CODEWITHSUNDEEP
pythontensorflowmatplotlib
yudhiesh
yudhiesh

Reputation: 6799

Unable to plot images

I am trying to plot the 3 images(img.jpg) belonging to the 3 classes in my dataset with the following code:

plt.figure(figsize=(10, 10))
for images, labels in train_dataset.take(1):
  for i in range(9):
    ax = plt.subplot(3, 3, i + 1)
    plt.imshow(images[i].numpy().astype("uint8"))
    plt.title(class_names[labels[i]])
    plt.axis("off")

But I am getting the following error :

InvalidArgumentError: assertion failed: [Unable to decode bytes as JPEG, PNG, GIF, or BMP]

EDIT:

train_dataset:

train_dataset = tf.keras.preprocessing.image_dataset_from_directory(
    directory=TRAIN_DIR,
    labels="inferred",
    label_mode="int",
    class_names=["0","5","10"],
    batch_size=BATCH_SIZE,
    image_size=(TARGETX, TARGETY),
    shuffle=True,
    seed=SEED,
    validation_split=None,
    subset=None,
    interpolation="bilinear",
    follow_links=False,
)

Upvotes: 0

Views: 674

Answers (2)

yudhiesh
yudhiesh

Reputation: 6799

So I got the outcome I wanted issue was that my images are nested deeply in their folders so I needed to extract their paths before plotting them.

  1. Found the paths to the images and stored them in a list based on their category
IMG_0 = [files for root, directories, files in os.walk(test_img_0, topdown=False)]
IMG_5 = [files for root, directories, files in os.walk(test_img_5, topdown=False)]
IMG_10 = [files for root, directories, files in os.walk(test_img_10, topdown=False)]
  1. Function to plot the images:
def show_images(images, cols = 1, titles = None):
    """Display a list of images in a single figure with matplotlib.
    
    Parameters
    ---------
    images: List of np.arrays compatible with plt.imshow.
    
    cols (Default = 1): Number of columns in figure (number of rows is 
                        set to np.ceil(n_images/float(cols))).
    
    titles: List of titles corresponding to each image. Must have
            the same length as titles.
    """
    assert((titles is None)or (len(images) == len(titles)))
    n_images = len(images)
    if titles is None: titles = ['Image (%d)' % i for i in range(1,n_images + 1)]
    fig = plt.figure()
    for n, (image, title) in enumerate(zip(images, titles)):
        a = fig.add_subplot(cols, np.ceil(n_images/float(cols)), n + 1)
        if image.ndim == 2:
            plt.gray()
        plt.imshow(image)
        a.set_title(title)
    fig.set_size_inches(np.array(fig.get_size_inches()) * n_images)
    plt.show()

Then I got the absolute paths to the images and sent them as a list into the function:

img1 = mpimg.imread(os.path.join(test_img_0, IMG_0[0][0]))
img2 = mpimg.imread(os.path.join(test_img_5, IMG_5[0][0]))
img3 = mpimg.imread(os.path.join(test_img_10, IMG_10[0][0]))

show_images([img1, img2, img3], cols = 2, titles=["Class 0", "Class 5", "Class 10"])

Upvotes: 0

Aniket Bote
Aniket Bote

Reputation: 3564

There is nothing wrong with the plotting function. This most likely an issue with your dataset.

See this issue.

If you want to check whether the plotting function is correct.

import pathlib
dataset_url = "https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz"
data_dir = tf.keras.utils.get_file(origin=dataset_url, 
                                   fname='flower_photos', 
                                   untar=True)
data_dir = pathlib.Path(data_dir) 

batch_size = 32
img_height = 180
img_width = 180

train_dataset = tf.keras.preprocessing.image_dataset_from_directory(
  data_dir,
  validation_split=0.2,
  subset="training",
  seed=123,
  image_size=(img_height, img_width),
  batch_size=batch_size)
class_names = train_dataset.class_names

Note: This will download the flowers dataset and build a tf.dataset object. You can pass the train_dataset to plotting function to see if there are any issues.

Checks for your dataset:

  1. Ensure that you have only images in your directories.
  2. Make sure the depts of images match. ie All images should have the same channels.
  3. Make sure all images have the same extensions.

Upvotes: 1

Related Questions