Only for Notebook Run

ongoing

This code has it's implementation in different languages, refer to "implementation" folder.

v1.0 | v1.1 | v1.2(current)

v0.1 | v0.2

Colab Preview for Code

Website for eagleview

eagleview

A package for greatly increasing the visualisation of datasets, in minimal lines of code.

Usage

It is advised to use `Method 1`, as the github repository is updated from time to time, it may occour that new changes are not still updated.

Method 1

1. Clone the repository:

!git clone https://github.com/hexronuspi/eagleview

2. Upload the folder containing your images. If they are in a zip or any other file, extract them first.

3. Navigate to the cloned repository:

%cd eagleview

4. Install the package:

!pip install .

5. Use the package in your Python code:

Method2

1. using pip install

!pip install eagleview

import eagleview

---

Input:

#v0.x
from eagleview.figshow import ImageMatrix

image_display = ImageMatrix('/content/path_to_folder_containing_images')

#v1.1
from eagleview.figshow import ImageMatrix

im = ImageMatrix('/content/path_to_folder_containing_images', '/content/path_to_file_containing_label.csv')
#only .csv extension

#v1.2
from eagleview.figshow import MultiImageMatrix

im = MultiImageMatrix(
    folder_paths=['random_uploads/images', 'random_uploads/Football Dataset'],
    file_paths=['random_uploads/csv/Book1.csv', '']
)
#only .csv extension

Output:

#v1.2


im.rand((2,4)).display_image(
    {'check_col': 'Image_Name', 'display_cols': ['Color_code', 'type'], 'display_name': True},
    {}, # replace this with first for label displaying, this number should be the same of folder_paths number
    fig_size=(20, 16),
    print_all=False,
    display_label = True,
    x=200,
    y=1600,
    fontsize=(10,10),
    text_color='white',
    label_background_color='black'
)

#v1.1
im.rand((a, b)).display_image( # a,b is the size of matrix
    check_col='col_name', # replace 'col_name' with the name of column, which has image name
    display_label=True,  # by default this is False
    display_cols=['col1', 'col2'], # replace '', to the column name which you want to print as label
    display_name=True, # by default this is False
    print_all=False, # by default this is True
    x= , # X-coordinate of label
    y= , # Y-coordinate of label
    fig_size=(m, n) # by default this is set to max(image_size), maximum size of all the images which will be displayed
    fontsize= , # font size is 10, by default
    text_color='',  # Specify hex code for text color (e.g., white)
    label_background_color='' # Specify hex code for label_background_color (e.g., black)
)

#v0.2
(ImageMatrix('/content/path_to_folder_containing_images').rand()).display_image((2,2), print_all=False, display_name=True)

# v0.1
(ImageMatrix('/content/path_to_folder_containing_images')).display_image((2,2), print_all=False)

Test Run

This is done with a sample dataset which can be found, here.

• Image : `random_uploads/images`
• Label : `random_uploads/csv/Book1.csv`

Output

v1.1

Label Data

Image_Name	Color_code	Type
123	rgb	hole
456	rgb	hole
cube	bw	obstacle
cyllinder	bw	pick
depth-cyllinder	bw	pick
edge1	rgb	map
edge2	rgb	map
edge3	rgb	map
edge4	rgb	map

Releases:

• v0.1

  • This will display the images in a grid with 3 columns and 2 rows, without printing the left out images. If you want to print all images, you can call image_display.display_image((2, 3)) or image_display.display_image((2, 3), True).

• v0.2

  • This will display the name of the images, as their titles. In v0.1, the images were printed starting from first and consecutive until upper bound is reached, this is an optional and random option is added.

• v1.0

  • Added capability to display label with a variey of options.

• v1.1

  • Added capability to change label display and visuals with a variey of options.