{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"
\n\n

\n

Author: @ammarnassanalhajali

\n
Original Notebook: Notebook

\n

\n
","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2021-08-20T17:42:42.629428Z","iopub.execute_input":"2021-08-20T17:42:42.630187Z","iopub.status.idle":"2021-08-20T17:42:42.644664Z","shell.execute_reply.started":"2021-08-20T17:42:42.630091Z","shell.execute_reply":"2021-08-20T17:42:42.6433Z"}}},{"cell_type":"markdown","source":"

In this Notebook we will be going through and annotating @ammarnassanalhajali 's \"šŸ§ Brain Tumor 3D [Training]\" Notebook
( please see the original work )

\n\n### The Goal: provide clear analysis as to what the logic in each function is doing such that anyone new to ml or python can replicate a given concept with their own logic after some critical thought\n\n#### Last Couple of Remarks before the Annotations: I am currently working through annotating the model section of this notebook as well as a few more preprocessing notebooks. Thanks for Reading!\n\n#### (8/25 update) I apologize for the late annotations on this, school started up and things got busy\n\n#### (8/25 update) see author's continuation of this nb w/ inference @ https://www.kaggle.com/ammarnassanalhajali/brain-tumor-3d-inference/data","metadata":{}},{"cell_type":"markdown","source":"#### **
Imports
**","metadata":{}},{"cell_type":"code","source":"import os\n#for file management\nimport json\n#for standardized data storage\nimport glob\n#for precise file selection(low verbosity)\nimport random\n#random amount generator\nimport collections\n#has premade datastructure objects that can be implemented\nimport time\n#return the time in seconds since the epoch as a floating point number. A good way to watch your training progress quantitatively\nimport re\n#import regular expression matching operations. alows for facilitated use of string comparison. Docs:https://docs.python.org/3/library/re.html\nimport math\n#provides mathematical functions established within the c-standard. gives ceil, factorial, etc. Docs:https://docs.python.org/3/library/math.html\nimport numpy as np\n#linear algebra\nimport pandas as pd\n#succinct array and data handling\nimport cv2\n#image data handler\n\nimport matplotlib.pyplot as plt\n#data visualization library\nimport seaborn as sns\n#data visualization library\n\nimport pydicom\n#c based dicom modulation tool\nfrom pydicom.pixel_data_handlers.util import apply_voi_lut\n#pixel handler that uses voi_lut to grab pixel data from within the frame of a window\n\nfrom random import shuffle\n#random library allows for generation of pseudo-random number and ways to use os to get as random as feasible. shuffle is method provided that will use a random function to effectively randomize the order of an array\nfrom sklearn import model_selection as sk_model_selection\n#the sklearn library provides the model_slection library and we set the name for this notebook to be 'sk_model_selection'. it wraps the input validation data(X and true for training purposes ) and splits it according \n #to a random function that can thereby be set to have a seed that will assure reproducible results accross training sessions. Stratification(approximately even proportions of 0 and 1 data) is the default\n #Docs: https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html\n\nfrom tensorflow import keras\n#the notorious tf library provises the keras framework that can be used to interface with artifical neural networks via methods and pre-built objects from the libary\n #Docs: https://faroit.com/keras-docs/1.2.0/\nfrom tensorflow.keras import layers\n#from the keras library import layers: \"A layer encapsulates both a state (the layer's \"weights\") and a transformation from inputs to outputs (a \"call\", the layer's forward pass).\". source: https://www.tensorflow.org/guide/keras/custom_layers_and_models\nimport tensorflow as tf\n#importing the tensor flow library as tf","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:01.889129Z","iopub.execute_input":"2021-08-25T09:24:01.890525Z","iopub.status.idle":"2021-08-25T09:24:09.216177Z","shell.execute_reply.started":"2021-08-25T09:24:01.890429Z","shell.execute_reply":"2021-08-25T09:24:09.215244Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated
**","metadata":{}},{"cell_type":"code","source":"data_directory = '../input/rsna-miccai-brain-tumor-radiogenomic-classification'\n#set data_directory variable to file location of the main dir\npytorch3dpath = \"../input/efficientnetpyttorch3d/EfficientNet-PyTorch-3D\"\n#here we are loading a pretrained model from the EfficientNet-PyTorch-3D library","metadata":{"_kg_hide-input":false,"execution":{"iopub.status.busy":"2021-08-25T09:24:09.217736Z","iopub.execute_input":"2021-08-25T09:24:09.218034Z","iopub.status.idle":"2021-08-25T09:24:09.222549Z","shell.execute_reply.started":"2021-08-25T09:24:09.218007Z","shell.execute_reply":"2021-08-25T09:24:09.221551Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**I went ahead and grabbed the model details and will leave it in the next code block for those interested:**\n> I can, and likely will have to come back and review this in more detail and annotate some more when I get to my model collection period for ensembling this competition(will get back with this)","metadata":{}},{"cell_type":"code","source":"# import torch\n# from torch import nn\n# from torch.nn import functional as F\n\n# from .utils import (\n# round_filters,\n# round_repeats,\n# drop_connect,\n# get_same_padding_conv3d,\n# get_model_params,\n# efficientnet_params,\n# Swish,\n# MemoryEfficientSwish,\n# )\n\n# class MBConvBlock3D(nn.Module):\n# \"\"\"\n# Mobile Inverted Residual Bottleneck Block\n\n# Args:\n# block_args (namedtuple): BlockArgs, see above\n# global_params (namedtuple): GlobalParam, see above\n\n# Attributes:\n# has_se (bool): Whether the block contains a Squeeze and Excitation layer.\n# \"\"\"\n\n# def __init__(self, block_args, global_params):\n# super().__init__()\n# self._block_args = block_args\n# self._bn_mom = 1 - global_params.batch_norm_momentum\n# self._bn_eps = global_params.batch_norm_epsilon\n# self.has_se = (self._block_args.se_ratio is not None) and (0 < self._block_args.se_ratio <= 1)\n# self.id_skip = block_args.id_skip # skip connection and drop connect\n\n# # Get static or dynamic convolution depending on image size\n# Conv3d = get_same_padding_conv3d(image_size=global_params.image_size)\n\n# # Expansion phase\n# inp = self._block_args.input_filters # number of input channels\n# oup = self._block_args.input_filters * self._block_args.expand_ratio # number of output channels\n# if self._block_args.expand_ratio != 1:\n# self._expand_conv = Conv3d(in_channels=inp, out_channels=oup, kernel_size=1, bias=False)\n# self._bn0 = nn.BatchNorm3d(num_features=oup, momentum=self._bn_mom, eps=self._bn_eps)\n\n# # Depthwise convolution phase\n# k = self._block_args.kernel_size\n# s = self._block_args.stride\n# self._depthwise_conv = Conv3d(\n# in_channels=oup, out_channels=oup, groups=oup, # groups makes it depthwise\n# kernel_size=k, stride=s, bias=False)\n# self._bn1 = nn.BatchNorm3d(num_features=oup, momentum=self._bn_mom, eps=self._bn_eps)\n\n# # Squeeze and Excitation layer, if desired\n# if self.has_se:\n# num_squeezed_channels = max(1, int(self._block_args.input_filters * self._block_args.se_ratio))\n# self._se_reduce = Conv3d(in_channels=oup, out_channels=num_squeezed_channels, kernel_size=1)\n# self._se_expand = Conv3d(in_channels=num_squeezed_channels, out_channels=oup, kernel_size=1)\n\n# # Output phase\n# final_oup = self._block_args.output_filters\n# self._project_conv = Conv3d(in_channels=oup, out_channels=final_oup, kernel_size=1, bias=False)\n# self._bn2 = nn.BatchNorm3d(num_features=final_oup, momentum=self._bn_mom, eps=self._bn_eps)\n# self._swish = MemoryEfficientSwish()\n\n# def forward(self, inputs, drop_connect_rate=None):\n# \"\"\"\n# :param inputs: input tensor\n# :param drop_connect_rate: drop connect rate (float, between 0 and 1)\n# :return: output of block\n# \"\"\"\n\n# # Expansion and Depthwise Convolution\n# x = inputs\n# if self._block_args.expand_ratio != 1:\n# x = self._swish(self._bn0(self._expand_conv(inputs)))\n# x = self._swish(self._bn1(self._depthwise_conv(x)))\n\n# # Squeeze and Excitation\n# if self.has_se:\n# x_squeezed = F.adaptive_avg_pool3d(x, 1)\n# x_squeezed = self._se_expand(self._swish(self._se_reduce(x_squeezed)))\n# x = torch.sigmoid(x_squeezed) * x\n\n# x = self._bn2(self._project_conv(x))\n\n# # Skip connection and drop connect\n# input_filters, output_filters = self._block_args.input_filters, self._block_args.output_filters\n# if self.id_skip and self._block_args.stride == 1 and input_filters == output_filters:\n# if drop_connect_rate:\n# x = drop_connect(x, p=drop_connect_rate, training=self.training)\n# x = x + inputs # skip connection\n# return x\n\n# def set_swish(self, memory_efficient=True):\n# \"\"\"Sets swish function as memory efficient (for training) or standard (for export)\"\"\"\n# self._swish = MemoryEfficientSwish() if memory_efficient else Swish()\n\n\n# class EfficientNet3D(nn.Module):\n# \"\"\"\n# An EfficientNet model. Most easily loaded with the .from_name or .from_pretrained methods\n\n# Args:\n# blocks_args (list): A list of BlockArgs to construct blocks\n# global_params (namedtuple): A set of GlobalParams shared between blocks\n\n# Example:\n# model = EfficientNet3D.from_pretrained('efficientnet-b0')\n\n# \"\"\"\n\n# def __init__(self, blocks_args=None, global_params=None, in_channels=3):\n# super().__init__()\n# assert isinstance(blocks_args, list), 'blocks_args should be a list'\n# assert len(blocks_args) > 0, 'block args must be greater than 0'\n# self._global_params = global_params\n# self._blocks_args = blocks_args\n\n# # Get static or dynamic convolution depending on image size\n# Conv3d = get_same_padding_conv3d(image_size=global_params.image_size)\n\n# # Batch norm parameters\n# bn_mom = 1 - self._global_params.batch_norm_momentum\n# bn_eps = self._global_params.batch_norm_epsilon\n\n# # Stem\n# out_channels = round_filters(32, self._global_params) # number of output channels\n# self._conv_stem = Conv3d(in_channels, out_channels, kernel_size=3, stride=2, bias=False)\n# self._bn0 = nn.BatchNorm3d(num_features=out_channels, momentum=bn_mom, eps=bn_eps)\n\n# # Build blocks\n# self._blocks = nn.ModuleList([])\n# for block_args in self._blocks_args:\n\n# # Update block input and output filters based on depth multiplier.\n# block_args = block_args._replace(\n# input_filters=round_filters(block_args.input_filters, self._global_params),\n# output_filters=round_filters(block_args.output_filters, self._global_params),\n# num_repeat=round_repeats(block_args.num_repeat, self._global_params)\n# )\n\n# # The first block needs to take care of stride and filter size increase.\n# self._blocks.append(MBConvBlock3D(block_args, self._global_params))\n# if block_args.num_repeat > 1:\n# block_args = block_args._replace(input_filters=block_args.output_filters, stride=1)\n# for _ in range(block_args.num_repeat - 1):\n# self._blocks.append(MBConvBlock3D(block_args, self._global_params))\n\n# # Head\n# in_channels = block_args.output_filters # output of final block\n# out_channels = round_filters(1280, self._global_params)\n# self._conv_head = Conv3d(in_channels, out_channels, kernel_size=1, bias=False)\n# self._bn1 = nn.BatchNorm3d(num_features=out_channels, momentum=bn_mom, eps=bn_eps)\n\n# # Final linear layer\n# self._avg_pooling = nn.AdaptiveAvgPool3d(1)\n# self._dropout = nn.Dropout(self._global_params.dropout_rate)\n# self._fc = nn.Linear(out_channels, self._global_params.num_classes)\n# self._swish = MemoryEfficientSwish()\n\n# def set_swish(self, memory_efficient=True):\n# \"\"\"Sets swish function as memory efficient (for training) or standard (for export)\"\"\"\n# self._swish = MemoryEfficientSwish() if memory_efficient else Swish()\n# for block in self._blocks:\n# block.set_swish(memory_efficient)\n\n\n# def extract_features(self, inputs):\n# \"\"\" Returns output of the final convolution layer \"\"\"\n\n# # Stem\n# x = self._swish(self._bn0(self._conv_stem(inputs)))\n\n# # Blocks\n# for idx, block in enumerate(self._blocks):\n# drop_connect_rate = self._global_params.drop_connect_rate\n# if drop_connect_rate:\n# drop_connect_rate *= float(idx) / len(self._blocks)\n# x = block(x, drop_connect_rate=drop_connect_rate)\n\n# # Head\n# x = self._swish(self._bn1(self._conv_head(x)))\n\n# return x\n\n# def forward(self, inputs):\n# \"\"\" Calls extract_features to extract features, applies final linear layer, and returns logits. \"\"\"\n# bs = inputs.size(0)\n# # Convolution layers\n# x = self.extract_features(inputs)\n\n# # Pooling and final linear layer\n# x = self._avg_pooling(x)\n# x = x.view(bs, -1)\n# x = self._dropout(x)\n# x = self._fc(x)\n# return x\n\n# @classmethod\n# def from_name(cls, model_name, override_params=None, in_channels=3):\n# cls._check_model_name_is_valid(model_name)\n# blocks_args, global_params = get_model_params(model_name, override_params)\n# return cls(blocks_args, global_params, in_channels)\n\n# @classmethod\n# def get_image_size(cls, model_name):\n# cls._check_model_name_is_valid(model_name)\n# _, _, res, _ = efficientnet_params(model_name)\n# return res\n\n# @classmethod\n# def _check_model_name_is_valid(cls, model_name):\n# \"\"\" Validates model name. \"\"\" \n# valid_models = ['efficientnet-b'+str(i) for i in range(9)]\n# if model_name not in valid_models:\n# raise ValueError('model_name should be one of: ' + ', '.join(valid_models))","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-20T19:43:07.702115Z","iopub.execute_input":"2021-08-20T19:43:07.702762Z","iopub.status.idle":"2021-08-20T19:43:07.716402Z","shell.execute_reply.started":"2021-08-20T19:43:07.70271Z","shell.execute_reply":"2021-08-20T19:43:07.714899Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Data setup","metadata":{}},{"cell_type":"code","source":"mri_types = ['FLAIR','T1w','T1wCE','T2w']\n#setting the mri_types variable to the 4 image types we have in our folders \nSIZE = 256\n#setting a SIZE to be the magic number 256\nNUM_IMAGES = 64\n#setting the a magic number variable NUM_IMAGES to 64","metadata":{"_kg_hide-input":false,"execution":{"iopub.status.busy":"2021-08-25T09:24:15.504328Z","iopub.execute_input":"2021-08-25T09:24:15.504707Z","iopub.status.idle":"2021-08-25T09:24:15.509071Z","shell.execute_reply.started":"2021-08-25T09:24:15.50467Z","shell.execute_reply":"2021-08-25T09:24:15.508101Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# \"why 256 and 64?\"* \n[source](https://www.researchgate.net/post/Which_Image_resolution_should_I_use_for_training_for_deep_neural_network)\n******\n\n**The answer lies in the object we will be declaring later that has been provided to us via this pytorch library, time constraints, and opinion.**\n\n**256x256 is 'known-good' for image classification problems and we will be training our data in batches as we have quite a bit and the choice was 64 here.**\n\n**later on we will be generating a 3D object from our 2d slices, we will make the depth of said objects 64 pixels**\n\n**32-64 is also allegedly ideal for batches of image data but ultimately it depends on if you have a GPU cluster at your finger tips or not as to whether or not you want to do larger batch sizes with more epochs(not me sadly, taking donations;) )**","metadata":{}},{"cell_type":"code","source":"train_df = pd.read_csv(\"../input/rsna-miccai-brain-tumor-radiogenomic-classification/train_labels.csv\")\n#creating a dataframe from our train_labels.csv\n\ntrain_df['BraTS21ID5'] = [format(x, '05d') for x in train_df.BraTS21ID]\n#we are creating a column BraTS21ID5 that represents the 5 digit long ID and uses regular expressions in order to reformat the string contents that are read from the BraTS21ID column \n #side note: this was clever","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:19.334606Z","iopub.execute_input":"2021-08-25T09:24:19.335127Z","iopub.status.idle":"2021-08-25T09:24:19.360498Z","shell.execute_reply.started":"2021-08-25T09:24:19.335094Z","shell.execute_reply":"2021-08-25T09:24:19.359703Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## [Quick Demonstration] Printing some regular expressions","metadata":{}},{"cell_type":"code","source":"x = train_df.BraTS21ID[25]\nprint(str(x) + \" versus \" + format(x, '05d'))\n# format(x, '05d') is effectively in words: format the d(data) in x such that it 5 characters where the empty slots are zero","metadata":{"execution":{"iopub.status.busy":"2021-08-25T09:24:23.372638Z","iopub.execute_input":"2021-08-25T09:24:23.373029Z","iopub.status.idle":"2021-08-25T09:24:23.378301Z","shell.execute_reply.started":"2021-08-25T09:24:23.372997Z","shell.execute_reply":"2021-08-25T09:24:23.377057Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df.head(5)\n#prints the first 5 rows of the dataframe","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:26.04322Z","iopub.execute_input":"2021-08-25T09:24:26.043543Z","iopub.status.idle":"2021-08-25T09:24:26.063446Z","shell.execute_reply.started":"2021-08-25T09:24:26.043514Z","shell.execute_reply":"2021-08-25T09:24:26.062696Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: Loading Data
**","metadata":{}},{"cell_type":"code","source":"data_directory = '../input/rsna-miccai-brain-tumor-radiogenomic-classification'\npytorch3dpath = \"../input/efficientnetpyttorch3d/EfficientNet-PyTorch-3D\"\n \nmri_types = ['FLAIR','T1w','T1wCE','T2w']\nIMAGE_SIZE = 256\nNUM_IMAGES = 64\nBATCH_SIZE= 4\ntrain_df = pd.read_csv(\"../input/rsna-miccai-brain-tumor-radiogenomic-classification/train_labels.csv\")\ntrain_df['BraTS21ID5'] = [format(x, '05d') for x in train_df.BraTS21ID]\ntrain_df.head(3)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:30.783502Z","iopub.execute_input":"2021-08-25T09:24:30.78402Z","iopub.status.idle":"2021-08-25T09:24:30.80023Z","shell.execute_reply.started":"2021-08-25T09:24:30.783987Z","shell.execute_reply":"2021-08-25T09:24:30.799027Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated: Loading Data
**","metadata":{}},{"cell_type":"code","source":"data_directory = '../input/rsna-miccai-brain-tumor-radiogenomic-classification'\n#set data_directory variable to file location of the main dir\npytorch3dpath = \"../input/efficientnetpyttorch3d/EfficientNet-PyTorch-3D\"\n#here we are loading a pretrained model from the EfficientNet-PyTorch-3D library\n \nmri_types = ['FLAIR','T1w','T1wCE','T2w']\n#setting the mri_types variable to the 4 image types we have in our folders \nSIZE = 256\n#setting a SIZE to be the magic number 256\nNUM_IMAGES = 64\n#setting the a magic number variable NUM_IMAGES to 64\n\ntrain_df = pd.read_csv(\"../input/rsna-miccai-brain-tumor-radiogenomic-classification/train_labels.csv\")\n#creating a dataframe from our train_labels.csv\n\ntrain_df['BraTS21ID5'] = [format(x, '05d') for x in train_df.BraTS21ID]\n#we are creating a column BraTS21ID5 that represents the 5 digit long ID and uses regular expressions in order to reformat the string contents that are read from the BraTS21ID column ","metadata":{"execution":{"iopub.status.busy":"2021-08-20T19:43:07.792336Z","iopub.execute_input":"2021-08-20T19:43:07.792629Z","iopub.status.idle":"2021-08-20T19:43:07.814775Z","shell.execute_reply.started":"2021-08-20T19:43:07.792599Z","shell.execute_reply":"2021-08-20T19:43:07.813274Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: Functions to load images
**","metadata":{}},{"cell_type":"code","source":"def load_dicom_image(path, img_size=SIZE, voi_lut=True, rotate=0):\n dicom = pydicom.read_file(path)\n data = dicom.pixel_array\n if voi_lut:\n data = apply_voi_lut(dicom.pixel_array, dicom)\n else:\n data = dicom.pixel_array\n \n if rotate > 0:\n rot_choices = [0, cv2.ROTATE_90_CLOCKWISE, cv2.ROTATE_90_COUNTERCLOCKWISE, cv2.ROTATE_180]\n data = cv2.rotate(data, rot_choices[rotate])\n \n data = cv2.resize(data, (img_size, img_size))\n return data\n\n\ndef load_dicom_images_3d(scan_id, num_imgs=NUM_IMAGES, img_size=SIZE, mri_type=\"FLAIR\", split=\"train\", rotate=0):\n\n files = sorted(glob.glob(f\"{data_directory}/{split}/{scan_id}/{mri_type}/*.dcm\"), \n key=lambda var:[int(x) if x.isdigit() else x for x in re.findall(r'[^0-9]|[0-9]+', var)])\n\n middle = len(files)//2\n num_imgs2 = num_imgs//2\n p1 = max(0, middle - num_imgs2)\n p2 = min(len(files), middle + num_imgs2)\n img3d = np.stack([load_dicom_image(f, rotate=rotate) for f in files[p1:p2]]).T \n if img3d.shape[-1] < num_imgs:\n n_zero = np.zeros((img_size, img_size, num_imgs - img3d.shape[-1]))\n img3d = np.concatenate((img3d, n_zero), axis = -1)\n \n if np.min(img3d) < np.max(img3d):\n img3d = img3d - np.min(img3d)\n img3d = img3d / np.max(img3d)\n \n return np.expand_dims(img3d,0)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:33.925628Z","iopub.execute_input":"2021-08-25T09:24:33.925973Z","iopub.status.idle":"2021-08-25T09:24:33.936994Z","shell.execute_reply.started":"2021-08-25T09:24:33.925944Z","shell.execute_reply":"2021-08-25T09:24:33.935704Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated: Functions to load images
**\n","metadata":{}},{"cell_type":"code","source":"def load_dicom_image(path, img_size=SIZE, voi_lut=True, rotate=0):\n #load_dicom_image function provide from the pydicom library. pass path to point at to grab and load the dicom data to pixel data. \n #voi lut is true so we remove area in the image within the given frame of reference, rotate=0 is the degree angle relative to the positive X axis we want to rotate the image, so 0 degrees \n dicom = pydicom.read_file(path)\n #define the dicom variable that is the dcm object from the file within the given path\n data = dicom.pixel_array\n #define the data variable to the pixel_array data of the dicom image(see metadata for available attributes)\n if voi_lut: #if voi_lut is True (which it should be, as we just set it to be)\n #if we can isolate the data to more valuable information by removing window size-do it with apply_voi_lut\n data = apply_voi_lut(dicom.pixel_array, dicom) #\" Value Of Interest LUT to eliminate area's that are not\n #clinically significant (think Window width/level)\"-source: https://groups.google.com/g/comp.protocols.dicom/c/0YAVTRd3BZ0\n else:\n #if somehow the always True we setup in the if fails just give us the pixel_array of the loaded dicom and set the variable data to its value\n data = dicom.pixel_array\n \n #if somehow rotate is note what we set it to use the cv2 library's image handling capability to rotate the pixel_data on its axis by 0 degrees clockwise, so dont rotate it lol\n #a good resource for this: https://www.geeksforgeeks.org/python-opencv-cv2-rotate-method/\n if rotate > 0:\n rot_choices = [0, cv2.ROTATE_90_CLOCKWISE, cv2.ROTATE_90_COUNTERCLOCKWISE, cv2.ROTATE_180]\n #data = pixel_data in the same place it started because we rotate by 0 degrees clockwise\n data = cv2.rotate(data, rot_choices[rotate])\n \n #resize the data using our presets from earlier\n data = cv2.resize(data, (img_size, img_size))\n return data\n\n\n#constructing 3d dicom images with our given meta data, image num(depth), img size(height and width), scan type, split category(train) and with rotate set to 0 or rather, 'no rotation while processing'\ndef load_dicom_images_3d(scan_id, num_imgs=NUM_IMAGES, img_size=SIZE, mri_type=\"FLAIR\", split=\"train\", rotate=0):\n\n files = sorted(glob.glob(f\"{data_directory}/{split}/{scan_id}/{mri_type}/*.dcm\"), \n #we set files equal to a sorted function\n #glob.glob function parses an f string using our input parameters\n \n key=lambda var:[int(x) if x.isdigit() else x for x in re.findall(r'[^0-9]|[0-9]+', var)])\n #we set our key value that will dictate our sorting to a lambda function that utilizes an regexpression\n #logic annotated in expanded form example below\n \n #lambda function in expanded form\n '''\n --define lambda function--\ndef lambda(var):\n ---for x in the contents of the 1d array generated by parsing each individual character grouping via the regex expression and re.findall expression---\n for x in re.findall((r'[^0-9]|[0-9]+', var)):\n ---if x is a digit, return the value as an integer in an array and using the sorted value use that mapped array that was generated at the given-----\n ---values to return a files variable the is in numerical ascending order based on Image-#--------\n if x.isdigit():\n return int(x)\n else:\n return x\n \n ''' \n \n middle = len(files)//2\n #set variable middle to the index at the halfway point of the files array\n \n num_imgs2 = num_imgs//2\n #set num_imgs2 to half of the number of images present\n \n p1 = max(0, middle - num_imgs2)\n #sets p1 to the value of the middle array - run_imgs2 unless num_imgs > than halfway indice in which case it is 0\n \n p2 = min(len(files), middle + num_imgs2)\n #sets p2 to the value of the either the length of the files array or the value of half of the len of + 32(half of 64 from earlier)\n #im assuming the goal here is to create indices by which to iterate through while generating 3D arrays for each respective folder without compromising too much in terms of standardization(someone please correct me if I am wrong)\n \n img3d = np.stack([load_dicom_image(f, rotate=rotate) for f in files[p1:p2]]).T #uses a for loop in a \n #set img3d to the transposition(were flipping the result array) of the stacked array generated by stacking the each dicom image per file between the p1 amount and p2 amount, including p2 value\n #better conceptual explanation:\n '''\n #we are in effect grabbing images that are more likely to have pixel data as they aren't going to be the first few empty frames due to our p1 and p2 setup and\n #then from there we are iterating through the p1 and p2 amounts and loading the dicom data from each respective image, transposing it, and stacking it on the previous image(this is how we get that 3d image)\n i am curious however, why transpose? are the new dimensions more favorable and why?\n '''\n if img3d.shape[-1] < num_imgs:\n #conditional to check for the number of layers the 3d image has and if that is less than the number of images in the folder we used \n \n n_zero = np.zeros((img_size, img_size, num_imgs - img3d.shape[-1]))\n #returns a 3d array of the same dimensions as the generated 3d image that accounts for the missing layers we filtered out earlier\n \n img3d = np.concatenate((img3d, n_zero), axis = -1)\n #we add the 3d image with the blank array along the last axis of the array and remaining dimensions\n \n if np.min(img3d) < np.max(img3d):\n #conditional that normalizes the checks to see if the minimum value is less than the max value in the array(which it should be)\n \n #in these two statements we are normalizing the data contained within the 3d array to values between 0 and 1 via subtracting the im3d element amounts by the mean array element and then dividing the im3d array elements by the max element value in the array\n img3d = img3d - np.min(img3d)\n img3d = img3d / np.max(img3d)\n \n #returns what I believe is a 4d array via expanding the dimensions here, but that dim will remain as 1\n return np.expand_dims(img3d,0)\n\n#where a is an example 3d image object we made at id 00019\na = load_dicom_images_3d(\"00019\")\n\n#(1, 256, 256, 64)\nprint(a.shape)\n\n#0.0 1.0 0.07237920020074569 0.0\nprint(np.min(a), np.max(a), np.mean(a), np.median(a))\n\nimage = a[0]\n#set image to the slice(2d image) at a[0]\n\nprint(\"Dimension of the CT scan is:\", image.shape)\n#Dimension of the CT scan is: (256, 256, 64)\n\nplt.imshow(np.squeeze(image[:, :, 30]), cmap=\"gray\")\n'''comment from the author:\na = load_dicom_images_3d(\"00000\")\nprint(a.shape)--> the output is (1, 256, 256, 64) (channel,width,height,depth) // channel=1 because we have gray images.\nimage = a[0] // I selected first channel already I have one channel\n\nEach three-dimensional region is called a voxel. A typical input is 256x256x64 (width,height,depth) voxels, where 64 is the number of slices and 256x256 is the resolution of each image.\nwhen I selected \"image[:, :, 31]\" to get slice number 31 that is in the middle voxel, you can change the number to get another slice.\n\nGood Luck.\n\n'''\n\n#use the np.squeeze when accessing the image pixel data at a depth of the 30th pixel(?) and remove the unnecessary \n#dimension(we grabbed a single slice from a 3d image so it makes sense we still have 256,256,_ where the _ is the z or depth at which you took that slice)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-20T19:43:07.818302Z","iopub.execute_input":"2021-08-20T19:43:07.818638Z","iopub.status.idle":"2021-08-20T19:43:09.179296Z","shell.execute_reply.started":"2021-08-20T19:43:07.818607Z","shell.execute_reply":"2021-08-20T19:43:09.178226Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**reviewing the dicom variable's data type**","metadata":{}},{"cell_type":"code","source":"path = '../input/rsna-miccai-brain-tumor-radiogenomic-classification/train/00000/FLAIR/Image-1.dcm'\n\ndicom = pydicom.read_file(path)\ntype(dicom) #should return FileDataset which holds the metadata for a given dcm file","metadata":{"execution":{"iopub.status.busy":"2021-08-25T09:24:39.862206Z","iopub.execute_input":"2021-08-25T09:24:39.862545Z","iopub.status.idle":"2021-08-25T09:24:39.88763Z","shell.execute_reply.started":"2021-08-25T09:24:39.862516Z","shell.execute_reply":"2021-08-25T09:24:39.886817Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: Splitting Data
**","metadata":{}},{"cell_type":"code","source":"df_train, df_valid = sk_model_selection.train_test_split(\n train_df, \n test_size=0.2, \n random_state=12, \n stratify=train_df[\"MGMT_value\"],\n)\n\nlen(df_train)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:43.070015Z","iopub.execute_input":"2021-08-25T09:24:43.070343Z","iopub.status.idle":"2021-08-25T09:24:43.083012Z","shell.execute_reply.started":"2021-08-25T09:24:43.070312Z","shell.execute_reply":"2021-08-25T09:24:43.082099Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated: Splitting Data
**\n","metadata":{}},{"cell_type":"code","source":"#use the sklearn split mechanism and stratify train data\ndf_train, df_valid = sk_model_selection.train_test_split(\n train_df, \n test_size=0.2, \n random_state=12, \n stratify=train_df[\"MGMT_value\"],\n)\n\nlen(df_train)\n#pretty sure this was done just to double check his df was correct before proceeding","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: Custom Data Generator
**","metadata":{}},{"cell_type":"code","source":"from keras.utils import Sequence\nclass Dataset(Sequence):\n def __init__(self,df,is_train=True,batch_size=1,shuffle=True):\n self.idx = df[\"BraTS21ID\"].values\n self.paths = df[\"BraTS21ID5\"].values\n self.y = df[\"MGMT_value\"].values\n self.is_train = is_train\n self.batch_size = batch_size\n self.shuffle = shuffle\n def __len__(self):\n return math.ceil(len(self.idx)/self.batch_size)\n \n def __getitem__(self,ids):\n id_path= self.paths[ids]\n batch_paths = self.paths[ids * self.batch_size:(ids + 1) * self.batch_size]\n \n if self.y is not None:\n batch_y = self.y[ids * self.batch_size: (ids + 1) * self.batch_size]\n \n list_x = load_dicom_images_3d(id_path)#str(scan_id).zfill(5)\n #list_x = [load_dicom_images_3d(x) for x in batch_paths]\n batch_X = np.stack(list_x)\n if self.is_train:\n return batch_X,batch_y\n else:\n return batch_X\n \n def on_epoch_end(self):\n if self.shuffle and self.is_train:\n ids_y = list(zip(self.idx, self.y))\n shuffle(ids_y)\n self.idx, self.y = list(zip(*ids_y))","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:46.570253Z","iopub.execute_input":"2021-08-25T09:24:46.570807Z","iopub.status.idle":"2021-08-25T09:24:46.641388Z","shell.execute_reply.started":"2021-08-25T09:24:46.570768Z","shell.execute_reply":"2021-08-25T09:24:46.640611Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated: Custom Data Generator
**\n","metadata":{}},{"cell_type":"code","source":"from keras.utils import Sequence\n#Sequence are a safer way to do multiprocessing. This structure guarantees that the network will only train once on each sample per epoch which is not the case with generators. src: https://www.tensorflow.org/api_docs/python/tf/keras/utils/Sequence\n\nclass Dataset(Sequence):\n # override and add to parent class sequence that returns getiter, iter, and len magic functions\n def __init__(self,df,is_train=True,batch_size=BATCH_SIZE,shuffle=True):\n #set idx to the array of id values\n self.idx = df[\"BraTS21ID\"].values\n #set the paths values to the values inside the ID5 array we created earlier\n self.paths = df[\"BraTS21ID5\"].values\n #set the y values to the values of the true mgmt values \n self.y = df[\"MGMT_value\"].values\n #set is_train to True\n self.is_train = is_train\n #set batch_size to the parameter passed for batchsize\n self.batch_size = batch_size\n #set shuffle to True\n self.shuffle = shuffle\n \n def __len__(self):\n return math.ceil(len(self.idx)/self.batch_size)\n #find the ratio of idx to batch size and round up to find the declared len of this dataset object\n \n def __getitem__(self,ids):\n id_path= self.paths[ids]\n #grabs current id path from paths array at ids\n \n batch_paths = self.paths[ids * self.batch_size:(ids + 1) * self.batch_size]\n #creates and sets batch_paths to the range between the batch size at current ids*batch_size and next ids*batch_size\n \n if self.y is not None:\n batch_y = self.y[ids * self.batch_size: (ids + 1) * self.batch_size]\n #initializing matching y values for corresonding ids\n \n #author comment: list_x = load_dicom_images_3d(id_path) #str(scan_id).zfill(5)\n list_x = [load_dicom_images_3d(x) for x in batch_paths]\n #create a list of 3d dicom images for the number of batch paths(double check needed)\n batch_X = np.stack(list_x, axis=4)\n #create an array on a new axis, 4\n if self.is_train:\n return batch_X,batch_y\n #return the appropriately setup batch for training\n else:\n return batch_X\n \n def on_epoch_end(self):\n if self.shuffle and self.is_train:#which will be true\n ids_y = list(zip(self.idx, self.y))\n #create and set ids_y to a list of pairings via dictionary zipping of the idx and corresponding y\n shuffle(ids_y) #shuffle the order of the pairings within the array\n self.idx, self.y = list(zip(*ids_y))\n #unpacks the pairs and sets the idx and y to be the respective arrays for each with corresponding indexing","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: initializing custom dataset
**\n","metadata":{}},{"cell_type":"code","source":"train_dataset = Dataset(df_train)\nvalid_dataset = Dataset(df_valid)\n#test_dataset = Dataset(test,is_train=False)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:50.321377Z","iopub.execute_input":"2021-08-25T09:24:50.32334Z","iopub.status.idle":"2021-08-25T09:24:50.328214Z","shell.execute_reply.started":"2021-08-25T09:24:50.323298Z","shell.execute_reply":"2021-08-25T09:24:50.327443Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for i in range(1):\n images, label = train_dataset[i]\n print(\"Dimension of the CT scan is:\", images.shape)\n print(\"label=\",label)\n plt.imshow(images[0,:,:,32,0], cmap=\"gray\")\n plt.show()","metadata":{"_kg_hide-input":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated: initializing custom dataset
**","metadata":{}},{"cell_type":"code","source":"#iinitialize datasets for train and valid dfs\ntrain_dataset = Dataset(df_train)\nvalid_dataset = Dataset(df_valid)\n#test_dataset = Dataset(test,is_train=False)\nlen(train_dataset)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#test for that newly generated dataset is accessible and representative of goal dimensions\nfor i in range(1):\n images, label = train_dataset[i]\n #grabs a single image and its label\n print(\"Dimension of the CT scan is:\", images.shape)\n print(\"label=\",label)\n plt.imshow(images[0,:,:,32,0], cmap=\"gray\")\n #our image dimensionality at this point is a bit odd but all that matters it that we are grabbing the image at i in our data set where our axis=4 is at the value 0\n plt.show()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: Model
**\n","metadata":{}},{"cell_type":"code","source":"def get_model(width=256, height=256, depth=64):\n \"\"\"Build a 3D convolutional neural network model.\"\"\"\n\n inputs = keras.Input((width, height, depth, 1))\n\n x = layers.Conv3D(filters=64, kernel_size=3, activation=\"relu\")(inputs)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n\n x = layers.Conv3D(filters=64, kernel_size=3, activation=\"relu\")(x)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n\n x = layers.Conv3D(filters=128, kernel_size=3, activation=\"relu\")(x)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n\n x = layers.Conv3D(filters=256, kernel_size=3, activation=\"relu\")(x)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n\n x = layers.GlobalAveragePooling3D()(x)\n x = layers.Dense(units=512, activation=\"relu\")(x)\n x = layers.Dropout(0.3)(x)\n\n outputs = layers.Dense(units=1, activation=\"sigmoid\")(x)\n\n # Define the model.\n model = keras.Model(inputs, outputs, name=\"3dcnn\")\n\n return model\n\n\n# Build model.\nmodel = get_model(width=256, height=256, depth=64)\nmodel.summary()","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-25T09:24:56.263933Z","iopub.execute_input":"2021-08-25T09:24:56.264517Z","iopub.status.idle":"2021-08-25T09:24:56.651911Z","shell.execute_reply.started":"2021-08-25T09:24:56.264483Z","shell.execute_reply":"2021-08-25T09:24:56.650906Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated: Model
**\n","metadata":{}},{"cell_type":"code","source":"def get_model(width=IMAGE_SIZE, height=IMAGE_SIZE, depth=64):\n #grab voxel data from our data\n \"\"\"Build a 3D convolutional neural network model.\"\"\"\n\n inputs = keras.Input((width, height, depth, 1))\n #setting the shape of our keras object to be our voxel values with the inclusion of 1 for indicating the channel is gray scale(3 would have been RGB)\n #we are effectively stating that we have a depth number of inputs of widthxheight images per input src:https://stackoverflow.com/questions/55572325/how-can-i-use-neutral-network-with-gray-scale-images-in-keras\n #keras docs:https://keras.io/api/layers/core_layers/input/\n \n x = layers.Conv3D(filters=32, kernel_size=3, activation=\"relu\")(inputs)#input specifies what values are being used\n # filters ; Integer, the dimensionality of the output space (i.e. the number of output filters in the convolution)\n #the abstracted frame our model will continue to use to find features\n # kernel_size ; An integer or tuple/list of 3 integers, specifying the depth, height and width of the 3D convolution window. Can be a single integer to specify the same value for all spatial dimensions.\n #indicates the dimensionality of the 3D data we are passing\n #activation='relu' parses input and returns a max value based on threshold(we used default)\n #\"this returns the standard ReLU activation: max(x, 0), the element-wise maximum of 0 and the input tensor.\"-https://keras.io/api/layers/activations/\n\n x = layers.MaxPool3D(pool_size=2)(x)\n #Downsamples the input along its spatial dimensions (depth, height, and width) by taking the maximum value over an input window (of size defined by pool_size) for each channel of the input. \"-https://keras.io/api/layers/pooling_layers/max_pooling3d/\n #pool_size=2 specifies (2,2,2) implicitly thereby acting as the strides along each dimension that they are shifted by\n '''\n x=30\n y=30\n z=30\n color=1\n pool_size=3\n \n results in:\n x=10\n y=10\n z=10\n color=1\n \n '''\n x = layers.BatchNormalization()(x)\n #Batch normalization applies a transformation that maintains the mean output close to 0 and the output standard deviation close to 1.-https://keras.io/api/layers/normalization_layers/batch_normalization/\n #batch normalization is different for training than inference, here it normalizes output using the mean and standard deviation versus in inference it will be calculated using a moving avg and std\n \n x = layers.Conv3D(filters=32, kernel_size=3, activation=\"relu\")(inputs)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n x = layers.Dropout(0.01)(x)\n #The Dropout layer randomly sets input units to 0 with a frequency of rate at each step during training time, which helps prevent overfitting. Inputs not set to 0 are scaled up by 1/(1 - rate) suchh\n #that the sum over all inputs is unchanged.-https://keras.io/api/layers/regularization_layers/dropout/\n\n #repeat layers to increase predictive resoloution. note the dropout scaling to the change in size after previous dropout. filters continually expand and pool size changes\n\n x = layers.Conv3D(filters=64, kernel_size=3, activation=\"relu\")(inputs)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n x = layers.Dropout(0.01)(x)\n \n x = layers.Conv3D(filters=128, kernel_size=3, activation=\"relu\")(x)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n x = layers.Dropout(0.02)(x)\n\n x = layers.Conv3D(filters=256, kernel_size=3, activation=\"relu\")(x)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n x = layers.Dropout(0.03)(x)\n\n x = layers.Conv3D(filters=512, kernel_size=3, activation=\"relu\")(x)\n x = layers.MaxPool3D(pool_size=2)(x)\n x = layers.BatchNormalization()(x)\n x = layers.Dropout(0.04)(x)\n\n x = layers.GlobalAveragePooling3D()(x)\n #default shape input : (batch, spatial_dim1, spatial_dim2, spatial_dim3, channels)\n #default shape output : (batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)\n #reconfigures datastructure(?)\n\n x = layers.Dense(units=1024, activation=\"relu\")(x)\n x = layers.Dropout(0.08)(x)\n\n outputs = layers.Dense(units=1, activation=\"sigmoid\")(x)\n\n # Define the model.\n model = keras.Model(inputs, outputs, name=\"3dcnn\")\n #defining/storing keras model object with its learned parameters\n\n return model\n\n# Build model.\nmodel = get_model(width=IMAGE_SIZE, height=IMAGE_SIZE, depth=64)#generating model\nmodel.summary()#will print the type of all layers in a model, output shape for each layer, number of weight parameter of each layer, model general topology(pretty sure this is loss and similar components)\n #, and the total number of trainable and nontrainable params","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: Training
**\n","metadata":{}},{"cell_type":"code","source":"# Compile model.\ninitial_learning_rate = 0.0005\nlr_schedule = keras.optimizers.schedules.ExponentialDecay(\n initial_learning_rate, decay_steps=100000, decay_rate=0.96, staircase=True\n)\nmodel.compile(\n loss=\"binary_crossentropy\",\n optimizer=keras.optimizers.Adam(learning_rate=lr_schedule),\n metrics=[\"acc\"],\n)\n\n# Define callbacks.\ncheckpoint_cb = keras.callbacks.ModelCheckpoint(\n \"Brain_3d_classification.h5\", save_best_only=True\n)\nearly_stopping_cb = keras.callbacks.EarlyStopping(monitor=\"val_acc\", patience=15)\n\n# Train the model, doing validation at the end of each epoch\nepochs = 50\nmodel.fit(\n train_dataset,\n validation_data=valid_dataset,\n epochs=epochs,\n shuffle=True,\n verbose=2,\n callbacks=[checkpoint_cb, early_stopping_cb],\n)","metadata":{"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> see author's model feedback(currently working with a lot of limitations time wise)\nhttps://www.kaggle.com/ammarnassanalhajali/brain-tumor-3d-training?scriptVersionId=72507803&cellId=20","metadata":{}},{"cell_type":"markdown","source":"#### **
Annotated: Training
**\n","metadata":{}},{"cell_type":"code","source":"# Compile model.\ninitial_learning_rate = 0.0001 #sets the step size to 0.0001 for optimizer\nlr_schedule = keras.optimizers.schedules.ExponentialDecay(\n initial_learning_rate, decay_steps=100000, decay_rate=0.96, staircase=True#the following is true of these params;\n #initial_learning_rate * decay_rate ^ (step / decay_steps) is what generates the step size where the exponent exists if we set staircase=True which we have\n #When fitting a Keras model, decay every 100000 steps with a base of 96-learning_rate_schedules\n #we are effectively generating a decaying step learning rate as we get closer and closer to the minimum for a better chance at getting a min\n)\nmodel.compile( #generates model object with ;\n loss=\"binary_crossentropy\", #binary_crossentropy loss function\n optimizer=keras.optimizers.Adam(learning_rate=lr_schedule), #optimizer Adam with learning_rate prescribed the object lr_schedule \n metrics=[AUC(name='auc'),\"acc\"],# defines AUC metric\n '''quick AUC def:\n \"The Area Under the Curve (AUC) is the measure of the ability of a classifier to distinguish between \n classes and is used as a summary of the ROC curve. The higher the AUC, the better the performance of\n the model at distinguishing between the positive and negative classes\"-https://www.analyticsvidhya.com/blog/2020/06/auc-roc-curve-machine-learning/#:~:text=The%20Area%20Under%20the%20Curve,the%20positive%20and%20negative%20classes.'''\n)\n# Define callbacks.\nmodel_save = ModelCheckpoint('Brain_3d_cls_FLAIR.h5', #prevents overly random training and leverages ability to specify a successful model to iterate more effectively \n save_best_only = True, \n monitor = 'val_auc', \n mode = 'max', verbose = 1)\nearly_stop = EarlyStopping(monitor = 'val_auc', \n patience = 15, mode = 'max', verbose = 1,\n restore_best_weights = True)\n\n# Train the model, doing validation at the end of each epoch\nepochs = 100\nmodel.fit(\n train_dataset,\n validation_data=valid_dataset,\n epochs=epochs,\n shuffle=True,\n verbose=1,#prints training values per epoch if set=1\n callbacks = [model_save, early_stop],\n)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Unannotated: Visualizing model performance
**\n","metadata":{}},{"cell_type":"code","source":"fig, ax = plt.subplots(1, 2, figsize=(20, 7))\nax = ax.ravel()\n\nfor i, metric in enumerate([\"acc\", \"loss\"]):\n ax[i].plot(model.history.history[metric])\n ax[i].plot(model.history.history[\"val_\" + metric])\n ax[i].set_title(\"Model {}\".format(metric))\n ax[i].set_xlabel(\"epochs\")\n ax[i].set_ylabel(metric)\n ax[i].legend([\"train\", \"val\"])","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **
Annotated: Visualizing model performance
**\n","metadata":{}},{"cell_type":"code","source":"fig, ax = plt.subplots(1, 2, figsize=(20, 7))\n#create two figures with width 20 by 7 on one row\n\nax = ax.ravel()\n#create a flattened, 1d array\n\n#iterate through metric values then parse and wrap values for graph beautification\nfor i, metric in enumerate([\"acc\", \"loss\"]):\n ax[i].plot(model.history.history[metric])\n ax[i].plot(model.history.history[\"val_\" + metric])\n ax[i].set_title(\"Model {}\".format(metric))\n ax[i].set_xlabel(\"epochs\")\n ax[i].set_ylabel(metric)\n ax[i].legend([\"train\", \"val\"])","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2021-08-20T19:52:01.874996Z","iopub.status.idle":"2021-08-20T19:52:01.875461Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"![image.png](attachment:f629e45b-769b-4086-a106-fa3c195e3304.png)![image.png](attachment:6828f8c7-7180-4741-a14b-9ff71f09b6ef.png)","metadata":{},"attachments":{"f629e45b-769b-4086-a106-fa3c195e3304.png":{"image/png":"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"},"6828f8c7-7180-4741-a14b-9ff71f09b6ef.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"## images from original source:\n https://www.kaggle.com/ammarnassanalhajali/brain-tumor-3d-training/output","metadata":{}}]}