Python config.G属性代码示例

# 需要导入模块: import config [as 别名]
# 或者: from config import G [as 别名]
def setup_snapshot_image_grid(G, training_set,
    size    = '1080p',      # '1080p' = to be viewed on 1080p display, '4k' = to be viewed on 4k display.
    layout  = 'random'):    # 'random' = grid contents are selected randomly, 'row_per_class' = each row corresponds to one class label.

    # Select size.
    gw = 1; gh = 1
    if size == '1080p':
        gw = np.clip(1920 // G.output_shape[3], 3, 32)
        gh = np.clip(1080 // G.output_shape[2], 2, 32)
    if size == '4k':
        gw = np.clip(3840 // G.output_shape[3], 7, 32)
        gh = np.clip(2160 // G.output_shape[2], 4, 32)

    # Fill in reals and labels.
    reals = np.zeros([gw * gh] + training_set.shape, dtype=training_set.dtype)
    labels = np.zeros([gw * gh, training_set.label_size], dtype=training_set.label_dtype)
    masks = np.zeros([gw * gh] + [1, training_set.shape[-1], training_set.shape[-1]], dtype=training_set.dtype)
    for idx in range(gw * gh):
        x = idx % gw; y = idx // gw
        while True:
            real, label, mask = training_set.get_minibatch_np(1)
            if layout == 'row_per_class' and training_set.label_size > 0:
                if label[0, y % training_set.label_size] == 0.0:
                    continue
            reals[idx] = real[0]
            labels[idx] = label[0]
            masks[idx] = mask[0]
            break

    # Generate latents.
    latents = misc.random_latents(gw * gh, G)
    return (gw, gh), reals, labels, latents, masks

#----------------------------------------------------------------------------
# Just-in-time processing of training images before feeding them to the networks. 

# 需要导入模块: import config [as 别名]
# 或者: from config import G [as 别名]
def setup_snapshot_image_grid(G, training_set,
    size    = '1080p',      # '1080p' = to be viewed on 1080p display, '4k' = to be viewed on 4k display.
    layout  = 'random'):    # 'random' = grid contents are selected randomly, 'row_per_class' = each row corresponds to one class label.

    # Select size.
    gw = 1; gh = 1
    if size == '1080p':
        gw = np.clip(1920 // G.output_shape[3], 3, 32)
        gh = np.clip(1080 // G.output_shape[2], 2, 32)
    if size == '4k':
        gw = np.clip(3840 // G.output_shape[3], 7, 32)
        gh = np.clip(2160 // G.output_shape[2], 4, 32)

    # Fill in reals and labels.
    reals = np.zeros([gw * gh] + training_set.shape, dtype=training_set.dtype)
    labels = np.zeros([gw * gh, training_set.label_size], dtype=training_set.label_dtype)
    for idx in range(gw * gh):
        x = idx % gw; y = idx // gw
        while True:
            real, label = training_set.get_minibatch_np(1)
            if layout == 'row_per_class' and training_set.label_size > 0:
                if label[0, y % training_set.label_size] == 0.0:
                    continue
            reals[idx] = real[0]
            labels[idx] = label[0]
            break

    # Generate latents.
    latents = misc.random_latents(gw * gh, G)
    return (gw, gh), reals, labels, latents

#----------------------------------------------------------------------------
# Just-in-time processing of training images before feeding them to the networks. 

# 需要导入模块: import config [as 别名]
# 或者: from config import G [as 别名]
def load_G_weights(G, path, by_name = True):
    G_path = os.path.join(path,'Generator.h5')
    G.load_weights(G_path, by_name = by_name)
    return G 

# 需要导入模块: import config [as 别名]
# 或者: from config import G [as 别名]
def predict_gan():
    separate_funcs          = False
    drange_net              = [-1,1]
    drange_viz              = [-1,1]
    image_grid_size         = None
    image_grid_type         = 'default'
    resume_network          = 'pre-trained_weight'
    
    np.random.seed(config.random_seed)

    if resume_network:
        print("Resuming weight from:"+resume_network)
        G = Generator(num_channels=3, resolution=128, label_size=0, **config.G)
        G = load_G_weights(G,resume_network,True)

    print(G.summary())

    # Misc init.

    if image_grid_type == 'default':
        if image_grid_size is None:
            w, h = G.output_shape[1], G.output_shape[2]
            print("w:%d,h:%d"%(w,h))
            image_grid_size = np.clip(int(1920 // w), 3, 16).astype('int'), np.clip(1080 / h, 2, 16).astype('int')
        
        print("image_grid_size:",image_grid_size)
    else:
        raise ValueError('Invalid image_grid_type', image_grid_type)

    result_subdir = misc.create_result_subdir('pre-trained_result', config.run_desc)

    for i in range(1,6):
        snapshot_fake_latents = random_latents(np.prod(image_grid_size), G.input_shape)
        snapshot_fake_images = G.predict_on_batch(snapshot_fake_latents)
        misc.save_image_grid(snapshot_fake_images, os.path.join(result_subdir, 'pre-trained_%03d.png'%i), drange=drange_viz, grid_size=image_grid_size) 

# 需要导入模块: import config [as 别名]
# 或者: from config import G [as 别名]
def load_GD(path, compile = False):
    G_path = os.path.join(path,'Generator.h5')
    D_path = os.path.join(path,'Discriminator.h5')
    G = load_model(G_path, compile = compile)
    D = load_model(D_path, compile = compile)
    return G,D 

# 需要导入模块: import config [as 别名]
# 或者: from config import G [as 别名]
def save_GD(G,D,path,overwrite = False):

        os.makedirs(path);
        G_path = os.path.join(path,'Generator.h5')
        D_path = os.path.join(path,'Discriminator.h5')
        save_model(G,G_path,overwrite = overwrite)
        save_model(D,D_path,overwrite = overwrite)
        print("Save model to %s"%path) 

# 需要导入模块: import config [as 别名]
# 或者: from config import G [as 别名]
def load_GD_weights(G,D,path, by_name = True):
    G_path = os.path.join(path,'Generator.h5')
    D_path = os.path.join(path,'Discriminator.h5')
    G.load_weights(G_path, by_name = by_name)
    D.load_weights(D_path, by_name = by_name)
    return G,D