pranjalAI

class ResidualSep(nn.Module):
    def __init__(self, channels, dilation=1):
        super().__init__()

        self.blocks = nn.Sequential(
            nn.ReLU(),
            nn.ReflectionPad2d(dilation),
            nn.Conv2d(channels, channels, kernel_size=3, stride=1, 
                      padding=0, dilation=dilation, 
                      groups=channels, bias=False),

pranjalAI

# Convert the model.
converter = tf.lite.TFLiteConverter.from_keras_model(generator)
tflite_model = converter.convert()

pranjalAI

@tf.function
def train_step(images):
  seed = tf.random.normal([BATCH_SIZE, SEED_SIZE])

  with tf.GradientTape() as gen_tape, tf.GradientTape() as disc_tape:
    generated_images = generator(seed, training=True)

    real_output = discriminator(images, training=True)
    fake_output = discriminator(generated_images, training=True)

pranjalAI

cross_entropy = tf.keras.losses.BinaryCrossentropy(from_logits=True)

def discriminator_loss(real_output, fake_output):
    real_loss = cross_entropy(tf.ones_like(real_output), real_output)
    fake_loss = cross_entropy(tf.zeros_like(fake_output), fake_output)
    total_loss = real_loss + fake_loss
    return total_loss

def generator_loss(fake_output):
    return cross_entropy(tf.ones_like(fake_output), fake_output)

pranjalAI

def build_discriminator(image_shape):
    model = Sequential()

    model.add(Conv2D(32, kernel_size=3, strides=2, input_shape=image_shape, 
                     padding="same"))
    model.add(LeakyReLU(alpha=0.2))

    model.add(Dropout(0.25))
    model.add(Conv2D(64, kernel_size=3, strides=2, padding="same"))
    model.add(ZeroPadding2D(padding=((0,1),(0,1))))

pranjalAI

def build_generator(seed_size, channels):
    model = Sequential()

    model.add(Dense(4*4*256,activation="relu",input_dim=seed_size))
    model.add(Reshape((4,4,256)))

    model.add(UpSampling2D())
    model.add(Conv2D(256,kernel_size=3,padding="same"))
    model.add(BatchNormalization(momentum=0.8))
    model.add(Activation("relu"))

pranjalAI

model.eval()
dummy_input = torch.randn(1, 3, INPUT_HEIGHT, INPUT_WIDTH, 
                          dtype=torch.float32, device=DEVICE)

output = model(dummy_input.detach())

input_names = ['data']
output_names = ['output1']

torch.onnx.export(model, dummy_input, 

pranjalAI

class ResidualSep(nn.Module):
    def __init__(self, channels, dilation=1):
        super().__init__()

        self.blocks = nn.Sequential(
            nn.ReLU(),
            nn.ReflectionPad2d(dilation),
            nn.Conv2d(channels, channels, kernel_size=3, stride=1, 
                      padding=0, dilation=dilation, 
                      groups=channels, bias=False),

pranjalAI

download_file('http://images.cocodataset.org/zips/train2017.zip',
              COCO_IMAGES_ARCHIVE)

download_file('http://images.cocodataset.org/annotations/annotations_trainval2017.zip',
              COCO_ANNOTATIONS_ARCHIVE)
with ZipFile(COCO_ANNOTATIONS_ARCHIVE, 'r') as archive:
    archive.extractall()

pranjalAI

COCO_CATEGORY = 'person'  # target category

WORK_DIR = Path('.')

# Archives to be downloaded
COCO_IMAGES_ARCHIVE = WORK_DIR / 'train2017.zip'
COCO_ANNOTATIONS_ARCHIVE = WORK_DIR / 'annotations_trainval2017.zip'

# Paths where the dataset will be extracted to
COCO_ANNOTATIONS_PATH = WORK_DIR / 'annotations/instances_train2017.json'