-
Notifications
You must be signed in to change notification settings - Fork 4
/
utils.py
307 lines (273 loc) · 10.5 KB
/
utils.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
import argparse
import yaml
import importlib
from sklearn.metrics import confusion_matrix
from typing import Optional, Any, Union
import itertools
from matplotlib.colors import ListedColormap
import pandas as pd
import numpy as np
import torch
from matplotlib import pyplot as plt
import os
from PIL import Image
from pathlib import Path
Image.MAX_IMAGE_PIXELS = 100000000000
from .constants import MASK_VALUE_TO_TEXT, MASK_VALUE_TO_COLOR
def get_config():
parser = argparse.ArgumentParser()
parser.add_argument("--base_path",
type=str,
required=True)
parser.add_argument("--config",
type=str,
required=True)
args = parser.parse_args()
assert Path(args.base_path).exists() \
, f"Base path does not exist: {args.base_path}"
assert Path(args.config).exists()\
, f"Config path does not exist: {args.config}"
with open(args.config) as f:
config = yaml.load(f, Loader=yaml.FullLoader)
return Path(args.base_path), config
def get_metadata(constants):
preprocess_directory = constants.PREPROCESS_PATH
superpixel_directory = preprocess_directory / "superpixels"
tissue_mask_directory = preprocess_directory / "tissue_masks"
graph_directory = preprocess_directory / "graphs" / ("partial_" + str(constants.PARTIAL))
image_metadata = pd.read_pickle(constants.IMAGES_DF)
annotation_metadata = pd.read_pickle(constants.ANNOTATIONS_DF)
all_metadata = merge_metadata(
image_metadata=image_metadata,
annotation_metadata=annotation_metadata,
graph_directory=graph_directory,
superpixel_directory=superpixel_directory,
tissue_mask_directory=tissue_mask_directory,
add_image_sizes=True,
)
labels_metadata = pd.read_pickle(constants.LABELS_DF)
label_mapper = to_mapper(labels_metadata)
return all_metadata, label_mapper
def merge_metadata(
image_metadata: pd.DataFrame,
annotation_metadata: pd.DataFrame,
graph_directory: Optional[Path] = None,
superpixel_directory: Optional[Path] = None,
tissue_mask_directory: Optional[Path] = None,
add_image_sizes: bool = False,
):
# Join with image metadata
image_metadata = image_metadata.join(annotation_metadata)
if graph_directory is not None:
graph_metadata = pd.DataFrame(
[
(path.name.split(".")[0], path)
for path in filter(
lambda x: x.name.endswith(".bin"), graph_directory.iterdir()
)
],
columns=["name", "graph_path"],
)
graph_metadata = graph_metadata.set_index("name")
image_metadata = image_metadata.join(graph_metadata)
if superpixel_directory is not None:
superpixel_metadata = pd.DataFrame(
[
(path.name.split(".")[0], path)
for path in filter(
lambda x: x.name.endswith(".h5"), superpixel_directory.iterdir()
)
],
columns=["name", "superpixel_path"],
)
superpixel_metadata = superpixel_metadata.set_index("name")
image_metadata = image_metadata.join(superpixel_metadata)
if tissue_mask_directory is not None:
tissue_metadata = pd.DataFrame(
[
(path.name.split(".")[0], path)
for path in filter(
lambda x: x.name.endswith(".png"), tissue_mask_directory.iterdir()
)
],
columns=["name", "tissue_mask_path"],
)
tissue_metadata = tissue_metadata.set_index("name")
image_metadata = image_metadata.join(tissue_metadata)
# Add image sizes
if add_image_sizes:
image_heights, image_widths = list(), list()
for name, row in image_metadata.iterrows():
image = Image.open(row.annotation_mask_path)
width, height = image.size
image_heights.append(height)
image_widths.append(width)
image_metadata["height"] = image_heights
image_metadata["width"] = image_widths
return image_metadata
def to_mapper(df):
mapper = dict()
for name, row in df.iterrows():
mapper[name] = np.array(
[row["benign"], row["grade3"], row["grade4"], row["grade5"]]
)
return mapper
def create_directory(path):
if not os.path.isdir(path):
os.mkdir(path)
def dynamic_import_from(source_file: str, class_name: str) -> Any:
"""Do a from source_file import class_name dynamically
Args:
source_file (str): Where to import from
class_name (str): What to import
Returns:
Any: The class to be imported
"""
module = importlib.import_module(source_file)
return getattr(module, class_name)
def read_image(image_path: str) -> np.ndarray:
"""Reads an image from a path and converts it into a numpy array
Args:
image_path (str): Path to the image
Returns:
np.array: A numpy array representation of the image
"""
assert image_path.exists()
try:
with Image.open(image_path) as img:
image = np.array(img)
except OSError as e:
raise OSError(e)
return image
def fast_histogram(input_array: np.ndarray, nr_values: int) -> np.ndarray:
"""Calculates a histogram of a matrix of the values from 0 up to (excluding) nr_values
Args:
x (np.array): Input tensor
nr_values (int): Possible values. From 0 up to (exclusing) nr_values.
Returns:
np.array: Output tensor
"""
output_array = np.empty(nr_values, dtype=int)
for i in range(nr_values):
output_array[i] = (input_array == i).sum()
return output_array
def fast_confusion_matrix(y_true: Union[np.ndarray, torch.Tensor], y_pred: Union[np.ndarray, torch.Tensor], nr_classes: int):
"""Faster computation of confusion matrix according to https://stackoverflow.com/a/59089379
Args:
y_true (Union[np.ndarray, torch.Tensor]): Ground truth (1D)
y_pred (Union[np.ndarray, torch.Tensor]): Prediction (1D)
nr_classes (int): Number of classes
Returns:
np.ndarray: Confusion matrix of shape nr_classes x nr_classes
"""
assert y_true.shape == y_pred.shape
y_true = torch.as_tensor(y_true, dtype=torch.long)
y_pred = torch.as_tensor(y_pred, dtype=torch.long)
y = nr_classes * y_true + y_pred
y = torch.bincount(y)
if len(y) < nr_classes * nr_classes:
y = torch.cat((y, torch.zeros(nr_classes * nr_classes - len(y), dtype=torch.long)))
y = y.reshape(nr_classes, nr_classes)
return y.numpy()
def get_batched_segmentation_maps(
node_logits, superpixels, node_associations, NR_CLASSES
):
batch_node_predictions = node_logits.argmax(axis=1).detach().cpu().numpy()
segmentation_maps = np.empty((superpixels.shape), dtype=np.uint8)
start = 0
for i, end in enumerate(node_associations):
node_predictions = batch_node_predictions[start : start + end]
segmentation_maps[i] = get_segmentation_map(
node_predictions, superpixels[i], NR_CLASSES
)
start += end
return segmentation_maps
def get_segmentation_map(node_predictions, superpixels, NR_CLASSES):
all_maps = list()
for label in range(NR_CLASSES):
(spx_indices,) = np.where(node_predictions == label)
spx_indices = spx_indices + 1
map_l = np.isin(superpixels, spx_indices) * label
all_maps.append(map_l)
return np.stack(all_maps).sum(axis=0)
def save_confusion_matrix(prediction, ground_truth, classes, save_path):
cm = confusion_matrix(
y_true=ground_truth, y_pred=prediction, labels=np.arange(len(classes))
)
fig = plot_confusion_matrix(cm, classes, title=None, normalize=False)
fig.savefig(str(save_path), dpi=300, bbox_inches="tight")
def plot_confusion_matrix(
cm, classes, normalize=False, title=None, cmap=plt.cm.Blues
):
"""
This function prints and plots the confusion matrix.
Normalization can be applied by setting `normalize=True`.
(This function is copied from the scikit docs.)
"""
fig, ax = plt.subplots(figsize=(7, 7))
tick_marks = np.arange(len(classes))
plt.xticks(tick_marks, classes, rotation=45, fontsize=18)
plt.yticks(tick_marks, classes, fontsize=18)
if normalize:
cm = cm.astype("float") / cm.sum(axis=1)[:, np.newaxis]
thresh = cm.max() / 2.0
for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
if normalize:
plt.text(
j,
i,
"%.2f" % cm[i, j],
horizontalalignment="center",
color="white" if cm[i, j] > thresh else "black",
fontsize=16,
)
else:
plt.text(
j,
i,
cm[i, j],
horizontalalignment="center",
color="white" if cm[i, j] > thresh else "black",
fontsize=16,
)
plt.tight_layout()
plt.ylabel('True label', fontsize=20)
plt.xlabel('Predicted label', fontsize=20)
ax.imshow(cm, interpolation="nearest", cmap=cmap)
if title is not None:
ax.set_title(title)
return fig
def show_class_activation(per_class_output):
fig, axes = plt.subplots(ncols=2, nrows=2, figsize=(8, 8))
for i, ax in enumerate(axes.flat):
im = ax.imshow(per_class_output[i], vmin=0, vmax=1, cmap="viridis")
ax.set_axis_off()
ax.set_title(MASK_VALUE_TO_TEXT[i])
fig.subplots_adjust(right=0.8)
cbar_ax = fig.add_axes([0.85, 0.15, 0.05, 0.7])
fig.colorbar(im, cax=cbar_ax)
return fig
def show_segmentation_masks(output, annotation=None, **kwargs):
height = 4
width = 5
ncols = 1
if annotation is not None:
width += 5
ncols += 1
fig, ax = plt.subplots(nrows=1, ncols=ncols, figsize=(width, height))
cmap = ListedColormap(MASK_VALUE_TO_COLOR.values())
mask_ax = ax if annotation is None else ax[0]
im = mask_ax.imshow(output, cmap=cmap, vmin=-0.5, vmax=4.5, interpolation="nearest")
mask_ax.axis("off")
if annotation is not None:
ax[1].imshow(
annotation, cmap=cmap, vmin=-0.5, vmax=4.5, interpolation="nearest"
)
ax[1].axis("off")
ax[0].set_title("Prediction")
ax[1].set_title("Ground Truth")
fig.subplots_adjust(right=0.8)
cbar_ax = fig.add_axes([0.85, 0.15, 0.016, 0.7])
cbar = fig.colorbar(im, ticks=[0, 1, 2, 3, 4], cax=cbar_ax)
cbar.ax.set_yticklabels(MASK_VALUE_TO_TEXT.values())
return fig