Initial Commit
Browse files- .gitignore +176 -0
- SessionState.py +117 -0
- app.py +406 -0
- download_utils.py +55 -0
- helper.py +23 -0
- image_utils.py +139 -0
.gitignore
ADDED
@@ -0,0 +1,176 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
# Created by https://www.toptal.com/developers/gitignore/api/python
|
3 |
+
# Edit at https://www.toptal.com/developers/gitignore?templates=python
|
4 |
+
|
5 |
+
### Python ###
|
6 |
+
# Byte-compiled / optimized / DLL files
|
7 |
+
__pycache__/
|
8 |
+
*.py[cod]
|
9 |
+
*$py.class
|
10 |
+
|
11 |
+
# C extensions
|
12 |
+
*.so
|
13 |
+
|
14 |
+
# Distribution / packaging
|
15 |
+
.Python
|
16 |
+
build/
|
17 |
+
develop-eggs/
|
18 |
+
dist/
|
19 |
+
downloads/
|
20 |
+
eggs/
|
21 |
+
.eggs/
|
22 |
+
lib/
|
23 |
+
lib64/
|
24 |
+
parts/
|
25 |
+
sdist/
|
26 |
+
var/
|
27 |
+
wheels/
|
28 |
+
share/python-wheels/
|
29 |
+
*.egg-info/
|
30 |
+
.installed.cfg
|
31 |
+
*.egg
|
32 |
+
MANIFEST
|
33 |
+
|
34 |
+
# PyInstaller
|
35 |
+
# Usually these files are written by a python script from a template
|
36 |
+
# before PyInstaller builds the exe, so as to inject date/other infos into it.
|
37 |
+
*.manifest
|
38 |
+
*.spec
|
39 |
+
|
40 |
+
# Installer logs
|
41 |
+
pip-log.txt
|
42 |
+
pip-delete-this-directory.txt
|
43 |
+
|
44 |
+
# Unit test / coverage reports
|
45 |
+
htmlcov/
|
46 |
+
.tox/
|
47 |
+
.nox/
|
48 |
+
.coverage
|
49 |
+
.coverage.*
|
50 |
+
.cache
|
51 |
+
nosetests.xml
|
52 |
+
coverage.xml
|
53 |
+
*.cover
|
54 |
+
*.py,cover
|
55 |
+
.hypothesis/
|
56 |
+
.pytest_cache/
|
57 |
+
cover/
|
58 |
+
|
59 |
+
# Translations
|
60 |
+
*.mo
|
61 |
+
*.pot
|
62 |
+
|
63 |
+
# Django stuff:
|
64 |
+
*.log
|
65 |
+
local_settings.py
|
66 |
+
db.sqlite3
|
67 |
+
db.sqlite3-journal
|
68 |
+
|
69 |
+
# Flask stuff:
|
70 |
+
instance/
|
71 |
+
.webassets-cache
|
72 |
+
|
73 |
+
# Scrapy stuff:
|
74 |
+
.scrapy
|
75 |
+
|
76 |
+
# Sphinx documentation
|
77 |
+
docs/_build/
|
78 |
+
|
79 |
+
# PyBuilder
|
80 |
+
.pybuilder/
|
81 |
+
target/
|
82 |
+
|
83 |
+
# Jupyter Notebook
|
84 |
+
.ipynb_checkpoints
|
85 |
+
|
86 |
+
# IPython
|
87 |
+
profile_default/
|
88 |
+
ipython_config.py
|
89 |
+
|
90 |
+
# pyenv
|
91 |
+
# For a library or package, you might want to ignore these files since the code is
|
92 |
+
# intended to run in multiple environments; otherwise, check them in:
|
93 |
+
# .python-version
|
94 |
+
|
95 |
+
# pipenv
|
96 |
+
# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
|
97 |
+
# However, in case of collaboration, if having platform-specific dependencies or dependencies
|
98 |
+
# having no cross-platform support, pipenv may install dependencies that don't work, or not
|
99 |
+
# install all needed dependencies.
|
100 |
+
#Pipfile.lock
|
101 |
+
|
102 |
+
# poetry
|
103 |
+
# Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
|
104 |
+
# This is especially recommended for binary packages to ensure reproducibility, and is more
|
105 |
+
# commonly ignored for libraries.
|
106 |
+
# https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
|
107 |
+
#poetry.lock
|
108 |
+
|
109 |
+
# PEP 582; used by e.g. github.com/David-OConnor/pyflow
|
110 |
+
__pypackages__/
|
111 |
+
|
112 |
+
# Celery stuff
|
113 |
+
celerybeat-schedule
|
114 |
+
celerybeat.pid
|
115 |
+
|
116 |
+
# SageMath parsed files
|
117 |
+
*.sage.py
|
118 |
+
|
119 |
+
# Environments
|
120 |
+
.env
|
121 |
+
.venv
|
122 |
+
env/
|
123 |
+
venv/
|
124 |
+
ENV/
|
125 |
+
env.bak/
|
126 |
+
venv.bak/
|
127 |
+
|
128 |
+
# Spyder project settings
|
129 |
+
.spyderproject
|
130 |
+
.spyproject
|
131 |
+
|
132 |
+
# Rope project settings
|
133 |
+
.ropeproject
|
134 |
+
|
135 |
+
# mkdocs documentation
|
136 |
+
/site
|
137 |
+
|
138 |
+
# mypy
|
139 |
+
.mypy_cache/
|
140 |
+
.dmypy.json
|
141 |
+
dmypy.json
|
142 |
+
|
143 |
+
# Pyre type checker
|
144 |
+
.pyre/
|
145 |
+
|
146 |
+
# pytype static type analyzer
|
147 |
+
.pytype/
|
148 |
+
|
149 |
+
# Cython debug symbols
|
150 |
+
cython_debug/
|
151 |
+
|
152 |
+
# PyCharm
|
153 |
+
# JetBrains specific template is maintainted in a separate JetBrains.gitignore that can
|
154 |
+
# be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
|
155 |
+
# and can be added to the global gitignore or merged into this file. For a more nuclear
|
156 |
+
# option (not recommended) you can uncomment the following to ignore the entire idea folder.
|
157 |
+
#.idea/
|
158 |
+
|
159 |
+
# End of https://www.toptal.com/developers/gitignore/api/python
|
160 |
+
#
|
161 |
+
#
|
162 |
+
|
163 |
+
ImageNet-HARD-EMD-Real.zip
|
164 |
+
demonstrations.zip
|
165 |
+
ImageNet-HARD-Normal.zip
|
166 |
+
demonstrations/
|
167 |
+
visualizations_feb2022/
|
168 |
+
ImageNet-HARD-EMD-5-Patches-Real.zip
|
169 |
+
visualizations
|
170 |
+
imagenet1k-pilot.tar.gz
|
171 |
+
predictions/
|
172 |
+
imagenet1k-pilot.zip
|
173 |
+
Final.zip
|
174 |
+
imagenet1k-val-50k-emd_results_rosy-brook-184.pickle
|
175 |
+
CUB-Final.zip
|
176 |
+
CUB-Demonstrations/
|
SessionState.py
ADDED
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
"""Hack to add per-session state to Streamlit.
|
2 |
+
|
3 |
+
Usage
|
4 |
+
-----
|
5 |
+
|
6 |
+
>>> import SessionState
|
7 |
+
>>>
|
8 |
+
>>> session_state = SessionState.get(user_name='', favorite_color='black')
|
9 |
+
>>> session_state.user_name
|
10 |
+
''
|
11 |
+
>>> session_state.user_name = 'Mary'
|
12 |
+
>>> session_state.favorite_color
|
13 |
+
'black'
|
14 |
+
|
15 |
+
Since you set user_name above, next time your script runs this will be the
|
16 |
+
result:
|
17 |
+
>>> session_state = get(user_name='', favorite_color='black')
|
18 |
+
>>> session_state.user_name
|
19 |
+
'Mary'
|
20 |
+
|
21 |
+
"""
|
22 |
+
try:
|
23 |
+
import streamlit.ReportThread as ReportThread
|
24 |
+
from streamlit.server.Server import Server
|
25 |
+
except Exception:
|
26 |
+
# Streamlit >= 0.65.0
|
27 |
+
import streamlit.report_thread as ReportThread
|
28 |
+
from streamlit.server.server import Server
|
29 |
+
|
30 |
+
|
31 |
+
class SessionState(object):
|
32 |
+
def __init__(self, **kwargs):
|
33 |
+
"""A new SessionState object.
|
34 |
+
|
35 |
+
Parameters
|
36 |
+
----------
|
37 |
+
**kwargs : any
|
38 |
+
Default values for the session state.
|
39 |
+
|
40 |
+
Example
|
41 |
+
-------
|
42 |
+
>>> session_state = SessionState(user_name='', favorite_color='black')
|
43 |
+
>>> session_state.user_name = 'Mary'
|
44 |
+
''
|
45 |
+
>>> session_state.favorite_color
|
46 |
+
'black'
|
47 |
+
|
48 |
+
"""
|
49 |
+
for key, val in kwargs.items():
|
50 |
+
setattr(self, key, val)
|
51 |
+
|
52 |
+
|
53 |
+
def get(**kwargs):
|
54 |
+
"""Gets a SessionState object for the current session.
|
55 |
+
|
56 |
+
Creates a new object if necessary.
|
57 |
+
|
58 |
+
Parameters
|
59 |
+
----------
|
60 |
+
**kwargs : any
|
61 |
+
Default values you want to add to the session state, if we're creating a
|
62 |
+
new one.
|
63 |
+
|
64 |
+
Example
|
65 |
+
-------
|
66 |
+
>>> session_state = get(user_name='', favorite_color='black')
|
67 |
+
>>> session_state.user_name
|
68 |
+
''
|
69 |
+
>>> session_state.user_name = 'Mary'
|
70 |
+
>>> session_state.favorite_color
|
71 |
+
'black'
|
72 |
+
|
73 |
+
Since you set user_name above, next time your script runs this will be the
|
74 |
+
result:
|
75 |
+
>>> session_state = get(user_name='', favorite_color='black')
|
76 |
+
>>> session_state.user_name
|
77 |
+
'Mary'
|
78 |
+
|
79 |
+
"""
|
80 |
+
# Hack to get the session object from Streamlit.
|
81 |
+
|
82 |
+
ctx = ReportThread.get_report_ctx()
|
83 |
+
|
84 |
+
this_session = None
|
85 |
+
|
86 |
+
current_server = Server.get_current()
|
87 |
+
if hasattr(current_server, '_session_infos'):
|
88 |
+
# Streamlit < 0.56
|
89 |
+
session_infos = Server.get_current()._session_infos.values()
|
90 |
+
else:
|
91 |
+
session_infos = Server.get_current()._session_info_by_id.values()
|
92 |
+
|
93 |
+
for session_info in session_infos:
|
94 |
+
s = session_info.session
|
95 |
+
if (
|
96 |
+
# Streamlit < 0.54.0
|
97 |
+
(hasattr(s, '_main_dg') and s._main_dg == ctx.main_dg)
|
98 |
+
or
|
99 |
+
# Streamlit >= 0.54.0
|
100 |
+
(not hasattr(s, '_main_dg') and s.enqueue == ctx.enqueue)
|
101 |
+
or
|
102 |
+
# Streamlit >= 0.65.2
|
103 |
+
(not hasattr(s, '_main_dg') and s._uploaded_file_mgr == ctx.uploaded_file_mgr)
|
104 |
+
):
|
105 |
+
this_session = s
|
106 |
+
|
107 |
+
if this_session is None:
|
108 |
+
raise RuntimeError(
|
109 |
+
"Oh noes. Couldn't get your Streamlit Session object. "
|
110 |
+
'Are you doing something fancy with threads?')
|
111 |
+
|
112 |
+
# Got the session object! Now let's attach some state into it.
|
113 |
+
|
114 |
+
if not hasattr(this_session, '_custom_session_state'):
|
115 |
+
this_session._custom_session_state = SessionState(**kwargs)
|
116 |
+
|
117 |
+
return this_session._custom_session_state
|
app.py
ADDED
@@ -0,0 +1,406 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
import os
|
3 |
+
import pickle
|
4 |
+
import random
|
5 |
+
import time
|
6 |
+
from collections import Counter
|
7 |
+
from datetime import datetime
|
8 |
+
from glob import glob
|
9 |
+
|
10 |
+
import gdown
|
11 |
+
import matplotlib.pyplot as plt
|
12 |
+
import numpy as np
|
13 |
+
import pandas as pd
|
14 |
+
import seaborn as sns
|
15 |
+
import streamlit as st
|
16 |
+
from PIL import Image
|
17 |
+
|
18 |
+
import SessionState
|
19 |
+
from download_utils import *
|
20 |
+
from image_utils import *
|
21 |
+
|
22 |
+
random.seed(datetime.now())
|
23 |
+
np.random.seed(int(time.time()))
|
24 |
+
|
25 |
+
NUMBER_OF_TRIALS = 20
|
26 |
+
CLASSIFIER_TAG = ""
|
27 |
+
selected_xai_tool = None
|
28 |
+
|
29 |
+
# Config
|
30 |
+
folder_to_name = {}
|
31 |
+
# class_descriptions = {}
|
32 |
+
classifier_predictions = {}
|
33 |
+
selected_dataset = "Task-1-CUB-iNat-HumanStudy"
|
34 |
+
|
35 |
+
root_visualization_dir = "./visualizations/"
|
36 |
+
viz_url = "https://static.taesiri.com/xai/CUB-Task1.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-CHMOnly.pickle"
|
43 |
+
prediction_root = "./predictions/"
|
44 |
+
prediction_pickle = f"{prediction_root}predictions.pickle"
|
45 |
+
|
46 |
+
|
47 |
+
# Get the Data
|
48 |
+
download_files(
|
49 |
+
root_visualization_dir,
|
50 |
+
viz_url,
|
51 |
+
viz_archivefile,
|
52 |
+
demonstration_url,
|
53 |
+
demonst_zipfile,
|
54 |
+
picklefile_url,
|
55 |
+
prediction_root,
|
56 |
+
prediction_pickle,
|
57 |
+
)
|
58 |
+
################################################
|
59 |
+
# GLOBAL VARIABLES
|
60 |
+
app_mode = ""
|
61 |
+
|
62 |
+
# Shared/Global Information
|
63 |
+
birds_list = list(
|
64 |
+
sorted([x.replace(".jpg", "") for x in os.listdir("./CUB-Demonstrations")])
|
65 |
+
)
|
66 |
+
id_to_bird = {i: x for i, x in enumerate(birds_list)}
|
67 |
+
folder_to_name = {x: x for x in birds_list} #
|
68 |
+
################################################
|
69 |
+
|
70 |
+
with open(prediction_pickle, "rb") as f:
|
71 |
+
classifier_predictions = pickle.load(f)
|
72 |
+
|
73 |
+
# SESSION STATE
|
74 |
+
session_state = SessionState.get(
|
75 |
+
page=1,
|
76 |
+
first_run=1,
|
77 |
+
user_feedback={},
|
78 |
+
queries=[],
|
79 |
+
is_classifier_correct={},
|
80 |
+
XAI_tool="Unselected",
|
81 |
+
)
|
82 |
+
################################################
|
83 |
+
|
84 |
+
|
85 |
+
def resmaple_queries():
|
86 |
+
if session_state.first_run == 1:
|
87 |
+
both_correct = glob(
|
88 |
+
root_visualization_dir + selected_dataset + "/Both_correct/*.jpg"
|
89 |
+
)
|
90 |
+
both_wrong = glob(
|
91 |
+
root_visualization_dir + selected_dataset + "/Both_wrong/*.jpg"
|
92 |
+
)
|
93 |
+
|
94 |
+
correct_samples = list(
|
95 |
+
np.random.choice(a=both_correct, size=NUMBER_OF_TRIALS // 2, replace=False)
|
96 |
+
)
|
97 |
+
wrong_samples = list(
|
98 |
+
np.random.choice(a=both_wrong, size=NUMBER_OF_TRIALS // 2, replace=False)
|
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
|
106 |
+
session_state.user_feedback = {}
|
107 |
+
session_state.is_classifier_correct = {}
|
108 |
+
|
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][
|
116 |
+
f"{CLASSIFIER_TAG}-confidence"
|
117 |
+
]
|
118 |
+
prediction_label = folder_to_name[predicted_wnid]
|
119 |
+
# class_def = class_descriptions[predicted_wnid]
|
120 |
+
|
121 |
+
session_state.is_classifier_correct[query_id] = classifier_predictions[query_id][
|
122 |
+
f"{CLASSIFIER_TAG.upper()}-Output"
|
123 |
+
]
|
124 |
+
|
125 |
+
# SHOW QUERY and PREDICTION
|
126 |
+
|
127 |
+
col1, col2 = st.columns(2)
|
128 |
+
with col1:
|
129 |
+
st.image(load_query(current_query), caption=f"Query ID: {query_id}")
|
130 |
+
with col2:
|
131 |
+
# SHOW DESCRIPTION OF CLASS
|
132 |
+
with st.expander("Show Class Description"):
|
133 |
+
st.write(f"**Name**: {prediction_label}")
|
134 |
+
st.write("**Class Definition**:")
|
135 |
+
# st.markdown("`" + class_def + "`")
|
136 |
+
st.image(
|
137 |
+
Image.open(f"CUB-Demonstrations/{predicted_wnid}.jpg"),
|
138 |
+
caption=f"Class Explanation",
|
139 |
+
use_column_width=True,
|
140 |
+
)
|
141 |
+
|
142 |
+
default_value = 0
|
143 |
+
if query_id in session_state.user_feedback.keys():
|
144 |
+
if session_state.user_feedback[query_id] == "Correct":
|
145 |
+
default_value = 1
|
146 |
+
elif session_state.user_feedback[query_id] == "Wrong":
|
147 |
+
default_value = 2
|
148 |
+
|
149 |
+
session_state.user_feedback[query_id] = st.radio(
|
150 |
+
"What do you think about model's prediction?",
|
151 |
+
("-", "Correct", "Wrong"),
|
152 |
+
key=query_id,
|
153 |
+
index=default_value,
|
154 |
+
)
|
155 |
+
st.write(f"**Model Prediction**: {prediction_label}")
|
156 |
+
st.write(f"**Model Confidence**: {prediction_confidence}")
|
157 |
+
|
158 |
+
# SHOW Model Explanation
|
159 |
+
if selected_xai_tool is not None:
|
160 |
+
st.image(
|
161 |
+
selected_xai_tool(current_query),
|
162 |
+
caption=f"Explaination",
|
163 |
+
use_column_width=True,
|
164 |
+
)
|
165 |
+
|
166 |
+
# SHOW DEBUG INFO
|
167 |
+
|
168 |
+
if st.button("Debug: Show Everything"):
|
169 |
+
st.image(Image.open(current_query))
|
170 |
+
|
171 |
+
|
172 |
+
def render_results():
|
173 |
+
user_correct_guess = 0
|
174 |
+
# st.write(session_state.user_feedback)
|
175 |
+
# st.write(session_state.is_classifier_correct)
|
176 |
+
for q in session_state.user_feedback.keys():
|
177 |
+
if session_state.user_feedback[q] != "-":
|
178 |
+
uf = True if session_state.user_feedback[q] == "Correct" else False
|
179 |
+
if session_state.is_classifier_correct[q] == uf:
|
180 |
+
user_correct_guess += 1
|
181 |
+
|
182 |
+
st.write(
|
183 |
+
f"User performance on {CLASSIFIER_TAG}: {user_correct_guess} out of {len( session_state.user_feedback)} Correct"
|
184 |
+
)
|
185 |
+
st.markdown("## User Performance Breakdown")
|
186 |
+
|
187 |
+
categories = [
|
188 |
+
"Correct",
|
189 |
+
"Wrong",
|
190 |
+
] # set(session_state.is_classifier_correct.values())
|
191 |
+
breakdown_stats_correct = {c: 0 for c in categories}
|
192 |
+
breakdown_stats_wrong = {c: 0 for c in categories}
|
193 |
+
|
194 |
+
experiment_summary = []
|
195 |
+
|
196 |
+
for q in session_state.user_feedback.keys():
|
197 |
+
category = "Correct" if session_state.is_classifier_correct[q] else "Wrong"
|
198 |
+
is_user_correct = category == session_state.user_feedback[q]
|
199 |
+
|
200 |
+
if is_user_correct:
|
201 |
+
breakdown_stats_correct[category] += 1
|
202 |
+
else:
|
203 |
+
breakdown_stats_wrong[category] += 1
|
204 |
+
|
205 |
+
experiment_summary.append(
|
206 |
+
[
|
207 |
+
q,
|
208 |
+
classifier_predictions[q]["gt_wnid"],
|
209 |
+
folder_to_name[
|
210 |
+
classifier_predictions[q][f"{CLASSIFIER_TAG}-predictions"]
|
211 |
+
],
|
212 |
+
category,
|
213 |
+
session_state.user_feedback[q],
|
214 |
+
is_user_correct,
|
215 |
+
]
|
216 |
+
)
|
217 |
+
# Summary Table
|
218 |
+
experiment_summary_df = pd.DataFrame.from_records(
|
219 |
+
experiment_summary,
|
220 |
+
columns=[
|
221 |
+
"Query",
|
222 |
+
"GT Labels",
|
223 |
+
f"{CLASSIFIER_TAG} Prediction",
|
224 |
+
"Category",
|
225 |
+
"User Prediction",
|
226 |
+
"Is User Prediction Correct",
|
227 |
+
],
|
228 |
+
)
|
229 |
+
st.write("Summary", experiment_summary_df)
|
230 |
+
|
231 |
+
csv = convert_df(experiment_summary_df)
|
232 |
+
st.download_button(
|
233 |
+
"Press to Download", csv, "summary.csv", "text/csv", key="download-records"
|
234 |
+
)
|
235 |
+
# SHOW BREAKDOWN
|
236 |
+
user_pf_by_model_pred = experiment_summary_df.groupby("Category").agg(
|
237 |
+
{"Is User Prediction Correct": ["count", "sum", "mean"]}
|
238 |
+
)
|
239 |
+
# rename columns
|
240 |
+
user_pf_by_model_pred.columns = user_pf_by_model_pred.columns.droplevel(0)
|
241 |
+
user_pf_by_model_pred.columns = [
|
242 |
+
"Count",
|
243 |
+
"Correct User Guess",
|
244 |
+
"Mean User Performance",
|
245 |
+
]
|
246 |
+
user_pf_by_model_pred.index.name = "Model Prediction"
|
247 |
+
st.write("User performance break down by Model prediction:", user_pf_by_model_pred)
|
248 |
+
csv = convert_df(user_pf_by_model_pred)
|
249 |
+
st.download_button(
|
250 |
+
"Press to Download",
|
251 |
+
csv,
|
252 |
+
"user-performance-by-model-prediction.csv",
|
253 |
+
"text/csv",
|
254 |
+
key="download-performance-by-model-prediction",
|
255 |
+
)
|
256 |
+
# CONFUSION MATRIX
|
257 |
+
|
258 |
+
confusion_matrix = pd.crosstab(
|
259 |
+
experiment_summary_df["Category"],
|
260 |
+
experiment_summary_df["User Prediction"],
|
261 |
+
rownames=["Actual"],
|
262 |
+
colnames=["Predicted"],
|
263 |
+
)
|
264 |
+
st.write("Confusion Matrix", confusion_matrix)
|
265 |
+
csv = convert_df(confusion_matrix)
|
266 |
+
st.download_button(
|
267 |
+
"Press to Download",
|
268 |
+
csv,
|
269 |
+
"confusion-matrix.csv",
|
270 |
+
"text/csv",
|
271 |
+
key="download-confusiion-matrix",
|
272 |
+
)
|
273 |
+
|
274 |
+
|
275 |
+
def render_menu():
|
276 |
+
# Render the readme as markdown using st.markdown.
|
277 |
+
readme_text = st.markdown(
|
278 |
+
"""
|
279 |
+
# Instructions
|
280 |
+
```
|
281 |
+
When testing this study, you should first see the class definition, then hide the expander and see the query.
|
282 |
+
```
|
283 |
+
"""
|
284 |
+
)
|
285 |
+
|
286 |
+
app_mode = st.selectbox(
|
287 |
+
"Choose the page to show:",
|
288 |
+
["Experiment Instruction", "Start Experiment", "See the Results"],
|
289 |
+
)
|
290 |
+
|
291 |
+
if app_mode == "Experiment Instruction":
|
292 |
+
st.success("To continue select an option in the dropdown menu.")
|
293 |
+
elif app_mode == "Start Experiment":
|
294 |
+
# Clear Canvas
|
295 |
+
readme_text.empty()
|
296 |
+
|
297 |
+
page_id = session_state.page
|
298 |
+
col1, col4, col2, col3 = st.columns(4)
|
299 |
+
prev_page = col1.button("Previous Image")
|
300 |
+
|
301 |
+
if prev_page:
|
302 |
+
page_id -= 1
|
303 |
+
if page_id < 1:
|
304 |
+
page_id = 1
|
305 |
+
|
306 |
+
next_page = col2.button("Next Image")
|
307 |
+
|
308 |
+
if next_page:
|
309 |
+
page_id += 1
|
310 |
+
if page_id > NUMBER_OF_TRIALS:
|
311 |
+
page_id = NUMBER_OF_TRIALS
|
312 |
+
|
313 |
+
if page_id == NUMBER_OF_TRIALS:
|
314 |
+
st.success(
|
315 |
+
'You have reached the last image. Please go to the "Results" page to see your performance.'
|
316 |
+
)
|
317 |
+
if st.button("View"):
|
318 |
+
app_mode = "See the Results"
|
319 |
+
|
320 |
+
if col3.button("Resample"):
|
321 |
+
st.write("Restarting ...")
|
322 |
+
page_id = 1
|
323 |
+
session_state.first_run = 1
|
324 |
+
resmaple_queries()
|
325 |
+
|
326 |
+
session_state.page = page_id
|
327 |
+
st.write(f"Render Experiment: {session_state.page}")
|
328 |
+
render_experiment(session_state.page - 1)
|
329 |
+
elif app_mode == "See the Results":
|
330 |
+
readme_text.empty()
|
331 |
+
st.write("Results Summary")
|
332 |
+
render_results()
|
333 |
+
|
334 |
+
|
335 |
+
def main():
|
336 |
+
global app_mode
|
337 |
+
global session_state
|
338 |
+
global selected_xai_tool
|
339 |
+
global CLASSIFIER_TAG
|
340 |
+
|
341 |
+
# Set the session state
|
342 |
+
# State Management and General Setup
|
343 |
+
st.set_page_config(layout="wide")
|
344 |
+
st.title("TASK - 1 - CUB")
|
345 |
+
|
346 |
+
# st.write(classifier_predictions.keys())
|
347 |
+
# st.write(classifier_predictions["ILSVRC2012_val_00024646.JPEG"])
|
348 |
+
|
349 |
+
options = [
|
350 |
+
"Unselected",
|
351 |
+
"NOXAI",
|
352 |
+
"KNN",
|
353 |
+
# "EMD Nearest Neighbors",
|
354 |
+
# "EMD Correspondence",
|
355 |
+
"CHM Nearest Neighbors",
|
356 |
+
"CHM Correspondence",
|
357 |
+
]
|
358 |
+
|
359 |
+
st.markdown(
|
360 |
+
""" <style>
|
361 |
+
div[role="radiogroup"] > :first-child{
|
362 |
+
display: none !important;
|
363 |
+
}
|
364 |
+
</style>
|
365 |
+
""",
|
366 |
+
unsafe_allow_html=True,
|
367 |
+
)
|
368 |
+
|
369 |
+
if session_state.XAI_tool == "Unselected":
|
370 |
+
default = options.index(session_state.XAI_tool)
|
371 |
+
session_state.XAI_tool = st.radio(
|
372 |
+
"What explaination tool do you want to evaluate?",
|
373 |
+
options,
|
374 |
+
key="which_xai",
|
375 |
+
index=default,
|
376 |
+
)
|
377 |
+
# print(session_state.XAI_tool)
|
378 |
+
|
379 |
+
if session_state.XAI_tool != "Unselected":
|
380 |
+
st.markdown(f"## SELECTED METHOD ``{session_state.XAI_tool}``")
|
381 |
+
|
382 |
+
if session_state.XAI_tool == "NOXAI":
|
383 |
+
CLASSIFIER_TAG = "knn"
|
384 |
+
selected_xai_tool = None
|
385 |
+
elif session_state.XAI_tool == "KNN":
|
386 |
+
selected_xai_tool = load_knn_nns
|
387 |
+
CLASSIFIER_TAG = "knn"
|
388 |
+
elif session_state.XAI_tool == "CHM Nearest Neighbors":
|
389 |
+
selected_xai_tool = load_chm_nns
|
390 |
+
CLASSIFIER_TAG = "CHM"
|
391 |
+
elif session_state.XAI_tool == "CHM Correspondence":
|
392 |
+
selected_xai_tool = load_chm_corrs
|
393 |
+
CLASSIFIER_TAG = "CHM"
|
394 |
+
elif session_state.XAI_tool == "EMD Nearest Neighbors":
|
395 |
+
selected_xai_tool = load_emd_nns
|
396 |
+
CLASSIFIER_TAG = "EMD"
|
397 |
+
elif session_state.XAI_tool == "EMD Correspondence":
|
398 |
+
selected_xai_tool = load_emd_corrs
|
399 |
+
CLASSIFIER_TAG = "EMD"
|
400 |
+
|
401 |
+
resmaple_queries()
|
402 |
+
render_menu()
|
403 |
+
|
404 |
+
|
405 |
+
if __name__ == "__main__":
|
406 |
+
main()
|
download_utils.py
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
import os
|
3 |
+
import pickle
|
4 |
+
import random
|
5 |
+
import tarfile
|
6 |
+
import zipfile
|
7 |
+
from collections import Counter
|
8 |
+
from glob import glob
|
9 |
+
|
10 |
+
import gdown
|
11 |
+
import matplotlib.pyplot as plt
|
12 |
+
import numpy as np
|
13 |
+
import pandas as pd
|
14 |
+
import seaborn as sns
|
15 |
+
import streamlit as st
|
16 |
+
from PIL import Image
|
17 |
+
|
18 |
+
import SessionState
|
19 |
+
|
20 |
+
|
21 |
+
def download_files(
|
22 |
+
root_visualization_dir,
|
23 |
+
viz_url,
|
24 |
+
viz_archivefile,
|
25 |
+
demonstration_url,
|
26 |
+
demonst_zipfile,
|
27 |
+
picklefile_url,
|
28 |
+
prediction_root,
|
29 |
+
prediction_pickle,
|
30 |
+
):
|
31 |
+
# Get Visualization
|
32 |
+
if not os.path.exists(root_visualization_dir):
|
33 |
+
gdown.download(viz_url, viz_archivefile, quiet=False)
|
34 |
+
os.makedirs(root_visualization_dir, exist_ok=True)
|
35 |
+
|
36 |
+
if viz_archivefile.endswith("tar.gz"):
|
37 |
+
tar = tarfile.open(viz_archivefile, "r:gz")
|
38 |
+
tar.extractall(path=root_visualization_dir)
|
39 |
+
tar.close()
|
40 |
+
elif viz_archivefile.endswith("zip"):
|
41 |
+
with zipfile.ZipFile(viz_archivefile, "r") as zip_ref:
|
42 |
+
zip_ref.extractall(root_visualization_dir)
|
43 |
+
|
44 |
+
# Get Demonstrations
|
45 |
+
if not os.path.exists(demonst_zipfile):
|
46 |
+
gdown.download(demonstration_url, demonst_zipfile, quiet=False)
|
47 |
+
# os.makedirs(roo_demonstration_dir, exist_ok=True)
|
48 |
+
|
49 |
+
with zipfile.ZipFile(demonst_zipfile, "r") as zip_ref:
|
50 |
+
zip_ref.extractall("./")
|
51 |
+
|
52 |
+
# Get Predictions
|
53 |
+
if not os.path.exists(prediction_pickle):
|
54 |
+
os.makedirs(prediction_root, exist_ok=True)
|
55 |
+
gdown.download(picklefile_url, prediction_pickle, quiet=False)
|
helper.py
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import os
|
2 |
+
|
3 |
+
def get_label_for_query(image_url, model_name):
|
4 |
+
fourway_label = image_url.split('/')[-2]
|
5 |
+
|
6 |
+
if fourway_label=='both_correct':
|
7 |
+
return 'Correct'
|
8 |
+
|
9 |
+
if fourway_label=='both_wrong':
|
10 |
+
return 'Wrong'
|
11 |
+
|
12 |
+
if fourway_label == 'chm_correct_knn_incorrect' and model_name == 'CHM':
|
13 |
+
return 'Correct'
|
14 |
+
elif fourway_label == 'knn_correct_chm_incorrect' and model_name == 'KNN':
|
15 |
+
return 'Correct'
|
16 |
+
|
17 |
+
return 'Wrong'
|
18 |
+
|
19 |
+
def get_category(image_url):
|
20 |
+
return image_url.split('/')[-2]
|
21 |
+
|
22 |
+
def translate_winds_to_names(winds):
|
23 |
+
return [folder_to_name[x] for x in winds]
|
image_utils.py
ADDED
@@ -0,0 +1,139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
import os
|
3 |
+
import pickle
|
4 |
+
import random
|
5 |
+
from glob import glob
|
6 |
+
|
7 |
+
import matplotlib.pyplot as plt
|
8 |
+
import pandas as pd
|
9 |
+
import seaborn as sns
|
10 |
+
import streamlit as st
|
11 |
+
from PIL import Image
|
12 |
+
|
13 |
+
|
14 |
+
@st.cache(allow_output_mutation=True, max_entries=10, ttl=3600)
|
15 |
+
def load_query(image_path):
|
16 |
+
image = Image.open(image_path)
|
17 |
+
width, height = image.size
|
18 |
+
|
19 |
+
new_width = width
|
20 |
+
new_height = height
|
21 |
+
|
22 |
+
left = (width - new_width) / 2
|
23 |
+
top = (height - new_height) / 2
|
24 |
+
right = (width + new_width) / 2
|
25 |
+
bottom = (height + new_height) / 2
|
26 |
+
|
27 |
+
# Crop the center of the image
|
28 |
+
cropped_image = image.crop(
|
29 |
+
(left + 75, top + 145, right - 1790, bottom - (1140))
|
30 |
+
).resize((300, 300))
|
31 |
+
|
32 |
+
return cropped_image
|
33 |
+
|
34 |
+
|
35 |
+
# CHM ############################################################################
|
36 |
+
@st.cache(allow_output_mutation=True, max_entries=10, ttl=3600)
|
37 |
+
def load_chm_nns(image_path):
|
38 |
+
image = Image.open(image_path)
|
39 |
+
width, height = image.size
|
40 |
+
|
41 |
+
new_width = width
|
42 |
+
new_height = height
|
43 |
+
|
44 |
+
left = (width - new_width) / 2
|
45 |
+
top = (height - new_height) / 2
|
46 |
+
right = (width + new_width) / 2
|
47 |
+
bottom = (height + new_height) / 2
|
48 |
+
|
49 |
+
# Crop the center of the image
|
50 |
+
cropped_image = image.crop((left + 485, top + 145, right - 15, bottom - (1140)))
|
51 |
+
return cropped_image
|
52 |
+
|
53 |
+
|
54 |
+
@st.cache(allow_output_mutation=True, max_entries=10, ttl=3600)
|
55 |
+
def load_chm_corrs(image_path):
|
56 |
+
image = Image.open(image_path)
|
57 |
+
width, height = image.size
|
58 |
+
|
59 |
+
new_width = width
|
60 |
+
new_height = height
|
61 |
+
|
62 |
+
left = (width - new_width) / 2
|
63 |
+
top = (height - new_height) / 2
|
64 |
+
right = (width + new_width) / 2
|
65 |
+
bottom = (height + new_height) / 2
|
66 |
+
|
67 |
+
# Crop the center of the image
|
68 |
+
cropped_image = image.crop((left + 485, top + 900, right - 15, bottom - (25 + 10)))
|
69 |
+
return cropped_image
|
70 |
+
|
71 |
+
|
72 |
+
# CHM ############################################################################
|
73 |
+
|
74 |
+
# KNN ############################################################################
|
75 |
+
@st.cache(allow_output_mutation=True, max_entries=10, ttl=3600)
|
76 |
+
def load_knn_nns(image_path):
|
77 |
+
image = Image.open(image_path)
|
78 |
+
width, height = image.size
|
79 |
+
|
80 |
+
new_width = width
|
81 |
+
new_height = height
|
82 |
+
|
83 |
+
left = (width - new_width) / 2
|
84 |
+
top = (height - new_height) / 2
|
85 |
+
right = (width + new_width) / 2
|
86 |
+
bottom = (height + new_height) / 2
|
87 |
+
|
88 |
+
# Crop the center of the image
|
89 |
+
cropped_image = image.crop((left + 485, top + 525, right - 10, bottom - (770)))
|
90 |
+
return cropped_image
|
91 |
+
|
92 |
+
|
93 |
+
# KNN ############################################################################
|
94 |
+
|
95 |
+
# EMD ############################################################################
|
96 |
+
@st.cache(allow_output_mutation=True, max_entries=10, ttl=3600)
|
97 |
+
def load_emd_nns(image_path):
|
98 |
+
image = Image.open(image_path)
|
99 |
+
width, height = image.size
|
100 |
+
|
101 |
+
new_width = width
|
102 |
+
new_height = height
|
103 |
+
|
104 |
+
left = (width - new_width) / 2
|
105 |
+
top = (height - new_height) / 2
|
106 |
+
right = (width + new_width) / 2
|
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 |
+
|
116 |
+
@st.cache(allow_output_mutation=True, max_entries=10, ttl=3600)
|
117 |
+
def load_emd_corrs(image_path):
|
118 |
+
image = Image.open(image_path)
|
119 |
+
width, height = image.size
|
120 |
+
|
121 |
+
new_width = width
|
122 |
+
new_height = height
|
123 |
+
|
124 |
+
left = (width - new_width) / 2
|
125 |
+
top = (height - new_height) / 2
|
126 |
+
right = (width + new_width) / 2
|
127 |
+
bottom = (height + new_height) / 2
|
128 |
+
|
129 |
+
# Crop the center of the image
|
130 |
+
cropped_image = image.crop((left + 10, top + 2500, right - 20, bottom))
|
131 |
+
return cropped_image
|
132 |
+
|
133 |
+
|
134 |
+
# EMD ############################################################################
|
135 |
+
|
136 |
+
|
137 |
+
@st.cache()
|
138 |
+
def convert_df(df):
|
139 |
+
return df.to_csv().encode("utf-8")
|