Python cm.jet_r方法代码示例

# 需要导入模块: from matplotlib import cm [as 别名]
# 或者: from matplotlib.cm import jet_r [as 别名]
def plot_pc_old(pc_np, z_cutoff=70, birds_view=False, color='height', size=0.3, ax=None):
    # remove large z points
    valid_index = pc_np[:, 2] < z_cutoff
    pc_np = pc_np[valid_index, :]

    if ax is None:
        fig = plt.figure(figsize=(9, 9))
        ax = Axes3D(fig)
    if color == 'height':
        c = pc_np[:, 1]
        ax.scatter(pc_np[:, 0].tolist(), pc_np[:, 1].tolist(), pc_np[:, 2].tolist(), s=size, c=c, cmap=cm.jet_r)
    elif color == 'reflectance':
        assert False
    else:
        ax.scatter(pc_np[:, 0].tolist(), pc_np[:, 1].tolist(), pc_np[:, 2].tolist(), s=size, c=color)

    axisEqual3D(ax)
    if True == birds_view:
        ax.view_init(elev=0, azim=-90)
    else:
        ax.view_init(elev=-45, azim=-90)
    # ax.invert_yaxis()

    return ax 

# 需要导入模块: from matplotlib import cm [as 别名]
# 或者: from matplotlib.cm import jet_r [as 别名]
def colorize_depth(depth_map):
    # scale everything to [0, 255]
    sorted_depth = np.unique(np.sort(depth_map.flatten()))
    min_depth = sorted_depth[0]
    max_depth = sorted_depth[len(sorted_depth) - 1]

    depth_map = np.asarray(map(lambda pixel:
                               (pixel - min_depth) * 1.0 / (max_depth - min_depth),
                               depth_map))

    # Apply jet colormap to it
    depth_map = np.uint8(cm.jet_r(depth_map) * 255)
    return depth_map[:, :, 0:3]


# Given a CSV row of metadata, colorize the image and save into a destination 

# 需要导入模块: from matplotlib import cm [as 别名]
# 或者: from matplotlib.cm import jet_r [as 别名]
def save_gradcam(filename, gcam, raw_image, paper_cmap=False):
    gcam = gcam.cpu().numpy()
    cmap = cm.jet_r(gcam)[..., :3] * 255.0
    if paper_cmap:
        alpha = gcam[..., None]
        gcam = alpha * cmap + (1 - alpha) * raw_image
    else:
        gcam = (cmap.astype(np.float) + raw_image.astype(np.float)) / 2
    cv2.imwrite(filename, np.uint8(gcam)) 

# 需要导入模块: from matplotlib import cm [as 别名]
# 或者: from matplotlib.cm import jet_r [as 别名]
def draw_node(cls, node, surface, origin, size, config):
        import pygame
        cmap = cm.jet_r
        norm = mpl.colors.Normalize(vmin=0, vmax=1 / (1 - config["gamma"]))
        color = cmap(norm(node.get_value()), bytes=True)
        pygame.draw.rect(surface, color, (origin[0], origin[1], size[0], size[1]), 0) 

# 需要导入模块: from matplotlib import cm [as 别名]
# 或者: from matplotlib.cm import jet_r [as 别名]
def display_highway(cls, agent, surface):
        """
            Particular visualization of the state space that is used for highway_env environments only.

        :param agent: the agent to be displayed
        :param surface: the surface on which the agent is displayed.
        """
        import pygame
        norm = mpl.colors.Normalize(vmin=-2, vmax=2)
        cmap = cm.jet_r
        try:
            grid_shape = agent.mdp.original_shape
        except AttributeError:
            grid_shape = cls.highway_module.finite_mdp.compute_ttc_grid(agent.env, time_quantization=1., horizon=10.).shape
        cell_size = (surface.get_width() // grid_shape[2], surface.get_height() // (grid_shape[0] * grid_shape[1]))
        speed_size = surface.get_height() // grid_shape[0]
        value = agent.get_state_value().reshape(grid_shape)
        for h in range(grid_shape[0]):
            for i in range(grid_shape[1]):
                for j in range(grid_shape[2]):
                    color = cmap(norm(value[h, i, j]), bytes=True)
                    pygame.draw.rect(surface, color, (
                        j * cell_size[0], i * cell_size[1] + h * speed_size, cell_size[0], cell_size[1]), 0)
            pygame.draw.line(surface, cls.BLACK,
                             (0, h * speed_size), (grid_shape[2] * cell_size[0], h * speed_size), 1)
        states, actions = agent.plan_trajectory(agent.mdp.state)
        for state in states:
            (h, i, j) = np.unravel_index(state, grid_shape)
            pygame.draw.rect(surface, cls.RED,
                             (j * cell_size[0], i * cell_size[1] + h * speed_size, cell_size[0], cell_size[1]), 1) 

# 需要导入模块: from matplotlib import cm [as 别名]
# 或者: from matplotlib.cm import jet_r [as 别名]
def display(cls, agent, surface, sim_surface=None, display_text=True):
        """
            Display the action-values for the current state

        :param agent: the DQNAgent to be displayed
        :param surface: the pygame surface on which the agent is displayed
        :param sim_surface: the pygame surface on which the env is rendered
        :param display_text: whether to display the action values as text
        """
        import pygame
        action_values = agent.get_state_action_values(agent.previous_state)
        action_distribution = agent.action_distribution(agent.previous_state)

        cell_size = (surface.get_width() // len(action_values), surface.get_height())
        pygame.draw.rect(surface, cls.BLACK, (0, 0, surface.get_width(), surface.get_height()), 0)

        # Display node value
        for action, value in enumerate(action_values):
            cmap = cm.jet_r
            norm = mpl.colors.Normalize(vmin=0, vmax=1/(1-agent.config["gamma"]))
            color = cmap(norm(value), bytes=True)
            pygame.draw.rect(surface, color, (cell_size[0]*action, 0, cell_size[0], cell_size[1]), 0)

            if display_text:
                font = pygame.font.Font(None, 15)
                text = "v={:.2f} / p={:.2f}".format(value, action_distribution[action])
                text = font.render(text,
                                   1, (10, 10, 10), (255, 255, 255))
                surface.blit(text, (cell_size[0]*action, 0))

        if sim_surface and hasattr(agent.value_net, "get_attention_matrix"):
            cls.display_vehicles_attention(agent, sim_surface) 

# 需要导入模块: from matplotlib import cm [as 别名]
# 或者: from matplotlib.cm import jet_r [as 别名]
def live(config_path, model_path, cuda, crf, camera_id):
    """
    Inference from camera stream
    """

    # Setup
    CONFIG = OmegaConf.load(config_path)
    device = get_device(cuda)
    torch.set_grad_enabled(False)
    torch.backends.cudnn.benchmark = True

    classes = get_classtable(CONFIG)
    postprocessor = setup_postprocessor(CONFIG) if crf else None

    model = eval(CONFIG.MODEL.NAME)(n_classes=CONFIG.DATASET.N_CLASSES)
    state_dict = torch.load(model_path, map_location=lambda storage, loc: storage)
    model.load_state_dict(state_dict)
    model.eval()
    model.to(device)
    print("Model:", CONFIG.MODEL.NAME)

    # UVC camera stream
    cap = cv2.VideoCapture(camera_id)
    cap.set(cv2.CAP_PROP_FOURCC, cv2.VideoWriter_fourcc(*"YUYV"))

    def colorize(labelmap):
        # Assign a unique color to each label
        labelmap = labelmap.astype(np.float32) / CONFIG.DATASET.N_CLASSES
        colormap = cm.jet_r(labelmap)[..., :-1] * 255.0
        return np.uint8(colormap)

    def mouse_event(event, x, y, flags, labelmap):
        # Show a class name of a mouse-overed pixel
        label = labelmap[y, x]
        name = classes[label]
        print(name)

    window_name = "{} + {}".format(CONFIG.MODEL.NAME, CONFIG.DATASET.NAME)
    cv2.namedWindow(window_name, cv2.WINDOW_AUTOSIZE)

    while True:
        _, frame = cap.read()
        image, raw_image = preprocessing(frame, device, CONFIG)
        labelmap = inference(model, image, raw_image, postprocessor)
        colormap = colorize(labelmap)

        # Register mouse callback function
        cv2.setMouseCallback(window_name, mouse_event, labelmap)

        # Overlay prediction
        cv2.addWeighted(colormap, 0.5, raw_image, 0.5, 0.0, raw_image)

        # Quit by pressing "q" key
        cv2.imshow(window_name, raw_image)
        if cv2.waitKey(10) == ord("q"):
            break