vashineyu

def bin2img_mix(file_path, channel_first = False):
    w, h = 70, 70
    
    with open(file_path, 'rb') as f:
        affine_info = np.fromfile(f, dtype='float64', count=6)
        img1 = np.fromfile(f, dtype='single',count=245000)
        img2 = np.fromfile(f, dtype='single',count=245000)
        img2[img2 < -900] = -900
    if channel_first:
        return img1.reshape(-1, w, h), img2.reshape(-1,w,h), affine_info

vashineyu

#! /usr/bin/env python
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import tf_parameter_mgr
import monitor_cb
from tensorflow.python.framework import ops
#from monitor_cb import CMonitor

vashineyu

class GroupNorm(Layer):
    '''Group normalization layer
    Group Normalization divides the channels into groups and computes within each group
    the mean and variance for normalization. GN's computation is independent of batch sizes,
    and its accuracy is stable in a wide range of batch sizes
    # Arguments
        groups: Integer, the number of groups for Group Normalization.
        axis: Integer, the axis that should be normalized
            (typically the features axis).
            For instance, after a `Conv2D` layer with

vashineyu

import numpy as np
import os
import pickle as pkl
from tqdm import tqdm

"""
Write data into GPFS
"""
NUM_WRITE_PKL = 100
DATA_PATH = "/mnt/work/debug_data/"

vashineyu

# model.py

import tensorflow as tf
import tensorflow.keras.layers as layers
import tensorflow.keras.models as models
import tensorflow.nn as F

class Conv_bn_relu(layers.Layer):
    """Stack blocks of Conv2D->BN->relu.
    

vashineyu

# model.py

import tensorflow as tf
import tensorflow.keras.layers as layers
import tensorflow.keras.models as models
import tensorflow.nn as F

class Conv_bn_relu(models.Model):
    """Stack blocks of Conv2D->BN->relu.
    

vashineyu

# model.py

import tensorflow as tf
import tensorflow.keras.layers as layers
import tensorflow.keras.models as models
import tensorflow.nn as F

class Conv_bn_relu(models.Model):
    """Stack blocks of Conv2D->BN->relu.
    

vashineyu

"""Data generator for model

GetDataset: Get single data with next
Customized dataloader: Compose multiple dataset object together and put them into multi-processing flow

ver1. all patches were taken from single slide util N patches have been taken.
"""
import cv2
import os
import json

vashineyu

def get_large_image(slide_handler, dpl_list, sz = 256):
    """
    Given partial coordinate list, open them at once and return a large image.
    This is a modification from last version, which dynamically generate patches from slide and predict. However, that method request a interacte with openslide object too frequently, which cause a core-dump. In this version, we read the slide image as a whole then split them with normal operations.
    Args:
        - slide_handler: openslide object (SLIDE_OPENER object)
        - dpl_list: list of coordinates in tuple
        - sz: patch size
    Return:
        - large image: numpy array of partial WSI in high resolution (about 1/n_gpus patches of a WSI)

vashineyu

class GetDataset():
    """Claim Dataset object for inferencing.
    
    Args:
        slide_name (str): full path to slide
        f_inputs_preproc (function): preprocessing function to array
        patch_size (int): patch size of array
        stride (int): stride of image when get patches
        level (int): get patch at level