icas 1.0

Tool for labeling images

icas

Tool for clustering and segmenting image datasets. Detailed descriptions for packages and pipelines can be found in githup repo. Below is just basic usage sytle.

Test the whole system pipeline with:

* clusterimg test

import icas.clusterimg.clusterimg_test as ct
ct()

* Deep Learning clusterimg test

import icas.clusterimg.DL_clusterimg_test as dl_ct
dl_ct()

* segmentimg test

import icas.segmentimg.segmentimg_test as st
st()

Use the icas systems with:

* clusterimg

from icas.clusterimg.Clusteror import  Clusteror

test_path = "path/to/images"
method = "TM"
batch_size = 500
threshold = 0.5

cl = Clusteror(test_path, method, method, threshold=threshold, overwrite=True)
cl()

* Deep Learning clusterimg

from icas.clusterimg.DL_ModelTrainer import ModelTrainer
from icas.clusterimg.DL_Datasets import ImageDataset
from icas.clusterimg.DL_Models import PowerOf2s32to128
from icas.clusterimg.DL_Clusteror import DL_Clusteror

device = "cpu"
method = "kmeans"
loss = "mse"
test_path = "path/to/images"

mt = ModelTrainer(num_of_epochs=1, lr=0.001, batch_size=2, loss_type=loss, dataset=ImageDataset(test_path), model=PowerOf2s32to128(), verbose=0, device=device)

dlc = DL_Clusteror(model_trainer=mt, method=method, batch_size=100, overwrite=True, verbose=0)
dlc()

* segmentimg

from icas.segmentimg.Segmentator import  Segmentator

test_path = "path/to/images"
method = "graph"

sg = Segmentator(test_path, method=method)
sg()

To user attention:

We insist you to add comments, descriptions and example usage to your contributions

Below libraries are not installed with icas since aim for icas is to be lightweight. Advanced usage such as Deep Learning Clustering and SAM segmentation requires below installations:

pip install torch
pip install torchvision
pip install scikit-learn
pip install git+https://github.com/facebookresearch/segment-anything.git