |
# Crop the center of the image
|
128 |
+
cropped_image = image.crop((left + 90, top + 880, right - 75, bottom - 1438))
|
129 |
return cropped_image
|
130 |
|
131 |
|