Skip to content
Snippets Groups Projects
Commit 8fbacdab authored by Nathan Zhao's avatar Nathan Zhao
Browse files

fix imagenet, remove cv2

parent 17846ffc
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
hpvm/projects/keras/README.md
+ 3
− 3
View file @ 8fbacdab
...@@ -48,14 +48,14 @@ List of benchmarks and the expected accuracies: ...@@ -48,14 +48,14 @@ List of benchmarks and the expected accuracies:
| ----------- | ----------- | | ----------- | ----------- |
| AlexNet-CIFAR10 | 79.16 | | AlexNet-CIFAR10 | 79.16 |
| AlexNet2-CIFAR10 | 85.10 | | AlexNet2-CIFAR10 | 85.10 |
| AlexNet-ImageNet | 56.23 | todo: fix broken | AlexNet-ImageNet | 56.30 |
| LeNet-MNIST | 99.11 | todo: fix broken | LeNet-MNIST | 99.11 | todo: fix broken
| MobileNet-CIFAR10 | 82.40 | | MobileNet-CIFAR10 | 82.40 |
| ResNet18-CIFAR10 | 89.52 | | ResNet18-CIFAR10 | 89.52 |
| ResNet50-ImageNet | 74.50 | | ResNet50-ImageNet | 75.10 |
| VGG16-CIFAR10 | 89.42 | | VGG16-CIFAR10 | 89.42 |
| VGG16-CIFAR100 | 66.20 | | VGG16-CIFAR100 | 66.20 |
| VGG16-ImageNet | 72.50 | todo: fix broken | VGG16-ImageNet | 69.46 |
Activate conda environment (above) before running benchmarks Activate conda environment (above) before running benchmarks
......
hpvm/projects/keras/keras_environment.yml
+ 0
− 1
View file @ 8fbacdab
...@@ -32,5 +32,4 @@ dependencies: ...@@ -32,5 +32,4 @@ dependencies:
- msgpack==0.5.6 - msgpack==0.5.6
- tables==3.4.4 - tables==3.4.4
- torch==0.4.1 - torch==0.4.1
- opencv-python==4.5.1.48
hpvm/projects/keras/src/alexnet_imagenet.py
+ 6
− 67
View file @ 8fbacdab
...@@ -4,7 +4,6 @@ import glob ...@@ -4,7 +4,6 @@ import glob
import numpy as np import numpy as np
import tensorflow as tf import tensorflow as tf
import cv2
import scipy import scipy
import scipy.io import scipy.io
import keras import keras
...@@ -29,75 +28,15 @@ IMAGES_PER_CLASS = 50 ...@@ -29,75 +28,15 @@ IMAGES_PER_CLASS = 50
class AlexNet(Benchmark): class AlexNet(Benchmark):
def load_image(self, x):
image = cv2.imread(x)
height, width, _ = image.shape
new_height = height * 256 // min(image.shape[:2])
new_width = width * 256 // min(image.shape[:2])
image = cv2.resize(image, (new_width, new_height), interpolation=cv2.INTER_CUBIC)
# center crop 224 x 224
height, width, _ = image.shape
startx = width // 2 - (224 // 2)
starty = height // 2 - (224 // 2)
image = image[starty:starty + 224, startx:startx + 224]
image = image / 255 # normalize [0, 1]
image = image[:, :, ::-1] # BGR -> RGB
image = np.transpose(image, (2, 0, 1)) # (H, W, C) -> (C, H, W)
image[0, :, :] = (image[0, :, :] - 0.485) / 0.229
image[1, :, :] = (image[1, :, :] - 0.456) / 0.224
image[2, :, :] = (image[2, :, :] - 0.406) / 0.225
return image.astype(np.float32)
def data_preprocess(self): def data_preprocess(self):
self.synset_to_keras_idx = {} X_val = np.fromfile(MODEL_PARAMS_DIR + '/alexnet_imagenet/test_input.bin', dtype=np.float32)
y_val = np.fromfile(MODEL_PARAMS_DIR + '/alexnet_imagenet/test_labels.bin', dtype=np.uint32)
f = open(IMAGENET_DIR + 'ILSVRC2012_devkit_t12/data/synset_words.txt', 'r')
c = 0 X_val = X_val.reshape((-1, 3, 224, 224))
for line in f: X_train, y_train = None, None
parts = line.split(' ')
self.synset_to_keras_idx[parts[0]] = c
c += 1
f.close()
X_train, X_val = [], []
y_train, y_val = [], []
classes = glob.glob(IMAGENET_DIR + 'val/*')
for c in np.random.permutation(len(classes))[:NUM_TUNE_CLASSES]:
x = glob.glob(classes[c] + '/*')
x = np.array(x)
idx = np.random.permutation(len(x))
idx = idx[:min(len(idx), IMAGES_PER_CLASS)]
synset = classes[c].split('/')[-1]
images = list(map(lambda x : self.load_image(x), x[idx]))
labels = [self.synset_to_keras_idx[synset]] * len(x[idx])
split = int(len(idx) * 0.5)
X_val += images[:split]
y_val += labels[:split]
X_train += images[split:]
y_train += labels[split:]
X_train = np.array(X_train)
y_train = np.array(y_train)
X_val = np.array(X_val)
y_val = np.array(y_val)
X_test = X_val[0:5000] X_test = X_val[0:5000]
y_test = y_val[0:5000] y_test = y_val[0:5000]
......
hpvm/projects/keras/src/lenet.py
+ 2
− 0
View file @ 8fbacdab
...@@ -110,6 +110,8 @@ if __name__ == '__main__': ...@@ -110,6 +110,8 @@ if __name__ == '__main__':
src_dir = 'data/lenet_mnist_src/' src_dir = 'data/lenet_mnist_src/'
num_classes = 10 num_classes = 10
batch_size = 500 batch_size = 500
print (reload_dir)
model = LeNet_MNIST('LeNet_MNIST', reload_dir, keras_model_file, data_dir, src_dir, num_classes, batch_size) model = LeNet_MNIST('LeNet_MNIST', reload_dir, keras_model_file, data_dir, src_dir, num_classes, batch_size)
......
hpvm/projects/keras/src/resnet50_imagenet.py
+ 7
− 62
View file @ 8fbacdab
...@@ -4,7 +4,6 @@ import glob ...@@ -4,7 +4,6 @@ import glob
import numpy as np import numpy as np
import tensorflow as tf import tensorflow as tf
import cv2
import scipy import scipy
import scipy.io import scipy.io
import keras import keras
...@@ -30,27 +29,6 @@ IMAGES_PER_CLASS = 50 ...@@ -30,27 +29,6 @@ IMAGES_PER_CLASS = 50
class ResNet50(Benchmark): class ResNet50(Benchmark):
def load_image(self, x):
image = cv2.imread(x)
height, width, _ = image.shape
new_height = height * 256 // min(image.shape[:2])
new_width = width * 256 // min(image.shape[:2])
image = cv2.resize(image, (new_width, new_height), interpolation=cv2.INTER_CUBIC)
height, width, _ = image.shape
startx = width // 2 - (224 // 2)
starty = height // 2 - (224 // 2)
image = image[starty:starty + 224, startx:startx + 224]
image = image[:, :, ::-1]
image = np.transpose(image, (2, 0, 1))
image = preprocess_input(image.astype(np.float32), data_format="channels_first")
return image.astype(np.float32)
def buildModel(self): def buildModel(self):
...@@ -143,51 +121,18 @@ class ResNet50(Benchmark): ...@@ -143,51 +121,18 @@ class ResNet50(Benchmark):
def data_preprocess(self): def data_preprocess(self):
self.synset_to_keras_idx = {} X_val = np.fromfile(MODEL_PARAMS_DIR + '/resnet50_imagenet/test_input.bin', dtype=np.float32)
y_val = np.fromfile(MODEL_PARAMS_DIR + '/resnet50_imagenet/test_labels.bin', dtype=np.uint32)
f = open(IMAGENET_DIR + 'ILSVRC2012_devkit_t12/data/synset_words.txt', 'r')
c = 0 X_val = X_val.reshape((-1, 3, 224, 224))
for line in f: X_train, y_train = None, None
parts = line.split(' ')
self.synset_to_keras_idx[parts[0]] = c
c += 1
f.close()
X_train, X_val = [], []
y_train, y_val = [], []
classes = glob.glob(IMAGENET_DIR + 'val/*')
for c in np.random.permutation(len(classes))[:NUM_TUNE_CLASSES]:
x = glob.glob(classes[c] + '/*')
x = np.array(x)
idx = np.random.permutation(len(x))
idx = idx[:min(len(idx), IMAGES_PER_CLASS)]
synset = classes[c].split('/')[-1]
images = list(map(lambda x : self.load_image(x), x[idx]))
labels = [self.synset_to_keras_idx[synset]] * len(x[idx])
split = int(len(idx) * 0.5)
X_val += images[:split]
y_val += labels[:split]
X_train += images[split:]
y_train += labels[split:]
X_train = np.array(X_train)
y_train = np.array(y_train)
X_val = np.array(X_val)
y_val = np.array(y_val)
X_test = X_val[0:5000] X_test = X_val[0:5000]
y_test = y_val[0:5000] y_test = y_val[0:5000]
X_tuner = X_val[5000:] X_tuner = X_val[5000:]
y_tuner = y_val[5000:] y_tuner = y_val[5000:]
return X_train, y_train, X_test, y_test, X_tuner, y_tuner return X_train, y_train, X_test, y_test, X_tuner, y_tuner
......
hpvm/projects/keras/src/vgg16_imagenet.py
+ 7
− 67
View file @ 8fbacdab
...@@ -4,7 +4,6 @@ import glob ...@@ -4,7 +4,6 @@ import glob
import numpy as np import numpy as np
import tensorflow as tf import tensorflow as tf
import cv2
import scipy import scipy
import scipy.io import scipy.io
import keras import keras
...@@ -29,33 +28,7 @@ IMAGES_PER_CLASS = 50 ...@@ -29,33 +28,7 @@ IMAGES_PER_CLASS = 50
class VGG16(Benchmark): class VGG16(Benchmark):
def load_image(self, x):
image = cv2.imread(x)
height, width, _ = image.shape
new_height = height * 256 // min(image.shape[:2])
new_width = width * 256 // min(image.shape[:2])
image = cv2.resize(image, (new_width, new_height), interpolation=cv2.INTER_CUBIC)
# center crop 224 x 224
height, width, _ = image.shape
startx = width // 2 - (224 // 2)
starty = height // 2 - (224 // 2)
image = image[starty:starty + 224, startx:startx + 224]
image = image / 255 # normalize [0, 1]
image = image[:, :, ::-1] # BGR -> RGB
image = np.transpose(image, (2, 0, 1)) # (H, W, C) -> (C, H, W)
image[0, :, :] = (image[0, :, :] - 0.485) / 0.229
image[1, :, :] = (image[1, :, :] - 0.456) / 0.224
image[2, :, :] = (image[2, :, :] - 0.406) / 0.225
return image.astype(np.float32)
def buildModel(self): def buildModel(self):
img_input = Input(shape=(3, 224, 224)) img_input = Input(shape=(3, 224, 224))
...@@ -132,51 +105,18 @@ class VGG16(Benchmark): ...@@ -132,51 +105,18 @@ class VGG16(Benchmark):
def data_preprocess(self): def data_preprocess(self):
self.synset_to_keras_idx = {} X_val = np.fromfile(MODEL_PARAMS_DIR + '/vgg16_imagenet/test_input.bin', dtype=np.float32)
y_val = np.fromfile(MODEL_PARAMS_DIR + '/vgg16_imagenet/test_labels.bin', dtype=np.uint32)
f = open(IMAGENET_DIR + 'ILSVRC2012_devkit_t12/data/synset_words.txt', 'r')
c = 0 X_val = X_val.reshape((-1, 3, 224, 224))
for line in f: X_train, y_train = None, None
parts = line.split(' ')
self.synset_to_keras_idx[parts[0]] = c
c += 1
f.close()
X_train, X_val = [], []
y_train, y_val = [], []
classes = glob.glob(IMAGENET_DIR + 'val/*')
for c in np.random.permutation(len(classes))[:NUM_TUNE_CLASSES]:
x = glob.glob(classes[c] + '/*')
x = np.array(x)
idx = np.random.permutation(len(x))
idx = idx[:min(len(idx), IMAGES_PER_CLASS)]
synset = classes[c].split('/')[-1]
images = list(map(lambda x : self.load_image(x), x[idx]))
labels = [self.synset_to_keras_idx[synset]] * len(x[idx])
split = int(len(idx) * 0.5)
X_val += images[:split]
y_val += labels[:split]
X_train += images[split:]
y_train += labels[split:]
X_train = np.array(X_train)
y_train = np.array(y_train)
X_val = np.array(X_val)
y_val = np.array(y_val)
X_test = X_val[0:5000] X_test = X_val[0:5000]
y_test = y_val[0:5000] y_test = y_val[0:5000]
X_tuner = X_val[5000:] X_tuner = X_val[5000:]
y_tuner = y_val[5000:] y_tuner = y_val[5000:]
return X_train, y_train, X_test, y_test, X_tuner, y_tuner return X_train, y_train, X_test, y_test, X_tuner, y_tuner
......
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment