ajayarunachalam

def _autoLabel(self, img, rects, masks):
	with open(self._via_export_json, 'r') as fr:
		json_file = json.load(fr)

	regions = []
	N = len(rects)		
	for i in range(N):
		mask = masks[..., i]
		mask = np.where(mask, 255, 0).astype(np.uint8)
		contours, _ = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

ajayarunachalam

# Object Detection using MaskRCNN
def MaskRCNNDetection(self, frame, epsilon=None, auto_mask=True):
	'''
	Detects objects in frame by masking them.
	
	:param frame: An image in BGR format.
	:type frame: :class:`numpy.ndarray`
	:param None epsilon: Used only to have consistency.
		Not used in the algorithm.
	:param bool auto_mask: Should the system perform auto-masking

ajayarunachalam

display_FP_FN(model, test_loader, criterion = nn.CrossEntropyLoss(), display = 'fn')

get_confusion_matrix(y_truth, y_proba, labels = [0, 1])

plot_ROCAUC_curve(y_truth, y_proba, (8, 8))

ajayarunachalam

# performance on test data

def test(model, test_loader, criterion):
    '''
    Model testing 
    
    Args:
        model: model used during training and validation
        test_loader: data loader object containing testing data
        criterion: loss function used

ajayarunachalam

# Save model
PATH = ".\ViTModel.pt" # Use your own path
torch.save(model.state_dict(), PATH)

ajayarunachalam

efficient_transformer = Linformer(
    dim=128,#128
    seq_len=49+1,  # 7x7 patches + 1 cls-token
    depth=12,
    heads=8,
    k=64
)

model = ViT(
    dim=128,

ajayarunachalam

# ### Image Augmentation

#Tune the transforms
ORIG_RES = False

if ORIG_RES:
    resize = 512
else:
    resize = 224

ajayarunachalam

device = 'cpu'

os.makedirs('lung_ct_scan_data', exist_ok=True)

train_dir = 'lung_ct_scan_data/train'
test_dir = 'lung_ct_scan_data/test'

train_list = glob.glob(os.path.join(train_dir,'*.png')) + glob.glob(os.path.join(train_dir,'*.jpg'))
test_list = glob.glob(os.path.join(test_dir, '*.png')) + glob.glob(os.path.join(test_dir, '*.jpg'))

ajayarunachalam

### Training settings

batch_size = 64
epochs = 20
lr = 3e-5
gamma = 0.7
seed = 142
IMG_SIZE = 224

ajayarunachalam

from __future__ import print_function
import glob
from itertools import chain
import os
import random
import zipfile
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import torch