From 028534a5b9cc5e48b364b07f6a857e4ad1f0058a Mon Sep 17 00:00:00 2001 From: Michal Futrega Date: Wed, 20 Oct 2021 10:03:05 -0700 Subject: [PATCH] [nnUNet/PyT] Add Jupyter notebook with BraTS21 solution --- PyTorch/Segmentation/nnUNet/Dockerfile | 28 +- PyTorch/Segmentation/nnUNet/README.md | 15 +- .../nnUNet/data_loading/dali_loader.py | 73 +- .../nnUNet/data_loading/data_module.py | 3 +- .../nnUNet/data_preprocessing/configs.py | 13 +- .../nnUNet/data_preprocessing/preprocessor.py | 41 +- .../Segmentation/nnUNet/images/unet-brats.jpg | Bin 0 -> 65453 bytes PyTorch/Segmentation/nnUNet/main.py | 28 +- PyTorch/Segmentation/nnUNet/models/layers.py | 25 - PyTorch/Segmentation/nnUNet/models/loss.py | 33 +- PyTorch/Segmentation/nnUNet/models/metrics.py | 68 +- PyTorch/Segmentation/nnUNet/models/nn_unet.py | 117 ++- PyTorch/Segmentation/nnUNet/models/unet.py | 37 +- .../nnUNet/notebooks/BraTS21.ipynb | 746 ++++++++++++++++++ PyTorch/Segmentation/nnUNet/preprocess.py | 5 +- PyTorch/Segmentation/nnUNet/requirements.txt | 13 +- .../Segmentation/nnUNet/scripts/benchmark.py | 2 +- PyTorch/Segmentation/nnUNet/scripts/train.py | 2 + .../Segmentation/nnUNet/utils/scheduler.py | 17 + PyTorch/Segmentation/nnUNet/utils/utils.py | 32 +- 20 files changed, 1064 insertions(+), 234 deletions(-) create mode 100644 PyTorch/Segmentation/nnUNet/images/unet-brats.jpg create mode 100644 PyTorch/Segmentation/nnUNet/notebooks/BraTS21.ipynb create mode 100644 PyTorch/Segmentation/nnUNet/utils/scheduler.py diff --git a/PyTorch/Segmentation/nnUNet/Dockerfile b/PyTorch/Segmentation/nnUNet/Dockerfile index edfa0a3f..a66dd985 100755 --- a/PyTorch/Segmentation/nnUNet/Dockerfile +++ b/PyTorch/Segmentation/nnUNet/Dockerfile @@ -1,30 +1,24 @@ ARG FROM_IMAGE_NAME=nvcr.io/nvidia/pytorch:21.02-py3 -FROM ${FROM_IMAGE_NAME} +FROM ${FROM_IMAGE_NAME} -ADD . /workspace/nnunet_pyt -WORKDIR /workspace/nnunet_pyt +ADD ./triton/requirements.txt / +RUN pip install --disable-pip-version-check -r /requirements.txt +RUN apt-get update && apt-get install -y libb64-dev libb64-0d +ADD ./requirements.txt / RUN pip install --upgrade pip -RUN pip install --disable-pip-version-check -r requirements.txt -RUN pip install --disable-pip-version-check -r triton/requirements.txt -RUN pip install pytorch-lightning==1.0.0 --no-dependencies -RUN pip install torchtext==0.6.0 --no-dependencies -RUN pip install monai==0.4.0 --no-dependencies -RUN pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==0.30.0 -RUN pip install torch_optimizer==0.0.1a15 --no-dependencies -RUN pip install numpy==1.20.3 +RUN pip install --disable-pip-version-check -r /requirements.txt +RUN pip install monai==0.7.0 --no-dependencies +RUN pip install numpy --upgrade RUN pip install nvidia-pyindex==1.0.9 RUN pip install nvidia-dlprof==1.2.0 RUN pip install nvidia_dlprof_pytorch_nvtx==1.2.0 +RUN pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.5.0 RUN curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip" RUN unzip -qq awscliv2.zip RUN ./aws/install RUN rm -rf awscliv2.zip aws -# Install Perf Client required library -RUN apt-get update && apt-get install -y libb64-dev libb64-0d - -# Install Triton Client Python API and copy Perf Client -#COPY --from=triton-client /workspace/install/ /workspace/install/ -#RUN pip install /workspace/install/python/triton*.whl +WORKDIR /workspace/nnunet_pyt +ADD . /workspace/nnunet_pyt diff --git a/PyTorch/Segmentation/nnUNet/README.md b/PyTorch/Segmentation/nnUNet/README.md index 343b2b44..b6b1f5e6 100755 --- a/PyTorch/Segmentation/nnUNet/README.md +++ b/PyTorch/Segmentation/nnUNet/README.md @@ -69,7 +69,7 @@ The following figure shows the architecture of the 3D U-Net model and its differ All convolution blocks in U-Net in both encoder and decoder are using two convolution layers followed by instance normalization and a leaky ReLU nonlinearity. For downsampling we are using stride convolution whereas transposed convolution for upsampling. -All models were trained with RAdam optimizer, learning rate 0.001 and weight decay 0.0001. For loss function we use the average of [cross-entropy](https://en.wikipedia.org/wiki/Cross_entropy) and [dice coefficient](https://en.wikipedia.org/wiki/S%C3%B8rensen%E2%80%93Dice_coefficient). +All models were trained with Adam optimizer, learning rate 0.0008 and weight decay 0.0001. For loss function we use the average of [cross-entropy](https://en.wikipedia.org/wiki/Cross_entropy) and [dice coefficient](https://en.wikipedia.org/wiki/S%C3%B8rensen%E2%80%93Dice_coefficient). Early stopping is triggered if validation dice score wasn't improved during the last 100 epochs. @@ -304,7 +304,7 @@ To see the full list of available options and their descriptions, use the `-h` o The following example output is printed when running the model: ``` -usage: main.py [-h] [--exec_mode {train,evaluate,predict}] [--data DATA] [--results RESULTS] [--logname LOGNAME] [--task TASK] [--gpus GPUS] [--learning_rate LEARNING_RATE] [--gradient_clip_val GRADIENT_CLIP_VAL] [--negative_slope NEGATIVE_SLOPE] [--tta] [--amp] [--benchmark] [--residual] [--focal] [--sync_batchnorm] [--save_ckpt] [--nfolds NFOLDS] [--seed SEED] [--skip_first_n_eval SKIP_FIRST_N_EVAL] [--ckpt_path CKPT_PATH] [--fold FOLD] [--patience PATIENCE] [--lr_patience LR_PATIENCE] [--batch_size BATCH_SIZE] [--val_batch_size VAL_BATCH_SIZE] [--steps STEPS [STEPS ...]] [--profile] [--momentum MOMENTUM] [--weight_decay WEIGHT_DECAY] [--save_preds] [--dim {2,3}] [--resume_training] [--factor FACTOR] [--num_workers NUM_WORKERS] [--min_epochs MIN_EPOCHS] [--max_epochs MAX_EPOCHS] [--warmup WARMUP] [--norm {instance,batch,group}] [--nvol NVOL] [--data2d_dim {2,3}] [--oversampling OVERSAMPLING] [--overlap OVERLAP] [--affinity {socket,single,single_unique,socket_unique_interleaved,socket_unique_continuous,disabled}] [--scheduler {none,multistep,cosine,plateau}] [--optimizer {sgd,radam,adam}] [--blend {gaussian,constant}] [--train_batches TRAIN_BATCHES] [--test_batches TEST_BATCHES] +usage: main.py [-h] [--exec_mode {train,evaluate,predict}] [--data DATA] [--results RESULTS] [--logname LOGNAME] [--task TASK] [--gpus GPUS] [--learning_rate LEARNING_RATE] [--gradient_clip_val GRADIENT_CLIP_VAL] [--negative_slope NEGATIVE_SLOPE] [--tta] [--amp] [--benchmark] [--residual] [--focal] [--sync_batchnorm] [--save_ckpt] [--nfolds NFOLDS] [--seed SEED] [--skip_first_n_eval SKIP_FIRST_N_EVAL] [--ckpt_path CKPT_PATH] [--fold FOLD] [--patience PATIENCE] [--lr_patience LR_PATIENCE] [--batch_size BATCH_SIZE] [--val_batch_size VAL_BATCH_SIZE] [--steps STEPS [STEPS ...]] [--profile] [--momentum MOMENTUM] [--weight_decay WEIGHT_DECAY] [--save_preds] [--dim {2,3}] [--resume_training] [--factor FACTOR] [--num_workers NUM_WORKERS] [--min_epochs MIN_EPOCHS] [--max_epochs MAX_EPOCHS] [--warmup WARMUP] [--norm {instance,batch,group}] [--nvol NVOL] [--data2d_dim {2,3}] [--oversampling OVERSAMPLING] [--overlap OVERLAP] [--affinity {socket,single,single_unique,socket_unique_interleaved,socket_unique_continuous,disabled}] [--scheduler {none,multistep,cosine,plateau}] [--optimizer {sgd,adam}] [--blend {gaussian,constant}] [--train_batches TRAIN_BATCHES] [--test_batches TEST_BATCHES] optional arguments: -h, --help show this help message and exit @@ -370,8 +370,8 @@ optional arguments: type of CPU affinity (default: socket_unique_interleaved) --scheduler {none,multistep,cosine,plateau} Learning rate scheduler (default: none) - --optimizer {sgd,radam,adam} - Optimizer (default: radam) + --optimizer {sgd,adam} + Optimizer (default: adam) --blend {gaussian,constant} How to blend output of overlapping windows (default: gaussian) --train_batches TRAIN_BATCHES @@ -418,7 +418,7 @@ If you have dataset in other format or you want customize data preprocessing or ### Training process The model trains for at least `--min_epochs` and at most `--max_epochs` epochs. After each epoch evaluation, the validation set is done and validation loss is monitored for early stopping (see `--patience` flag). Default training settings are: -* RAdam optimizer with learning rate of 0.001 and weight decay 0.0001. +* Adam optimizer with learning rate of 0.0008 and weight decay 0.0001. * Training batch size is set to 2 for 3D U-Net and 16 for 2D U-Net. This default parametrization is applied when running scripts from the `scripts/` directory and when running `main.py` without explicitly overriding these parameters. By default, the training is in full precision. To enable AMP, pass the `--amp` flag. AMP can be enabled for every mode of execution. @@ -454,8 +454,6 @@ The script will then: ## Performance -The performance measurements in this document were conducted at the time of publication and may not reflect the performance achieved from NVIDIA’s latest software release. For the most up-to-date performance measurements, go to [NVIDIA Data Center Deep Learning Product Performance](https://developer.nvidia.com/deep-learning-performance-training-inference). - ### Benchmarking The following section shows how to run benchmarks to measure the model performance in training and inference modes. @@ -633,6 +631,9 @@ To achieve these same results, follow the steps in the [Quick Start Guide](#quic ### Changelog +October 2021 +- Add Jupyter Notebook with BraTS solution + May 2021 - Add Triton Inference Server support - Removed deep supervision, attention and drop block diff --git a/PyTorch/Segmentation/nnUNet/data_loading/dali_loader.py b/PyTorch/Segmentation/nnUNet/data_loading/dali_loader.py index 1fcbaaed..b694ce66 100644 --- a/PyTorch/Segmentation/nnUNet/data_loading/dali_loader.py +++ b/PyTorch/Segmentation/nnUNet/data_loading/dali_loader.py @@ -25,7 +25,7 @@ from nvidia.dali.plugin.pytorch import DALIGenericIterator def get_numpy_reader(files, shard_id, num_shards, seed, shuffle): - return ops.NumpyReader( + return ops.readers.Numpy( seed=seed, files=files, device="cpu", @@ -42,6 +42,7 @@ class TrainPipeline(Pipeline): def __init__(self, batch_size, num_threads, device_id, **kwargs): super(TrainPipeline, self).__init__(batch_size, num_threads, device_id) self.dim = kwargs["dim"] + self.internal_seed = kwargs["seed"] self.oversampling = kwargs["oversampling"] self.input_x = get_numpy_reader( num_shards=kwargs["gpus"], @@ -60,6 +61,7 @@ class TrainPipeline(Pipeline): self.patch_size = kwargs["patch_size"] if self.dim == 2: self.patch_size = [kwargs["batch_size_2d"]] + self.patch_size + self.crop_shape = types.Constant(np.array(self.patch_size), dtype=types.INT64) self.crop_shape_float = types.Constant(np.array(self.patch_size), dtype=types.FLOAT) @@ -69,7 +71,7 @@ class TrainPipeline(Pipeline): return img, lbl def random_augmentation(self, probability, augmented, original): - condition = fn.cast(fn.coin_flip(probability=probability), dtype=types.DALIDataType.BOOL) + condition = fn.cast(fn.random.coin_flip(probability=probability), dtype=types.DALIDataType.BOOL) neg_condition = condition ^ True return condition * augmented + neg_condition * original @@ -77,45 +79,60 @@ class TrainPipeline(Pipeline): def slice_fn(img): return fn.slice(img, 1, 3, axes=[0]) - def crop_fn(self, img, lbl): - center = fn.segmentation.random_mask_pixel(lbl, foreground=fn.coin_flip(probability=self.oversampling)) - crop_anchor = self.slice_fn(center) - self.crop_shape // 2 - adjusted_anchor = math.max(0, crop_anchor) - max_anchor = self.slice_fn(fn.shapes(lbl)) - self.crop_shape - crop_anchor = math.min(adjusted_anchor, max_anchor) - img = fn.slice(img.gpu(), crop_anchor, self.crop_shape, axis_names="DHW", out_of_bounds_policy="pad") - lbl = fn.slice(lbl.gpu(), crop_anchor, self.crop_shape, axis_names="DHW", out_of_bounds_policy="pad") - return img, lbl + def biased_crop_fn(self, img, label): + roi_start, roi_end = fn.segmentation.random_object_bbox( + label, + format="start_end", + foreground_prob=self.oversampling, + background=0, + seed=self.internal_seed, + device="cpu", + cache_objects=True, + ) + + anchor = fn.roi_random_crop(label, roi_start=roi_start, roi_end=roi_end, crop_shape=[1, *self.patch_size]) + anchor = fn.slice(anchor, 1, 3, axes=[0]) # drop channels from anchor + img, label = fn.slice( + [img, label], anchor, self.crop_shape, axis_names="DHW", out_of_bounds_policy="pad", device="cpu" + ) + + return img.gpu(), label.gpu() def zoom_fn(self, img, lbl): - resized_shape = self.crop_shape * self.random_augmentation(0.15, fn.uniform(range=(0.7, 1.0)), 1.0) - img, lbl = fn.crop(img, crop=resized_shape), fn.crop(lbl, crop=resized_shape) + scale = self.random_augmentation(0.15, fn.random.uniform(range=(0.7, 1.0)), 1.0) + d, h, w = [scale * x for x in self.patch_size] + if self.dim == 2: + d = self.patch_size[0] + img, lbl = fn.crop(img, crop_h=h, crop_w=w, crop_d=d), fn.crop(lbl, crop_h=h, crop_w=w, crop_d=d) img = fn.resize(img, interp_type=types.DALIInterpType.INTERP_CUBIC, size=self.crop_shape_float) lbl = fn.resize(lbl, interp_type=types.DALIInterpType.INTERP_NN, size=self.crop_shape_float) return img, lbl def noise_fn(self, img): - img_noised = img + fn.random.normal(img, stddev=fn.uniform(range=(0.0, 0.33))) + img_noised = img + fn.random.normal(img, stddev=fn.random.uniform(range=(0.0, 0.33))) return self.random_augmentation(0.15, img_noised, img) def blur_fn(self, img): - img_blured = fn.gaussian_blur(img, sigma=fn.uniform(range=(0.5, 1.5))) - return self.random_augmentation(0.15, img_blured, img) + img_blurred = fn.gaussian_blur(img, sigma=fn.random.uniform(range=(0.5, 1.5))) + return self.random_augmentation(0.15, img_blurred, img) def brightness_fn(self, img): - brightness_scale = self.random_augmentation(0.15, fn.uniform(range=(0.7, 1.3)), 1.0) + brightness_scale = self.random_augmentation(0.15, fn.random.uniform(range=(0.7, 1.3)), 1.0) return img * brightness_scale def contrast_fn(self, img): min_, max_ = fn.reductions.min(img), fn.reductions.max(img) - scale = self.random_augmentation(0.15, fn.uniform(range=(0.65, 1.5)), 1.0) + scale = self.random_augmentation(0.15, fn.random.uniform(range=(0.65, 1.5)), 1.0) img = math.clamp(img * scale, min_, max_) return img def flips_fn(self, img, lbl): - kwargs = {"horizontal": fn.coin_flip(probability=0.33), "vertical": fn.coin_flip(probability=0.33)} + kwargs = { + "horizontal": fn.random.coin_flip(probability=0.33), + "vertical": fn.random.coin_flip(probability=0.33), + } if self.dim == 3: - kwargs.update({"depthwise": fn.coin_flip(probability=0.33)}) + kwargs.update({"depthwise": fn.random.coin_flip(probability=0.33)}) return fn.flip(img, **kwargs), fn.flip(lbl, **kwargs) def transpose_fn(self, img, lbl): @@ -124,7 +141,7 @@ class TrainPipeline(Pipeline): def define_graph(self): img, lbl = self.load_data() - img, lbl = self.crop_fn(img, lbl) + img, lbl = self.biased_crop_fn(img, lbl) img, lbl = self.zoom_fn(img, lbl) img, lbl = self.flips_fn(img, lbl) img = self.noise_fn(img) @@ -141,15 +158,15 @@ class EvalPipeline(Pipeline): super(EvalPipeline, self).__init__(batch_size, num_threads, device_id) self.input_x = get_numpy_reader( files=kwargs["imgs"], - shard_id=device_id, - num_shards=kwargs["gpus"], + shard_id=0, + num_shards=1, seed=kwargs["seed"], shuffle=False, ) self.input_y = get_numpy_reader( files=kwargs["lbls"], - shard_id=device_id, - num_shards=kwargs["gpus"], + shard_id=0, + num_shards=1, seed=kwargs["seed"], shuffle=False, ) @@ -281,6 +298,12 @@ def fetch_dali_loader(imgs, lbls, batch_size, mode, **kwargs): imgs = list(itertools.chain(*(100 * [imgs])))[: nbs * kwargs["gpus"]] lbls = list(itertools.chain(*(100 * [lbls])))[: nbs * kwargs["gpus"]] + if mode == "eval": # To avoid padding for the multigpu evaluation. + rank = int(os.getenv("LOCAL_RANK", "0")) + imgs, lbls = np.array_split(imgs, kwargs["gpus"]), np.array_split(lbls, kwargs["gpus"]) + imgs, lbls = [list(x) for x in imgs], [list(x) for x in lbls] + imgs, lbls = imgs[rank], lbls[rank] + pipe_kwargs = { "imgs": imgs, "lbls": lbls, diff --git a/PyTorch/Segmentation/nnUNet/data_loading/data_module.py b/PyTorch/Segmentation/nnUNet/data_loading/data_module.py index 6b6dd22c..d5121f62 100644 --- a/PyTorch/Segmentation/nnUNet/data_loading/data_module.py +++ b/PyTorch/Segmentation/nnUNet/data_loading/data_module.py @@ -57,8 +57,7 @@ class DataModule(LightningDataModule): self.val_imgs = get_split(imgs, val_idx) self.val_lbls = get_split(lbls, val_idx) if is_main_process(): - ntrain, nval = len(self.train_imgs), len(self.val_imgs) - print(f"Number of examples: Train {ntrain} - Val {nval}") + print(f"Number of examples: Train {len(self.train_imgs)} - Val {len(self.val_imgs)}") elif is_main_process(): print(f"Number of test examples: {len(self.test_imgs)}") diff --git a/PyTorch/Segmentation/nnUNet/data_preprocessing/configs.py b/PyTorch/Segmentation/nnUNet/data_preprocessing/configs.py index 54835cc4..ff4e5236 100644 --- a/PyTorch/Segmentation/nnUNet/data_preprocessing/configs.py +++ b/PyTorch/Segmentation/nnUNet/data_preprocessing/configs.py @@ -23,6 +23,8 @@ task = { "08": "Task08_HepaticVessel", "09": "Task09_Spleen", "10": "Task10_Colon", + "11": "BraTS2021_train", + "12": "BraTS2021_val", } patch_size = { @@ -36,6 +38,8 @@ patch_size = { "08_3d": [64, 192, 192], "09_3d": [64, 192, 160], "10_3d": [56, 192, 160], + "11_3d": [128, 128, 128], + "12_3d": [128, 128, 128], "01_2d": [192, 160], "02_2d": [320, 256], "03_2d": [512, 512], @@ -60,7 +64,8 @@ spacings = { "08_3d": [1.5, 0.8, 0.8], "09_3d": [1.6, 0.79, 0.79], "10_3d": [3, 0.78, 0.78], - "11_3d": [5, 0.741, 0.741], + "11_3d": [1.0, 1.0, 1.0], + "12_3d": [1.0, 1.0, 1.0], "01_2d": [1.0, 1.0], "02_2d": [1.25, 1.25], "03_2d": [0.7676, 0.7676], @@ -80,7 +85,6 @@ ct_min = { "08": -3, "09": -41, "10": -30, - "11": -958, } ct_max = { @@ -90,9 +94,8 @@ ct_max = { "08": 243, "09": 176, "10": 165.82, - "11": 93, } -ct_mean = {"03": 99.4, "06": -158.58, "07": 77.9, "08": 104.37, "09": 99.29, "10": 62.18, "11": -547.7} +ct_mean = {"03": 99.4, "06": -158.58, "07": 77.9, "08": 104.37, "09": 99.29, "10": 62.18} -ct_std = {"03": 39.36, "06": 324.7, "07": 75.4, "08": 52.62, "09": 39.47, "10": 32.65, "11": 281.08} +ct_std = {"03": 39.36, "06": 324.7, "07": 75.4, "08": 52.62, "09": 39.47, "10": 32.65} diff --git a/PyTorch/Segmentation/nnUNet/data_preprocessing/preprocessor.py b/PyTorch/Segmentation/nnUNet/data_preprocessing/preprocessor.py index ece07c98..785c198b 100644 --- a/PyTorch/Segmentation/nnUNet/data_preprocessing/preprocessor.py +++ b/PyTorch/Segmentation/nnUNet/data_preprocessing/preprocessor.py @@ -22,7 +22,6 @@ import monai.transforms as transforms import nibabel import numpy as np from joblib import Parallel, delayed -from skimage.morphology import dilation, erosion, square from skimage.transform import resize from utils.utils import get_task_code, make_empty_dir @@ -39,32 +38,34 @@ class Preprocessor: self.target_spacing = None self.task = args.task self.task_code = get_task_code(args) + self.verbose = args.verbose self.patch_size = patch_size[self.task_code] self.training = args.exec_mode == "training" self.data_path = os.path.join(args.data, task[args.task]) + metadata_path = os.path.join(self.data_path, "dataset.json") + self.metadata = json.load(open(metadata_path, "r")) + self.modality = self.metadata["modality"]["0"] self.results = os.path.join(args.results, self.task_code) if not self.training: self.results = os.path.join(self.results, self.args.exec_mode) self.crop_foreg = transforms.CropForegroundd(keys=["image", "label"], source_key="image") - self.normalize_intensity = transforms.NormalizeIntensity(nonzero=False, channel_wise=True) - metadata_path = os.path.join(self.data_path, "dataset.json") + nonzero = True if self.modality != "CT" else False # normalize only non-zero region for MRI + self.normalize_intensity = transforms.NormalizeIntensity(nonzero=nonzero, channel_wise=True) if self.args.exec_mode == "val": dataset_json = json.load(open(metadata_path, "r")) dataset_json["val"] = dataset_json["training"] with open(metadata_path, "w") as outfile: json.dump(dataset_json, outfile) - self.metadata = json.load(open(metadata_path, "r")) - self.modality = self.metadata["modality"]["0"] def run(self): make_empty_dir(self.results) - print(f"Preprocessing {self.data_path}") try: self.target_spacing = spacings[self.task_code] except: self.collect_spacings() - print(f"Target spacing {self.target_spacing}") + if self.verbose: + print(f"Target spacing {self.target_spacing}") if self.modality == "CT": try: @@ -77,7 +78,8 @@ class Preprocessor: _mean = round(self.ct_mean, 2) _std = round(self.ct_std, 2) - print(f"[CT] min: {self.ct_min}, max: {self.ct_max}, mean: {_mean}, std: {_std}") + if self.verbose: + print(f"[CT] min: {self.ct_min}, max: {self.ct_max}, mean: {_mean}, std: {_std}") self.run_parallel(self.preprocess_pair, self.args.exec_mode) @@ -86,7 +88,7 @@ class Preprocessor: "patch_size": self.patch_size, "spacings": self.target_spacing, "n_class": len(self.metadata["labels"]), - "in_channels": len(self.metadata["modality"]), + "in_channels": len(self.metadata["modality"]) + int(self.args.ohe), }, open(os.path.join(self.results, "config.pkl"), "wb"), ) @@ -99,9 +101,10 @@ class Preprocessor: image, label = data["image"], data["label"] test_metadata = None else: + orig_shape = image.shape[1:] bbox = transforms.utils.generate_spatial_bounding_box(image) - test_metadata = np.vstack([bbox, image.shape[1:]]) image = transforms.SpatialCrop(roi_start=bbox[0], roi_end=bbox[1])(image) + test_metadata = np.vstack([bbox, orig_shape, image.shape[1:]]) if label is not None: label = transforms.SpatialCrop(roi_start=bbox[0], roi_end=bbox[1])(label) if self.args.dim == 3: @@ -111,11 +114,16 @@ class Preprocessor: image = self.normalize(image) if self.training: image, label = self.standardize(image, label) - if self.args.dilation: - new_lbl = np.zeros(label.shape, dtype=np.uint8) - for depth in range(label.shape[1]): - new_lbl[0, depth] = erosion(dilation(label[0, depth], square(3)), square(3)) - label = new_lbl + + if self.args.ohe: + mask = np.ones(image.shape[1:], dtype=np.float32) + for i in range(image.shape[0]): + zeros = np.where(image[i] <= 0) + mask[zeros] *= 0.0 + image = self.normalize_intensity(image).astype(np.float32) + mask = np.expand_dims(mask, 0) + image = np.concatenate([image, mask]) + self.save(image, label, fname, test_metadata) def resample(self, image, label, image_spacings): @@ -145,7 +153,8 @@ class Preprocessor: def save(self, image, label, fname, test_metadata): mean, std = np.round(np.mean(image, (1, 2, 3)), 2), np.round(np.std(image, (1, 2, 3)), 2) - print(f"Saving {fname} shape {image.shape} mean {mean} std {std}") + if self.verbose: + print(f"Saving {fname} shape {image.shape} mean {mean} std {std}") self.save_npy(image, fname, "_x.npy") if label is not None: self.save_npy(label, fname, "_y.npy") diff --git a/PyTorch/Segmentation/nnUNet/images/unet-brats.jpg b/PyTorch/Segmentation/nnUNet/images/unet-brats.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b93626e71a60a5be1c10615b74d8debc81b34bbd GIT binary patch literal 65453 zcmeFZcU)83x-L8c0V#qYAU!BmX-e-=5fG3j(wj=}AiV}rK)Mt`x^$4a0{JB)bG&1Ww?5Apmm0)oqg#1vPEiHL}4$gh!7FwjC68EEP0nOJz(nQn41)6;VZ z-{QJ+S3p1j!Y(Q)!Y9GQFTnTnORxwC2#AS@sjpn2=DR_EgYSR&foTQEE@KDdRO4XX z1hC1laLBMQ9RLLECoa}+58(F)7B9!g8EYpO)YI5U1JkdGjj_|D<@|cS2uSL&)07P0)yTLheXH3zK@Il zkdT;>nU$TBo0tEoyrQzIx~8_SzOB8Zv#YzO_sh`m$mrPk#3X!vVR7ll^2+Mk`riJ* z;nDF4^7QPdTv!0kKg0rm|3k1}(c%=}MwDPO$&PDsTfFweU8Q?%bC`=1lc|NoO@|3k3pn@g2^#dX@O&4^##T=dzBq^yWCw z`*O!Z@y8{bC7TH7<_=P!>{#W7ciQQt?eh=aGb_119kqOzE7I4F;J;GMYMB9}3q-;ZvG_zO6v%$8GDId*gpb5oiwdb%?4o@SL ze@V-sZMG{VKQc6^jY3W4{(Tt=RjmI`mwr&h=9UYAx1hC`?Unw@!~*+36O!;y#b`cK z{CA}D20ky*cQ61rj6|>r4GeG?v7GGoAnxme3pbr&d743K<*?0MMbbG@e@zIk_R=)$ zX7vvod?vtYM6-1l9iD|OX**b}mkCJj&wCWkT2(IF>D%kzpzx~ep2#t!mL3LpS6zYu z97s_jeR}a2V4u%j3*Gh-w#LwJH5NAbNn%C%du5)t_`D*r5+TqiOsh^8@ao~4%H3`8 zvgG9Cv!yZNw}*WMwvFl@tl<^YL{0K<`pM%jDe;hOnls}-6xJ#uPp>tyV*rE7NE8J3 zh5KvteZ$)rprlA01IYF3Q^$>HG=wyi8|HooXV*Ci*E?kaa!MzN0nQ0$V5=u3h1N?u zS1~}(B-qOd%>%Ge(@SGmV3SJ5eA-DJEC6kB48ena6-IX|K`?+F2L||n0Z{$0^PtSn zX$My@K$*H620(;Bm!TNI$p}5a@y&=(>iNavRc;Khl)eT#CTzw4->ctGXhG~T04u2? z1}N1yL5Eisx8=ad9FPw}IPy3+GD3j=P8VnVUeE;$sfTU@hc(Mha72rDkrasmc56@p zMi}7Hwq>{TEt>(Ozgr*AH(>izqZnWt3j^fRSii&oXjv6PaF&!+P|S5jhF=FiiP$ou zAMOxffN@)9aOOU1r!s*HPAjBkW8*0}=m(lJQ5fK++Zzn9Q4Ri=IajE171qPqX>@e- z3ATD#@S1YBdnECkSA8{vBcooatyy`G^ddYFoZtX^=%Ls6nBc$KvX*vmEjbLHPqnmR z00A%3zcfVO=LSy$!LH_KVt}nq$N{)?>VB%h(JF5M?|)x(x^TqD>jLSd6L zvIYZ02W5jzZl7ZSS-947b?Dn#?;A1{Pbt8K3;cJwDEpA@Xy^r~AJ4Tgz>s+2`Y+2x zV6mQ}+MKrtR7*YZ6Ses0WYeNP3~)*#!3j1Jg@Q(^^r+>O$O;3T02p96tf~%%0hrkX zV4Ljh=%}f&5;OBrdZE9ceEUZjAWFOqc7y=``M#^k{ua7KxlnR;HVQrT9{n7GG+tNo ziO)Zjp-?kmn>S*24?@fK!_aLtX=jof8?>k59ngItYYcEq*$B4Z(f=?zU!uNRYmVAB znujGy2K}{pSlu%O zeZU25%UPPA)}b$TH6x^J4?jd;_KKYC@uGOY^AGN=t#r%m*jI7z5CSF=`7Kqo3K^#p zL&0yh5=_mvq7Pvv_4x7+lADzQqH$OE?y(x~6~zLJV&=j4+xYrn4zeh14` zK6=VS#oMt)>*Q}gzLGM7Zjnv;^qWs#JlH*oYjh6IP3xB*g6oV;e`>42ZEu(Op5eC=&s%$%T&yK>RgQaVFaaeHp8uZMWUT+J1hv|>kBU!F%?hw6kjCG(U&Ijf~#v7Zeh6{f*G}T7Tz*(nR%1brW z*5yAEtQH%gsqkK5fXhwsZ>jWxDqCXPE$JQj`^C3S;dyxSr%JN3WQ~%j80$P4)+r2N zm9n)^6*wkW%a_X&ct;rfqshuy2(ZENNldE5xWrQEEGJ1g#nPc6)0-@r{4&%2WhQ^g zPe=brl~ms~ba8NN%hI;+P<(7?DNg&vNINI9fQi2XPMzwzd&VQ*@Qr0&x?+H<0u5<& zR)+KsC6)^8FfV_#*K-6Z24Ba zRb?QxFf+L+F)`~jh%O+0;Tz7wx4I+Z zOYOV@PgsCLEPNTinU6}1G8}hBbIeyd-tbU0efPdcpF}&=-%l)&Git=!(#7_9wn{G2 z^+YWZJ-iig)8)~87>9AaZ0n!|NiJo)$DZzYuEOogX~ykSwqZKW4<5l}{?mHakPjDM zFGD##l0L{Nzp*D9#GPK}qd9=9=)P^Lv+~fw*nu$q(nVd*hK_DzQg5KXe zaM0gc(<(21sl55WkEQfH;;xEwmtzf9sj24E_ogpH5ogO+Kzm{_YF^2B-* zjGVwoW=)vzR|9FJJ>ENW1GKpnVAo2ER?v~2OI;Y?if5WkJ9Gzh??e~hH>HTd0LvZ~ zk_bglsbPS-V4McJ;}cT3zjU*|#ry5Zgl}LR@nbe0AIkI^^h zcLO}qI*NGQ_Ot@4f|*BW6{{6L@FdC4u5=j44hea;lHYlB##pr$`;@|+r)7TjE$xb5 zVX=#%w!P-~<0KT-kJ&9lCw|;_8TqqJp%y?Yf!*N_m8IB^%|!io7WF|%_BjtF?p3pX z=P7DWznXrVsE2~wv4@6nbcaF}KgB)sVRe$$u32u@8m{*GJAt8Y8;v2#VE>@S+Xbzk zyEq{ox{dYSk(opG0^DX0(g}lZe!=3zP2TONy93D9aFub9!?le{Rs0~mHP25M_amfp zt^5z&qZJdl6LC<<)!7P30V0THZ1Jb$+&67GS6D`Rm$%pt72p`!>vZ^z=hes2co4-=@ zGLC<}nYDb6C(~&j_0ZjaQ6YNlOV^@``<89dYtm`XJ9JB;)7!H6*iBzTfL$M`Qdj6g z6x;g+ZbjuZGRxhou*AF3WlyocL}>fplK?oU@;;I%&c81@ zif`ZIL8w+(%V&~Mu`5MZcv12+C5K(xR&uFSyN65hpYU?edA*A;8mjW*)H9H}BDz^z zoF&B+Vd7dFBJ(Wx$BH9UX^#z7&Fg+I0}8CAM0`I9rFoU>GVbwFR`ovBkzAL!D;=ta zD+4ZK>K{n+Gb~GQi;uPhN?jRiypcc4lbr*n7Ua-`i4kC(Gn;^l{H%->&z(8!?D&X~ z182~ta+U%Kli7=f8z#}iv(aUO1F4aGxF!yMqP?wM)N=>jO-}ESavdj9xJ|>w5)yl} zieC0^c{E2NOEKE29bb~hvG2C|@rTiSu1s2dB&BD>+q;A&Z2tD*dq(MN)zLR&rM)l9 zVI~iRVwj#y^1ged!A=(`_|3AJeslpT4x3MK&f`AfD752BCwr95Bmaczo>SNUOSfk$ zf~(G|_#fO$l=hOFslG`-v(|V8KU79YYPWk)kl)cDe6DHSDkA!fl-_epE5!SH;Rfe3 zF#CAhEtuuWuLEDp&%22M%=GO|MA-PswmcXxDmz<0Ezmc-6tEVci%Ypk))xZqMYL3}pT~ zZ~yVPaWP{J2Do8sgr0+(ImXrSf*GpGZVBqm(r;n-=;qh6%)lY&lJagE%5?~gf+>R+ zNKbLy3((C~DkzAc?HXjC-CYIUUoGV5M1}U4x0R?z6;(1xl$xQ`gzOoZ41!qJIY;v;%k zGbSa9W*Y8>oiC(70}n$BIJj1ZY*?#aqaO3=F|)w7^Qj>LfR(@=oA9H8)C7_K8+ETt z$!Idk8Q*G~ped7F=21NRdLBHLeYa+&u!t1x!i;e~K9Pe2$g_n$=lK}FudUm(U~Jp| zAm@I>2OBd#UZM}Q>Q@u_nbuzg%FQb&_L*~$yMOrD2h7cRjfxy9UM`tIixWL<^xm%3 ze)`C1oI7kt&%)Cu>FY7vC+mvP!yFEc#|h|Td}uhJr}SXpQR>tBSD7k}8%#HRbE_b5 zDwz2JdK5n<*(3ECBJJ_vr!tE~Z>k}L%yVmcatu%yd_%S5i28HpIjPF|^$GN6>th2% zV;ZZg`MBNPM4*f<|2Z*PKlfI-4EvPp`v*1ORuXlp#OYka%^GXMvuN(^=W$7vd*ngn z%Hw84SSqUBjnAVh6*PQb(?99Y*3}oV70hl7zuI;^T;3>0jslB+UGuI1cVT?on!^O* z+9RCgGvehaK?5npu8N4W$?bUYk2+6lyFaZ6(pkn0y7DB#uQ`Y*7}EtUjwRQ76F8V_ z-@L|Heu*w}am??sppVdxA9`NrI=WXM`-{F0A@ydwRko4%cx+g|Gk>!2Dy-me>7`S{$G(P66&78-v4UZb z-v)Ag$(_cP#UJv!SwDX0b&DxtN2Bc3HgKn~=5o$GPs-kv-CV_D%RB40auxF@vkw++ zE0i;22Osmiv@*6Tyl<)=k-%_EsXTs0!_9Ls5h!77%2LX&PNK>;IoU`VD{#nBi>Q0N z5}D06+~LW35l=R`LF?0wz#c9X=2&JDivuoN!u6;=Yrg~7^KpvAd!;I|((D&}@V2|W zHf*hx6NhF|&U@3$a@WtllHf1OU9g;8KQ=l}J8p_9ovqlq>9T*~X*a;HbrB)|S1O$=%{nO3ibY;}xWuiGUb-gd$yaC_g+1V3GEN@It8 z=H2H(T_B&QCHwv-wgm1A0&gS&3zoc+OCeLTroJATBLl;kl}q`Cf7&IsrzMf<`F-gC zyb;_z^ifOng1FwQnYF{?bP`vraRCI3eDy?=f9A6f9RiF4i{6DC1X zF~FN7IdmU-=-k(l9b{jehjq7%s!;@RoWt3yE*q2gFO?Y+(dQD3r%~K-6 zDG|t}YRc{ZB$Y&8Y!@RQ+cxJozRYaDp;Wm?lH|kMrq|CbSj)|}%HV*Xb8m1yu?3VU zruXC{@dk|CvIZ33!3o}FsZ*iy{0oX=y@-BzH%k)_)gVeE|honQqF57=ViYs zUuSn#wh-j=r<3Sg60Z`0bjt-I@kt(Xj*=0CzkYx2;8hXynllzUG|{%@?Pjr!d1C~_ zLpl6e9f6IvOVI61CNm7+_gdHRQD!goN8(xJqatWI2B?(k&pXvS_JBQ0>K{ZoAsQHw zo!1#}R#unQGHTmXF!j>X zj*z4e&ppV2TbCKiJA|6TAnypc1uK}mVQ!Mv(!DevA8_CZ&32GYX-^ptU6H*@s=ZWY z8ifI-4Ra#54$*z*4#P*}!dv@NUmvy@vQ;hUM~H->iR_V3@w3BMP^ z5KrucTu8Dn?vo7f?6ZJ}abJJLqJ7Z|ooh=e~W_ zP~nY} z_OjIZLxu-4+RZ;2hxB&hQA6eBkw`KMn9U3ZFhv!d6P|3*9it-qUw`dBsaUq@z914T z_`%KiWWkSbTQF(D#&#w0K~t)Z6slekaBrF6W>QC9Q*7jU!W@tS=_!DcnC4aBY#I!4G7#J9~D%znqLZ3GJsBNXmr;3{nAS$>_ z>Co78Zfm-Rpq#wffGON6@y)(IcUKf^eHTG^*y@bsA+?0B0K!M>NnDpl_IHy&cMLGv zYR7EM`~6kxg|*%nBL#`3CeL`}$ZPJJekqbWX)+EKwro3_hAatFl_ zlnumEV;=*|$-PEu*2Ff)dRVXBqjB?0%_ff9u+KuJn*d*CEuQX2I=&tUy+ods;wKie zwY+0aX*`-#$qK=rRNLf?JDMU#Tkp+$@H)`Nj-Kv9kvx@c%s8Y zwW1h-d(S)W`u&Uc2}t}8l(l*v*vl%gmq|WyI$m_a>_bHv^+rdkMf0XH>j=Erp6{s4 zLU1llXD2mhZD%7ZP^FuQjTZB@>Z-=7!%VO>&{I1im|a#AjZmzNUTYbb=H9s3efXTt z?jk^J`&>Ih(a|yo{x}RjdB{+}jDL!Qta@jt%Dd_?dw+b$kMBp*X#CSh8N*wKt&1-0 z2IW1Hj*vL8&d?k{W@xqbNy15cA}5u<+_}*;_RNg`Dh~QwF#ktMHlyyDDxOmIa^=20 zQNL6hg{t@KD+fLL$5X_LHM`tOHBIA&UGv4xTVkA57(kugZIHzteLx*sGM7^gO8Z;4 zyUz_9Q1LCBVs6KQ7am5|Fi!ho48SXTj6i;ObSo|jZ_v|T8+3(@ce5f2tWlDAwV6@bp^5}2v& zxoiq5y=4cLa#7T>y)yu#h1><$(N@J<|qOLBlu z90q8$`jcq+5}P!Y6%zup4UWUFflo+LyPR*om>po;M6k77NTEAA`&bofZyV!)0)+1*;G#J1u9%-~3`TXC26%ybl?#!N33cygc znP_9MKnDV<+l)=PyP3<7jsgE)YHyCme?q%7_b$v9mc?Nt8^U)#J95`0^4m}*3Y*y2 zeW15fA0g~&(pGm%lJ{7(RNztQ)!$exhN7>5xnQ~$=oOo{phy&CHx-H^tY1N$SYm+g zjkJvq_NS+iCff7cXwX9(L5V3o$I zyBb;FdbFbfowstAM#D;=yIL+7plkKk0~Hh3r&@Gi)m6O>a)1Jf8vYF>XkkOQ-Gi=^ zRH|Vc8yD9^AJ%lGucwwa2pp*&{39-gQw|IXYMDYwipJASqPxvod;PNkkpW`&49)J88SbpFJ6=ezs11R&2{lEYhVX2g$@+7B$1rBY` z*^7WG*ts8UtxI>R1*{v#w6z>xi}{HjTWeMDuSd}REr3_^H)=E2>3#Qzt+{XH>}Ci@ zag0MrOrtc2ADV%8d9F8zVmvwu|Ai+|{MdeXJpWLiAAQ2R3ldn`{|&%<=>FFw3{RqY zxv$XR6x^1JSV>;=XRX;Yp`lWN$Rp1nRDcU9iZiZe>rZ@v*rK`1otlosojh{_kgM+46k!EQFX9;eavS=m>1H&Eqbbk{7(g1Y?i2yrOPuq@pD&W`s;;i6Xry%yzcM;U%_oGn6j|*w zTF+ZI$75nvRtdK+8}{a`AG@ZJ+V=f%Jvko#j)o$B zRRX)(rkSb(T(c|@vs0;AsqzIHNfL?CD>2cU|8sgq(`WYb>~9o_6wH|aJwxZ%%tx)= z@^6ZM)rbEt^qZ1EsjTry&$NeXGL%Dhl)us`n#=x0^?Jk6v`12*fv}k$=a)UQ4CdS@dfj7?L2C&#YriOxa8-;K>bqu z)ttHWhH|u{R>f;!M&_m)L-bE)5B+Kudef)ISRLPB&vQZP^5?HVf&|R<4rGSyl?OM{0Q$X6QlL`Ug_10xP^!cTkKh` ze4;6->z=5Vkug&(*_=^R<3BVb?qy*dczN7Pjy6Dp(T@a-bqfc7V|4vOYx%z{sQ*QX?oyk6fheNkZ-g70u3v)%oNSf9Zc+`LBm1hhEOpa(sswkF^ zIiGU55V@qnAKS^Yyd=4Q>E&fm7b#aDy9pra$hQ)DLDdgp#gv)Giwv%dZHd`aEt9iw zDtJ|O8bzqh0(W_mV{PsSdFDa8F#cCs>Q|m5HjeoRSSZ1{cDQYAH*IfLRR!Fr33wn! zZylh)08k1#VgQT1AQbK)^nyYO{ANM6m;&p}PN6fye5YLfc1KgmwyL6})}HT;oUiM( zE|J|jT+>Y)k>*ZRLWWN7%9>_0|bE#an9fiLtNUy#ta7igks#`JWCd0n{8+T3l}Eyp-*Md!_pDWYEX z^Uu9_;ORds+-O^IVlVSs1_s$35kKWt z0>spwh6^-6_+;g21e8oq#u_Hr5kRM8=6p?Dl&*1(fUXDM-tK*$VN#;J8>My~tMM@C zB_>Ax3ga`B{^U&#)__tTprPcQQwIbZ4D;!hffUPIxSd!$>P2?-s@Od z`XT5^alhW4(URf6Ni`8QyPEkqsL*qMy^6@5p(5u=so~tphKq2dW3>q)?aq?ZHz&D^ z8@jnMD=^zhP#Hn?76!2NXqnssQO)j>$!ZT@SKl4c+3_mZTbwNJ#XMWlh(EJMZkKe- zB9C@YPyw*xv4tvu5|Z^e^o>xBkt03v61Fiti-qbk3+R}{V|i}Q6-HI+)^%&%1^Ww6 zq!+ra<=om)OMa%(D3Q%-Byr!IRnxiUltOU!oK|25Mc7rKk;03Xx#EWI3Kz!!56Vi^ z?}E_{2H+1R*lUq)`UI8={y+n5dCL8#9N_)3%DmY89zy_yn*Q><3`MVL#AQRv6yp#< zuI+BDdrQ$i99V~1;+gIn?Vcy_=Fj;BK5@KKMnb}&WU#^ejl{VScgH90ur6KF5Y|NQ zn}_`@w~Kpr8%;8uT}beLh^+H*;CP1QezK|-V@r_A%-7GZ$|9e# zvAP?6BU6WJ4qF=A;SMn1)O9y;Jn(6)zi6gtgbYb8br17RKZuMIygNcd^rMw!u9a?d z`^(Qdv9gfKfbwy0o+5H@{vJ`-*+lE=r*cCHq_#Kt3oZ=WOu$fh`n@QQpHR8 z|DD^R1zm()?VC5kpWV1~FQy;37kuM}F#jj)2{c&)`-Cd~eHjm#!A=IbhoyTA#X0o` z0|nWAw)Ff6e%8un*fe*v!q!r%$MJ?9Eqe!RWEWYDMU_|zrb`DjT+kG zRHSz?3gk+nC|4M7BC~Bt-scXcH!@jWAJUca^0{`3WekxBDL4=rz+R+b%c@5PI2rG* zUwIzah5@we$yu{BN%|?iHUxV^5;Qav??~PaTlSMA30Yu5#$>OdA>Vl=WNuZasSd?q zfUnwbgAvcnABsYSC3p>GUbts2_>9I?QKjWbr98t8G3&{b%MoxVv{wS20IuChjn9Q6N+W? z;_o83KfXy|;UUnxKYfRQ`E=|=HKv}D z6TV0DmwO>8mY)|Vk4S@e-Cet8goYlf7slI=n50TJyeDxCn8FfST5|nJvj3A@@%u%> z(+*P1DPa@JP!w}@k=WPLpH+uWqvLClVAWx(K)L?;Zf2@pa?l^iXq4mQH~v%%a5+rh zjewE0wu806MBekW6Ld0jAxZ@;VG3>*D1@yQYQBN)on~NwmJ--no7mt823YJeI;ZVM zCw9KnM6#~-v+qKswOU*DRp-F0TVos@UR|PF!n(_Q5i0`Sqy+0XnYIOE1zfzl@kb+x zkoNmvdGCJSuQ|9D(Vj3`(pwTNX&9lZYKzxvje=YDHs*KEK@R;^QKz*uvh2txfzp@2 zlm5NV&jNtzIBe5mxaE+Ow&>4RtjVSuX+3^1K+N8SB00_jz1knicDGTgzb0D=&^H@F z9j;(;z$qy}b!_YAhya7^7w5P~cHE}HL-h8hrc42YG@ln|O1|HNNWd-XXM=H9OXK#p zIiW-=(rmc)zpyQT;&#cd?)rN>NZWG1f zutp5EI}w|Ze4xiMJdgprF~9;hXSdjF`vyUk5`&$sZDk!@FQ3Pg!p*M*oDV6spEFMJ zQgd||>FvGPh|#xvsjI!d>*47P-Hnod&PjR4Ua5el^=rd@b~}<8N$Y8lP=&Y0LK~5= zFi1AN7Fm#W7UoL6>ZpsT`0P^kAU6KC>~#j}Fl-fljkIf)Qg=Zl{6KlE_>hC=2YNEx zCOcv?ilLr!E3A09lw6Vj+Bgma=hS~F!#|bcfBWAG7nIbZ<~MsAZae0hRQC{;=++7* z=oT|JHRBtI?hrUD)sQ!aW!cVr z!~z1#69daEYIIRwx$^}JiSxJ@Rh#wi2jBQWqJ4H)A_sogoVVDh?V%DQ9HCY1HI03g zGqhamSN`F0Yl!OQUYziD2K=#1qc>DB76Uc)DYM57p>e`g(UoNQ388l`lTHob@%K=! zfmF2kS#S3G5ed@hq*7?1(;?~oyuLsHIRZtbA;`2;orrgNkmWrC9lB5Z^=Ok`zmaF6H^<-Rj z@gVQ5Q)+Tw{d6}U4|_l$Y21NA^7Tz zOBzTl?hb1?7K}s*hAb8BDt=E?nez=Sc?DZa|NN8MTY{pqHLNmJoHIJ*aDYwR-T=iw zXccK{8_n}>wl?b6kq>i_+~Kb&~gp7INlC&sllCGdLd&$d&cW6e77&i z4@GZRWP@o-ItT?$e9wigY6|KWuEF*>tXqx+EN9Os!#yV>cgIUklc`n0<9THG~>x z>3^oCmGHp`#!269aunZhgwfR2Ma;I>_kVTJSF|$=L*SZ)Ptwul;M2t(VaNOed3fx1 z&M#~(m#jx*E7)sLxy0~iJvP=b2eybOE*jJCrVFYYc3#~cM?F@XZ2-yD zJECtWFDMIGkXwUBgw|;%)48Bi%IESnzcN!~nyZv1oeV^95zHj0Syn~NAw5*Qz zMD#lFa&0o-BYpS0-i^o04)1QwZof4Mmdkd60JfJ`lS`o;?Q);8Vi`I|WP3W({^Ua! zAsYC8+~ju73!A9i*}8(s=UQ>g@LY8>>L;BQ!Pb1YVjrCU*BJof%Tn*eZ% zN0<9@rFzbo4n{m|yL~iPR%@NaEy~k<_F=KXHN0iUn3@xbqeplBH|&`j16Ys4%1=YK1 z!f!fPpeYeCqOoVD3X4ZMaQzKi)@}s%CK(`gKhW((clGAS?)k*A-y3z;)QyLSx9cM; z`&^+Tg-#8!vpeK|k*%~+!ypK}eJ=D91m5=8`C$uUk7iCt&92VF%oiZ4FWEWJ$OX;? zLfN_kU1yRqCWrx?->{&6&c$E#{U1;{IHK{~1EqowwV#!X!RV%!9XXEL0=hRIYKeUw zWiayxq3uoQgs-{7x3wTeR=-c4iS9pSxKa8LIAwes(UFQM&x8c9Hgz@H*j@Rm>(BsM zfA~@~45SG@>bbzR`MsGjD$@|%H~2mmW?_HcY<#NcTRQ+FaXMK6OWH~u{g0K-Q?sn^Q?W^R z$2*eU*9qYY%NClk_1YAMSfU5cjV{XePflUGTnDHBV+}x4z%>FU9OCEFuybF{*_A{d zw(;h6`(yjfohl5_XU`9d#{gK}(3F49iODobZ~$|ZKNB3pbgCk^9`4J;AH(QQb`E^@ zA~@B5pAMc|;9GJB0pMHWLiPtf74RfWp?~S;Kh@k=udjrTF0!)uC#~XzvD|+5`J^Z` z@RoPK*Euhv*>8jf4^R)!>_9%lZv)F1ixug})^tqgXv>z&g8_b&MGN0eJGo~Q*)d$a zH1_ppU%h}2jIb2CXiy0yd*(fZ2pEF(S&9VT87gi6;|wQp0S? zcHrHEBZR?uePxr@I@a}hP4+X$sz94Hq2eR?e;SdK#+vpw!6Ntq)^=$$4RGVihs0Wa5l=z4>54Z7q&Fawy3HJ{P^X6$6?f?^nbm*&yk~ zDXpEdtkP4h5*z~V2J5!AIsGZieplE?rR#QC$$4uX`|izf_{p*1Y^CKaoDPivCTG7O z(AyU)dqqW=L*d}9?%L5A3V}?LMP)DV(wlud$rT8qxN6`a?+bYRv4}}V~t=R9&o)EXICEi zxW^RS`aO$XRUN4y1agbID-9f3YK7yEuQB5+UA8gg3#wi)wVn0TwXL__Fm{L+i%4Nb z!1Ck7@gEJNmiWnMUxVfS-+NnQgzf{0QkBkNgm8L*ZUrL*72bgXpA$5NesS~I9q#J% zTsha4jqK>pGr}0a2M0nF@;&kdl(8OTqy^x2Ko>7MZkjfVj5PdcTq3hSwFytrj0D~YS zvhAD0n_S*X&O_Y-k6w(hN)%j`(oxA#Aw50&_T)`gk^$+nj1Uf%g-z&^2}mV+v-orS zht+UlA1S!QS_3Q?i2OVU0$#b110fXsQP=ZFP6K#e#2>d~Xc;L$F0bs-6aU+ih6Tv1 z`0F}|7OTC#Vb~B1=TOiaWF4@Qke=fjfZO4%jy~-hVgD>LyatI9MrT6sCaZF=K1BdC zkWar50ILelv|vf}Px%S%FERXfK>#ud128r-f+4+UPsFgf&)46zzD@$TTDzTtinJoIb1Vkud6?zXP~xvs3zl6w)ihY z_v=vpCwa|Cut<8!VaN;~=J5e`7)RJs1a5)qeXs{Xd4r|3tWwP@Ke{Km?08(a_l+%u zR%luWe&=O$o?aWeN`3Vn(-B^M2~yMd4tg6@k5>nRM%j{7jBAmW3rhE^c#(36qCA_KZ|nO;CL`xHB}Wd`MvTKs zrjHc*z&&TVNNYvKD`GeENY+@K;@=NyIZdUZOIrcWY?DFStk|q6ji}BESWoXl$vHtc z*kjkxT%9-wcmhfRx($R4*BGiS#~({kyi|00_eb!f{%q-w{JP@iSShl>1rh3hpBjJr zw+VHz!tA9cnk+Q_sul7&T7e#NOj!T(G?LK`jiqMXpA-pvlz%aJ1WuLbsA&M299g`mR8j>8H%mVAN=Fxn8C`_X?x4bH z-OvHX8lMKh0PSXA1L$;RQ$U82_IS~UpId)1K*HxY$l3InKkJmyFTltF`)-N@GNP^S zH@XI?)!)!HGCz6d#G4nc?N7`;VD(qd6uB*TBQ`=Br-%8Sb)r_qkEvD7=G;_^H8Hpf zYi9$Z@uWrz`xcGj&0dLU4sm%Jc)JK&iU-$mVsU6*yQ$>Dq@h50Ck~TOk-dNi-K5-% z%g#_5i(4X9Mxy4YWi>XLlr`$>fKydj>pPOnOcFT`D3WER|uvxAgcDkEl$IDNi_ z+Js5%O4F5!tok6=NrbKF^z^*w9u?U>e`Gta8r4lZuOW92SINk*#8%}e)8H9s6aSN* zKJXv?a%T^NE~UFiqMJegHKfg%I0jq7&4Z0srkxQK?{ubK+Y=lNcS%BU6M8i$=D&!( zWnBB}KKA8zyZg5we2k;7VRJC*>9ujq-ucjv2M_l0tqb(A@6)4?E0wg5^0jjE@@k4A zKdwhaW?I7W5~xzLb>*xruTMunm*tnzocawjDkR>4UJobq*P?y}*Pq>B@ds~?mAI8( ziuy;tctH+t=Clq6kUG}A9EwIafk)I_-hiru2Px6OHmxCN)|Ba_7rquPI~%iS4iObc zP~=`bx~mKW`MM#*^ z9%Z4>Su&qIJ>Ti~Q>R`c&Mng(lV_g%BuyL2AwoDHTQ9(RKJs)v4|lV1(N&4- z>3SA5Vh=SdLd6rtmR0T9_-LM{YaS5oT=K9CBMZ-{S^e}fW;$4gJHj4))!;hXa|4PL z2c=xBvVOc<{61?PVqdM-RYxXpW7~eE79RGr6kF~YpPHv9A^ekPbv8+4=B!ox?JeAt z#(46}ty~%uWvwhuuQn7G%U*`e!{*Jv^4){6x>Gm?*gJ#%`Q`A023qTcPqGl2z~Gvj zJ!viG9VZr0KWNJ)%uFsy?|*ses>h|5x2^w5C7kNj+3wO{QbP^Yx(;3!kLuMXpC<=v_=rD>D1OX2<9uu7vwlQ zYk`p+!B{8;kf=(#AP4J_d!L@}bp;-@lZnO3T0Wt$JWR@q@ha88O}vxIb#E!6QYYa1 zw~9~3PPo0Ty4X%~2 z3J0Kf&U~!g+ar6mz9bCscsOE*e5%)8vMW;nX>c1vKA&%X#lI}J-P?I1wPl&d2-BP=%)^|ydB(!iFrm3LzT7(5Zm8vZ5rbSUR@#B*0fbq4Btk(39cA=$t^!!=Hy@gtXg%pWiDi@<3y1<4zM~<@cko5d z#v`RzAoKlMN6*xuaMS*eYoOUX4uQ{JeKO1{c3KR3;+7DBqIf|r;Y1C%Y9nQ>l#8?`C(YEm(6-M)Pk&tuCRq^oh;GD2LsEfGt-tChAJv{vq^{B(G zcciKw#u3oFOByOW*Iu0tZ?til?oJt9aW);Bi`Qr@ZK^M?7?$|Y?2+Av(!F#$s*2Yt8RwlUVGZ{fSpRp2SRUhD=HuKKL>?a5 z>;7HvY0BLF*x+oohZ-6VvYksY5z4??;-T)qc%?Xm9JC9vSz*WV00AX+zR0uosBK)p zp@rh6`!rjm-e~-jT0g`Zt!4`U!$GIy)1u0YR>zbnKvLp}=d?Upp^AED4IlATq+e>Kf&Kp9I@auM({^F z87S?v&ypI==+8{)eHpJ9q-?Dtk77}ROp!&M05o%|zKlqBu%igP87E=5@8ptC`bA_V zY|x%p-0gh9Pn6Tk8i{6u!!4zlMvgdp%;J2kcGNJ$kVP(eu=g>zMGC99GOOS=V>m!EXav$E92rXBf3tMm`~iC&=^S0DDa&bPof z8f6MRkmI#>@p3wjapbU1T$Xwi38214bknPQb2}Cwpl_>4uNWDu*9#V9WRb6rKVI{2 zeUo4)Wc>ayHvfs67l~MbN}v>7h5t77VxwtomC&X zEx*aK8k`+SLqleG2U`s5g!W9KuHZEARr(} zj*=uaL2{-Q1QZDZ0!q%AZV62e3P?tB&N)ZPp}obPJ?CuMGka#vJ@>iihtK2JU0-!o zQMGD?cfD)I;?A?wk6kaDrB)U#kSEMq+$Y=p@J$(akfD+JdEMc|Ke&XDt~1Bv)|}G? zJ&x4!rUib7Oei9@L*>Xp#Ta_3xIo)rR*Zfn9pf5e4!Z83)o7_~MhgEuS5&>I&8OMJ zp^lhB>+i>f@prUDZ)^o21Gu=Yrao5}Ms_hcYbVjR6_DYGOB(=JO>l2n#G0q1qdQ*t z=BpUh>YCz#UF~b`^sO9fE7r}GQlYc1p?7FGVB?=2jIMrSpNxGv>t$Nj3t0* z;((GQ8F-cWM6}2qc6p}XQR8``z(XD8_j>w8vP|JG%mPj)@1UgE?t4)V%BNAX@_DK> zITEg6I)3z+;BDTv4Pk#i7FWC=JYOLQzH7AhXP^2BxuALAt|6Fg5gi2JV5aVlWKx{j zxlN5240`qyyL^30l8j+$4`evc@_!o6t5<)2Cha)bK+7z=h5rV=NRBSYSvYb231R$E zU-LJ6N*)mD01!qW&5bMo1z~ZT*uw#)4luJ&Q#A*hV2gZ4qyV=-xEi+)P=QB4q!OrM zQX^^ueqi_1?xSnVL(q}(dNj1N<4+ z_4h~tM5BPBMVlw3TnJ=4c09R?+iI3^{Rd~go}o{HsWTz{Z?SGSbg*uJVYTv>T7CsM~9d>c}Ko+?FF7Ee>64 zJ4*{uQ%*r&tbIvLc=^Doe{XpkpSYGi24x(t@=D^KP34vSjTP0klS@-VQlWX3LKzQ@ zgOcsuvy2AbRyii>RZUpSsadT79{pz)qF$#pMiVykGsia9>HjsGkdA zT&n3HbgXohiN^Hh+o~wzKq>g}h}hN91NJ(y$jIU-L4pL%s^o~0w~5O67wnt@ZNmrp zi+0xdTOWN6Zd5=8^mlWt`jO(~gvorNnN4WO)c%XVOUPQqKVM+TP;x%vFBVy$qMs`- zwM3&x=s1bx(m2`6%bqbF0SeHeF)@lAHI;Hyn_IFLSCA(Me&9Gm-5huC4a}|8^fxHj zCqC&SD~cl)^?Q}h{w+3%#AFL&k`Rk7&Fj(M`r4v!x%}Z4^cPbeZxRUY#z$(Vu_u&vnJwev9tZ>e?+h>eAOst zzMyaQ(O4-L^F3S5X-1p5kJC)ppp%bT(8CpgQ$T+YD`jD0WKU!dZ4IqT5O@xq)Z9Cp zBgKqXgGKIyHaJiJS^w}&t7(ZwVE2!qbc zv-b&B9%-m~uC7k=i-JYxz2uv+cRMhyxpOs>gHR``IL0PVx82Cr_SUP+uN?Yz!_p-@ za_A*ZS>SGe%!2tmpy3bmusJu6Kt7-hj6#lb6Mbhto z3b5Cl&oHhL|GlGd|G1-)|H*C=qbO37OwV_;XR2*Ynizibisgs!MfueVMnhh7r@~)cXc)p_*Q~9@~iLonUT%DB=iPe!=W%#5} zR`>G2o#nS?<$ODzi+No_2tU+3F(zy*h=%kf7bN!~jQeQ1ieQV0I@4R@w5-)HAL%DJ zl8ne3F3-#PMoHA1@jcA{#`o~3-t;uF2y9Rz?G@*LdUyO!Wt9H$T#(>~MtO1Am|^F} z`s+k8Gd&31p4B!hh-R2No`IE>|H(w?|H&jUawLtlU}Wk%IvnNHrB{u+1Vr@mR*`;Zr}QfB3G{a!yT9)6%RHs2n%gO!Tta8 z-#r+w{&>KDJs8se=D`R#xP}8QM*M^npa)K0WgdzqFK|F zytlHZR^o0wfzG97E>AqB{&K3~DetHPQOEyPO;oqn5%@(j)OtGlJ-y+Z(Lp#`m;?CV zGvG_ceEAw^M&NJ6>aCT20bIdS3woOFZg zNEAMW|b7E=>RMZtAlS%0jtVDSNbPJ?iP~p8u|zKO0o!r{#U7gv8pDV zpOnU^+ezo+S-Ty^o=%ev_TD|#uHlv~+D3QbG#B!WF5JMr@#Vx9tjKG{I`|IvlK#OZ z+5`2D=|UHS<#GkikrRZQ!zLnCC0s-!Fxb3Wc_0E1w=z%F9FgrHPcDR6hJcLNdoC$HQSC%kc9k$7CZmC$$z!*WI5MMRCfrgC+ly&^v8ap zx_|qv-}r4Zy1;n;(C&|&{lj*D8~7i)`{yIq`fYAtSbym0{15(jPygKxtk}*!Zs!8F zPhh2nW1JHG7HQknx=fuqDj6%w2lS5@Q#gs3eW_*ABidP&Bh*{3je4Z*ImOoLI1 z0c5;`F6_zX_e@9X+kO+T>@C*btgC>ynWO*nwa!^vy_u;EB6IYu?*3!N_n~nG^)ti| zPWu{-j%ZSD&WEer_Sg|39Z)WCe&TeX8^G-0Omf+p*^K^R$umQ1dIkx(O=b|NBrx^ZvR>p^vz!- z%1Y--$6dukA8f(SYAsF-> z?-KPJ)POY;^E-&A9i`OqoyB082u8riaDYH4t_1M_8nT)_(q8yZfyKAK7C&|M03+CX zv=|~4wW;(kR#j8(xr+0{Ve~6R$Mh#hc`j-}m*)lEW0A{iQzmBP&znZBd0+nE=8z2^ ztOn7hL??fM^w6w1q^LN`%3LmY3bEvJ6gt)>$Gu2A+Lq616o#eoDxQ$0Z??bnb6@(h zO3=s;_`GvN)gvyADne+0t}OoZ1-U*>PJ?(#7i-cyYb_I#(9|ba%sRdCr1O{A65c^) z-s8JdDs;9(F36$&*@;>*cgb)aC!qAK7mbA#uf8MkkBm-p!d&Rp@p5$uWEM z*B?L*7bBy-b@@^`7pCWi>nAOC$HdW?+hi^8s7ZzdoZQdWO-oNeE~z_n$9*0dmu4u& zQXnB#ghUS@QNk}lWbXe&x;Ziy37zEw=(<6*{htt7V`K_!fxY7p9|qkc1%90FPi#qL zV=vJRJ0P5WVWB;MF~Bt~B0!qtY9(U!LL*LMcx`wOO3-C}&G3hHQI*;-XN#Y9kf2NF zbvkv9CP?#5JGbb8O*o*JAb=0FO&rOIxJ?8(uqO~7s4d7A#378i+Sb_5Qj|5D8m_)7`m z3yvW71O2V2q$P*5$}j&m?8S7Y+;nfkAH~ViK6_f;uWch>y%y7nunmnhz1MCAIJ514 zLU2}q;7AGwuDP#rAca$sb`yc>IJ$9nDCyE_<()L>)g~APZVq_c6zN|L zQrlNAsThGV6-eTTO9X9DUT17M{0ULRx-Hm~w27!O-e)yJEM-}7_U^NUFWEg>nQ!U7 zs-ciLSqAbkb%RJhWXmeg+bH+aM}&3G0+*r(SRC3&E{Z#yukZ-8JW*s zh*MwADN_01n*6ScSqa_Pr|sF$-9MmFJJuj{giU3wU;o0zb9YC}vm}niiOCz}zn9I~ z*Cq$ADvIkCmn3Vg4BKmtSu2wZKb3m1*(4|!U+?wMn95ty@W8&-nO}Wk*ln(jENTmr zx+Kik$XD<>41E_5i1uN;y2d0$^2g#Z~$4BdvCZ=6dpo*SFO7 zp9m-OIdwR#3ZY4@10dVv9;VY~XzC8QD;&>xVq`62bRrcarNF;!xcd^rCH2IyWnMK0 zLgE9TdN&#*%ye3t6y4HWS%zx##X1(LD9N5_o!Hp7?Wl6ReZa69PhH)ei;8L0s4xUb9;@z?dTtmR^KK;@V_P}*Ner0WF?(E ztmqM@#VVY>x6ZXQ#COFirO)W)t!LOokGFzf1xsW0^rvn`1$xMeQUt*t^uP;G1}R)` zRXYCQoM~CDcez>U6BRuB$U zb_##Ng`om@Jc1J&To=2g5d~03RyK~UF-h5Db?M`}-#neuvy)`f9~+bqm5JXZ#@3(3 z;6v84SXXL-$VTlFS!)jCdbrdVs{m5S+g1mb5kH~uhD*-}T~g|mP9F4IIc6r6d@k;4 zU%mdM)DH4KsO{jmo^|pC2KBbZZcSb0TIG}KTe^muE4SumqwTPU44z@Xf0`%;X?)?D zmogP&y5;!2hB0@|HX}%6t%=eJn=^K|`8x}y?}6ith`wImYJy7NC?W;MxHTk~et2h2 z8?7PNR~t;}TOWLI9j*;t4#p$jqe$dCM`I7C@JR2YDIB#9eu)OERZX}vw z7`ZY{xayPq@nuYtFmBM$*mv=zl6cf&`GEgZ&AuHGafB8_nhvTq@$uoO2c^uyj_rne zC(v-^tp^tB!=vF#TjNZtPV$y9^>0K~c|B*yg;&2po_8QzN*-_ZM;SP0n(ylmt1E}B z#+cbXD|#Z^W_<7dq$$J2uuHJT8X`-+G1Ft39{FCWq^nVD?^`L5gL_njx7Qonyw*n- zhO-vYkiMM|l5yDE>EdX)rOMV1;fjG#;@!7rBUPv%Sb{3UgmnX*9#VXZyCa`A);&7b zcp2eL^D`d~?YAu#@5J9r#LE+C6us#Or|v)uN!XmeHNVu3^1cV>_OKJ}e<5nCz4@4>EK9tZBfJI zd}@KlchkDuN|l?%QMQ@yY&hv$H4Oq>DKBOk+HYgE+KyoIHN$J`2RN(A%#ZR$JMp!^ z)ng^lA!oNlxItx2gyi;>c-0&0i0Ju>#_|8E4@31J=RLWsLa0;gZ2iO%d9gdTj2MlO zvK>(TNnq*<;$LyRuk5ZXw#;9WbG7ZMYv8g|xHs1Ho$;5fOVF$8k2eQv1`y*%7kjl@ z+y`^p@XdQ{h7B!A?%xvQ_53suW7pSlkO-yOjEX$rQ?KlqEVl+ldu%2J`Tka|N9H zd8VLnRL#*-CQnKYXBxas)L2^ynJ7h9db}DpV}V9}+p2;LBSMC`>h7wIf8MG|J1kxt zEHl8rQUUzyr<$EexBc?Q{LKXayRG#VJtt5=nBEu{u07O9iI5^7he%%xX((o~o<^z< zcSQGt`!Mv4Iw36OAVXf@g|>#-z*QM{dt0h73+0AyCZFaYnXg{n6cvn*7*&$zFTN(T z4zzt!z+!XZF7$-5{fwtYZ$)}bA=T<;S$TPAY)u;f!F0()cgzqTT{MFq2*dzKQO@UNU;ZMo z>G<54b6?yLCYu%_T%E{IkFcLLE7o!tcN`5hd?ef!Q7<1MY6=p@)7<1qVq-=vSS?+B zJ6?H6+oybP_)yZGhcDf7tm4Pby_#U!@Dsu=RQSVl)TqkJg|haj+BeIhBpNxWGFE3p zVh$5(4dG~!VG6yIaNBK%Z@5dTna0{Wc}{(P8AN-PKq~!zT>TMSWLlAI#KmI1Wms4A zXb)Bt`A}Plwk7`qo{Y0nzvNQ8NlBMfXA*6`le?4t{Ls_9B@e`ji`(0 zMUr7FOT@vD3U6%XyQv13#ob;RTf&fU&*AxEtYt zFX%8D?grO=m+>|>J-J(`0!jAKQF{f~3RAV3UztO>)w+Kcw!XlX@6~)g1&}Zb423UaeQbB3~^EsQoVNEVIIPR z$6UaER02qVIe-LUR2{ok56hj<(-i&WX=}9h38`ycn_ppblH%}S;*s|DM<M~JJdHg0fdpxvUD8JmVk9c^ORGxyR?|==Af1OtA&mhz zd>1af^3auGXyG$sXnlK20#w6qg_WM7>zL#r4kI$j03^YeX7mi%wU2L5lV} zYu{Q!WVUx!As}<)tU>_Y#HFuG0S@Kp2&~L#!6KP>p^VT6I=f;LX`IwCj8g)`k3BSMt$AR=G7r>{Oy5YYKtq=XS z#`M!%iT-0FXc`#}Q@54nv#RJ7tE$GMjP1eo6G+pE8Kzc1YsJ>}C?lpyX8|FKNvt=P z|5c>`F7Klv&o39KTg~5?A8g+fYqXn)&&JNr=`Nq3;>j-*G+-XrwQ3OPxIlhtUK~%? zW0P7MJZ^p?@2fd4ZYIQy`2Gp+*+k{RU@O32>7yks(yowyr4QY?WCzCp1tH4yK|zS9 zIOZYe)v;Nb*q@M!66Yl`ie5GOM(~O@rF(PwdvRNI-fLoFuS72CY1RQR`W{Q;lHh4` zQUqX3!z>ox{c`Wb_V?vXAOOo5R$A_1rGy*WaV%zaYW7a=bBv0opuoq`bZ1PPn@S;j zEo-M{n6d$_pq@g4J5&7&zOSGLM9i9r_^U-tsg1h3t-se%csM}wn05tMOE&L-^$94X z@RDyaH{?L(?gZGE)C17x6NOOYC@=WvNX1qI3)q=4Fr!PedG@m%=-j;2+=&2RCiqnr zOuc}X3h+ZFOqKSiFR+(*?wlsJKFg{K6Cgd>afRIRVNb#u*4$Q5aX|szQK@dq)?JJY z$>)+tP5Erm*msV5ufNM_$>wh@(dqhc-W^fAIyl~JOU3R)F-nX!`Op;%E!PyYIUq0t z@$9COty>B9Mz*dLLpQGsM(oJ#@lQsM?@_yIB7b^UrAuyZXfQ%izXDmh1OvrY?S_yKch5k80{@61_!jybQ!CeCO&>a7_q zO{k`2S2zLo1c6I%UHy^rghu{F>hK3A!rWsTZs^y~g7IP9w=Ej^9H)N}u&xc{Bn?P9x*|WaX_GmP)DbIXG zzucW!i`jc-dR_0~wFb^$F2QAh{(=K&e)NdNcn00)N88Oitt+@O8q*PBi}lWozP0a- zoT-hhOV^(C$z!9X^K8&xBhOs#=}%;IGTTcGBE5vdpX8TRp*9g^OsIMohhlZ7wbTu4 ztQS(^)E4k_H?5jFK%_|m?lT|GnjPi-IX?MoYd5e@Zjr%vSiqIJ-2tL+z}WwF28xo< z=RSYFwbiQqbG{*#WZ8 z93iqR{0<+KB%f=XXD|8>pD`AoMDSU&PUuR@J_x@QBizbq&nsE$mCdIJ* zB}H(^nl`2qj!vTh>o0RAnVQeuu@kxztlFXO%VY_xEAFiM0313S#N|9!2e#kaHn+V_VrF61dn zLd0hiv@SCDXC%M8xvju*JJvC4G4Zl+_=+@W^0yD&I_X6J{(cDK%->$x>iY;GAE7k7VOCHg}0nq^v|E8Hd8cBIch{SBytbJ^6?~{w7ig#W4_((7a!==N&{r1h4 zxX<4{kemIX+y#1}k7_36B*7pHyAvQB8W0%(dDl$h%E0&AM=aZ7S!RisD?_Ej?tsQb zXBAmgik@6Nt2uL{T}~3B7TIU=Afrm!DkH7JSap~MvqcoBnoag~HM?tP>bLg572Sz% z6Ndx`&gd#6e23j_iv7{5`#svJOQhp2*S^Hl44g37WLc;UJk@QrwL`3A>PX!#A!d7- zJ&DKm>dTsj&zF5NG_>D=lZlJdAx3k0!jU6{`$N#h$Er7)Ip*G7y+UUnK6vdDaeilo zyd`tpvAoto!=2H^b;FXEzP;%NAIC+L>7j2Ao6zfaF86Hgvb7Y#O_G-0FCBT{;ulHa zq=YHjDJsj%hBxy$B~`?mSj2Z*tKF+OpQ+39tMFO$uMNI$eP+?nsoteMq(RjjWBm;R zP;&11dq%CUNZSS$yy+dF2E)>ZFQ(>H4pjSZh3TxOwPV;nIpVIc<`HXjgF21c;0ZXX@ik3zs2LwM?4Q&NKN(;ojgcA_=UaLUd(}esP4NcWi zYKLg#J`b_3NOrMxfUUF+^ll*6ty2PSxhGv64vk`i1YLCuEnh%cbxBzHhT;4OAze9I zPdcrb5>FodVr^yONBMS_4C_&x>|51`j8vu(WA9zouk5%rZ(+T{sxizL`pA`J!_&Wn zrA4+&leoj)8O+X>ZH@ob$rJaT%<%{n>bc%BnHyG~4RYiw%$B+;b+21er0#)C=oEX2 zrM1SquvH_IAM1<~zu_|eGp4w)u=TQwC%*2FUhhiygn0Ng@fA+RU9ftv%;~P~o1$`9 zbj*r{CxLcrf-WRcV{iH_D~Kv`eVx9}E@`bhv^0<--?L$#*|mR z88GrI#~egcixOwNq6pLcILT`&yhj=i+x(&0Yjd4(voFMtUY={aWnNpeuX{I5H>);m zMCD}2%N_6uPKa-3I&}dHdEnFFCiN;+ zU!Si{-A=QSDg-X*LT!1IC?qV!z zF$b;g5*ABj*Ozmy#dT}RUFxehzCkTPpW@@kNaK@W;Z-xEO?``8?Hid|Rs4bZk`+wQ z#PntyqJHW!WYcvIfx~?XmQm})*qtXVKpGK`bi@IbHO3gdzgS(6^Zpw0f}AbP4x2>4 z+e4hp+hGl7=3&6mgnXZtc7+Q<4Z5b`WymzDO;;e!R&Pq}{scSmz`eokc~!>n_im=3 zzWT7*V&`i&-&a0-dtvfq(&)ayeLDAT#mEscfxCf<(U!4HU zZWeY@7HN2X+R6Bdewg1IOX^C4Gy+zJyuCT^Gb6c`_Pt~pU+8hXc3c-z$8hUu03MYz zmAlwVVC*c~6_QY=Phv6~KjE0|64xDo>_CKd*WGqIDRHY}l8XI>@ zT^fqx$DCWER2BbzMflXk`qgB}HM4IBzo-;RThnA}HgbaxAFA!@_6z2uu&^H8XAJMT zi!rcL4efTQB79ntAIf(>_hx{pH};HW>P3zNI&nNo(U-BRR=CsDd3~rKu(wtjd~vf* zR>A@_o<+gSH=LSC2oCNqy8qCLkJ)Uj*MfSZYH&t$hJ*#YXT7GLkIY|Jjz7NICeG~f`YNQ+5@-*?A~S0OYTdb9e3&#b_O2YAl9bKm_i5-Q zNWw8RY~Q=I+ZlZkr7PYHj!Qa?QK&BN+U4cZmSc{hNwfKJm5K5Z0j}wL@dnpSgcl|3 zD8oC?gWIm3zrSbsnqFH|gSf#B#zs8It9XSgGmy~lW|)avg!QGChCRkzoNK7d$fWow zE^BMA_A8I_OWAqi+T3wgc0`mMtFY=+H5gynaCY00B)AGdUcd!^SSfyuXX?GXc4_0T zC#@0;_ll)S4^xausojONT8~sy%ybOuD(OTQb-9WzVfJ+MsC_uBqPRq3E`|~0OvFE; z#sZ{MhQ9-42w4s)fKz&c0I0htV4K+JFu9WjRbGR><0=|?=cz12Rw(oSE$8E}^?tlV z1LWI{NUeG3l#)fl4(||1BF?HgI0ZsTgo^)GoS;X6puR7BThX#+W&Q_%Or#-N@;#Um z1^k2sY*MQ_%U$Bt&0p(j7%>LX)y~dPZuRcuW&3z5Y+C}H&6WSb_36>3CjF-?5jsT( zCX`eGqVhFP@RJY_A+{HtYN?>AIicH!?+5^zjr9(osVuL*o%+^=L~`ptt*$964`%ta z{#wF9N+T3>T%5xC930iAGkN0cHrkP`SJjkdC$5Cr~B}8w|#%Lf2w4Xxr_m6 z{XRs7Bah?QxuTijQT-Z@=#Ti9YC$CYNQ7zu0l4UDB9nSnd?y*wZI9p7a>@3 zqy)e~1VT?#X*c5zplH7;@F5leDs8DQkf^7eemdQ#Ltkm$?N~=M1BGMDB`i%ts9PYje zY8cV#pf{$Sj{K^?U0qRFUOiALBD`l)jqUk>0~npvWj7#`TOg?y-vK%jExjkbSF@KF zvS)gN_Dm8^s4Cf5uo<60>ajPkXe$E4R|dG#4FJ-orxanSaBVPk!#zd`xkjlbM0=ig z2RbFR2|vYm`<XTPblm5SKm@?$*U$jkzzE*<`)WiWW$jJs;G+W5|7iC%Y z^bFR70jlNo#@93Em__)(9e+@PCl8-jx0^4YwHu{PxpiH*qhOS=f;ZOG{N8uMS(sq9 z8FX6&2H(TW2i9SULmDo7#Er4R6_#9D64<^V-gR2z{D5uu5o95~+zSKM)7~&*4Tkv8 z&_L&0n*+JCjI#%O0Illd3O%GKT`^o#isF=EY@d;}R}TmAAP^3Y z7eE*O`aur&8VB=_xp`iSjH&q?W#AZzTCS3sC(oOP4 zi*ZD0vn-p^(5JgSg}~m({13x1pJ<98MSo|)P`@(MLAI<%IzkLE1Vt^33`@ z-RlcMLw;0#5V_E%O=Y_viVeq4GiXn43o=<9*G!Gou1O zcR-Ytqzir!n$L3o>n@d6E9-_`afBVGF1!#{OrE7HM#JuPRu=GQ%)njB&>PxB9s&tO4yQE%={#piPG<>AMx(N{22 zo=r~qW51W4(&+8D0Ay|n^Xcnd!c&gu^e1y{yz_#xEXj$V|Gz>P{z*c*@xAa8?dEe| zGm@_Cxm$*u@g84heaKVqi(AA5?-!i)S}T8ET-Pj>t3CL=z#5<7LvjYXaSIf)iw;3w z90op2X#NS1JP*==tbE0vaN<_gk*9<`Dch7+GHXuJe$hAK>VoqW zQjJ>H9j3BcIDarCB^QcL6N37)8B$GvjMXUA@+J|=3SZz zHvKh=iv947Gd~=o(%vZy{rc1t4JhT{+mcjU>r}ZQdiB?R-Jcn~ubFYquYvrvC(X_F z>eO+@J{0@dtp!Ctn4AW^_FKB=PC*OLp%u;$L&;@nx3|rFNiW_hDs$7ob2kT0eL)yj zhRed&$t^NG19rXDD%*Uoq>~++NcTZO=CjM+$%$gGbyR_jhuSM=ml`REos&uYgwzlC zhqRzelvEZ_5ydrd-fBPQA?e70+bo$Us#8_CQk9!?4N=KU#k|$2T}6?_S=UI9Ilm6e zzO;Q$AmS^dFA(zeR5>}L-92Z`dchN?x?i|IAN~ctA}j~)B2|Dum9to7580$W>L3On zRaN*-*l-2TF%|7G3c87G+Rpw7A=b-f=T<83n=1EpIHsh0LwGb$TIr(ciY6txXD#w5 zPu5&ak-q$KTbVV=Hk)3;+;|wU@9NFfrwm+X0M6=(2X7E4`)omZbK`QC@6F0{fF+g1 z2m+swLjAo~YI9Lpe>5Nq)E=aVc=+J0e+#i15)Y}nJK`P$ICt^e1nq#!1F&|en}NOd zFcbPdFN716Jgo*Cmdv9if z21QCyk479c?#=sBKydiZXeWJ`+EV6)4OcGpWzh$97K=}QV>g_<1Ou#;XS}^e+CeTF zqnz#QXH<=k^M~#QLB}OnIS+29Sl66J&4IEaXC#w97EWKCM!n;K9%s#a*Sw=}9Cd388;Q}C>c8Aecu(Y$J>5pp2Bq{Q%;@(w zKN>YpF5FF|(HgtWXd>?2cgP3u)!4R#vAeTH-7ak#LZXZ1U&<=V}@ zWUEu;QMHJ~#TpJ!sH>ks9vFt1Q7y z1zX=$9Uea^c|6l&eMhoAK8r7?@(%3jR~~q7=_7hh?#ULhlS^msJOi|wB4hC>!o_P? z`L>q%FBw(*ZV@LuEmMHvJxwdD_EhBxSFZ>jvyI^kW}x(xdGHewu2zhlMt(*>N*cH0 zmKvp)GuCQ7^esmEYLM6Kn_eQrz(K9ESrdq;G@u>}3EaiC5`I9EbJ3{6p7^3g_Q$WZ zye3t45TyE-o%E0IV ztyylW@l|WXCca?^w(N`-8@ySh3IJwepr>pBuT?}DT|)NwfnX6EfgW$xOiBOoK|Oo# zZQ=?IQ*~u=w{qht@x9at#5#@eq3;(=ORI!k)BBA|S``dq35DrpsYN_is~l@oRjAaf zW#pVx1~N3UI0K!Pa*mTLOMJNSaZPBuyM^`zF=&wemlr}i3_gQ?up_RVDeW8$RMNAz z_9O+VkptW}m2QZQlt`6=QX3*}6ssOrCIu?OsGW6R=2|?ARO7*;+KUul8B^I=cKxC< zoZq2`B9p7?P%J8wYrV&nLckEs7y9sx>PrTOo7;pU_;&oiU@n{@8roA`f_wD(xg3r~ z*ye}Pb6i9S?pfb~tCB>Q9&!mbDAC{3J{%eul+J_0lVa4ss>i3Rsf9H)gR*TF_ zT}k;SK#?A0^;<=)OBI}bGiMdG+Cz((FA}lrEGf^Hg|kX-7ilMFdL}f`?p!`ICu{_O za;CHgNCP_=fL#7sIdKxC+Cl8fsQ<58NefGaNglEJ51j!0PAmwRC0IY;+kCXYK%{!@ z>+dt_r@N4$CCi|z%$Xo1uTSyZdZ8-W({+~waYpnHFU%pVt*M{cEmm7@%R&sZ6)<1L z5=2}Jb0@ouDnm(<02k!{0DFY-zx9ri&XxAu)5BOD}JR#wQYimni?>GJ#c4}JvK6KVC;OxqAp>oQd3JNN)m z&=gU<`mXoeEy1Ldud{A6td8c0S$V;huHfQzLw@C0a&ja4N74^(V81y2f*4&V-9AL% z5(zHLjVnsd0eO3I=W}0HCg0~3p_;2p<8%oMo4XBhccUX}$J+7RW6Dazscd-*bF(i@ z-+5KS6f9^Q@-!DtaNtEqIK*=W|J&@^ysqxOy2t417Qsu4*wYH%+z(`*rbyk6sgTCk zWR)-SE<^>vA8nqN>sTBGp*|>#C}q6onpx!SFM=v;Y{$!-Il6r$Rd3B zO4#6gfxvhSlSu;f0V15&S!qk(f9D-cJ4iw^zu0fpx+F}d_lh(&n|;%)eNU*`#%_?% z;~s)EPtS|j#aMrgeR<3eS*W(Vqc1u-N_ZfBa%a9Ro`QBWXiSO=k^Cca(q?3i#X9hd z4K{)bBN`qYjDiDWR`JJ)kh@iqhB<%wh-GU&vk-1Ow1`v#-1t5ol6_??@?EZ;*eW0_%&EvJ^J1q#MeSVq}$ z?ksjoT5B&j>ZKCOZr?kRbP)yRAb?9($p#m0&eJHW&|fu3IOH+9(26bWwzZ)-Y&kr9 zqTRI%y=vy%Kd|5I#Kb9VK<|S?LWf`5P{z0^V`zps&X%f`7C0PxWT|c|vR2L!E|wP+ zq@Bc23~Q)ln2xj1b=##>?sa5IDk_8H*WKS9P_6sSv6c`Q1;N`}xR7yRTpkxy->yt8 zCtDqSw4xXQTUuufOt@PZcA$1o*-I5OQ7Ia-7yXCyQx?U%CH~=9EvtqMQ__>sZ&r-U z1<)>;=2R=^NAj8uk$oId$_Alkrb?{j9i!jf?7J(G(~blFQBNY)^ahR%`_bLnE0{#u z6+g;^(|kLG@n$ncKau)o%0UiZISPETzTi?!co4xaI>PSzv0K-s`Y(Z++s?g`58B4#Mdt1_2J_4Q@N+ ziG8#=C(W#{ zAFW+&vgYS*nXxxXwNvN{^;>3tURPhu+V64AOLkd6u&+Pa%8^p;k*aZb;sX6$!~5)A zZ?03oDVvbUv?Nd;JlD>g}2)rxyY+ZTpQ11?8P z(^)e-ugzqfyx>BeDJd~69i;BO`3@~nO%JpQ822g8bZJ8VGJNO40Ce-IJ<#&iT{ok@ z@8PTK#jhw<7X?4!B*O2@eQ(3o6ZysQM#J3FGP?|)QE!9F>ntr!-qJpgJ83I!4WF`> zuNHMLRU4E9g{*JS5zJFJv7|a-$Coy=lc<@7fq3ozJr+-OOq(sWmi1xwkHSI-Vcwfz z{68U*lpg|pd_>(;5;hB;0SgqSbFv0MIEAgU)fAwoMRJ8w%9Bg=4>Ui+J5dNPrxSzNL&ubKxKB6+p^I7rP$7LZRQi{P&p9W^b%Qm7lcx1d1R46eCG8F; zEFlj06GEHv>(g(Xk+=6gqm7+a)Ny|W%mNhEAK|xMt3YnCAAfY$fb$4tzqPXJURsa5 z>{}l|G`hAzFle{P*51rW@NvwhsgsrK!Bc2^O^NF8kBCfWOsu z;uJ@4;q;^sXl()a^`Fh_evfH=Y-Bidzd@)wLhU}=m8B+x^}x4zE47B&1+hL$`PCp2 z2>C>o&5rC-_5!)Y{+;krjU${(b~=#W9UfM&;Ep zMN1@q=3@JtY*(0y*caoZd(A5fe{vxPYnNQGJl2{%CCx!GH#)FEPBcDh)n4 zu85Y(vJ^JRkwcgllO1t>Aku%lnfuv-o1NiuYIwwLgcJh<((Of~LAHz(ul#2R3rXFS z0ze^<+XTW75N|nlP3x5Muwk;&Km#VTuZ}d0nTdnLeADBU8u;Bo$wuVk*y-yi5nD9> zRuTGId5_iFPU%Mc)-5~Z^4qV8J~kR#F0|3au}~S#t_By#xtN49hLMl;(B1ezTtp*I-}pl%VfN1SH#q9&AI$|BJ{)XExgK7@5BU&3bMsi=Us&2;+LP5L zSHqj2ImwHLY&};5{@}JWmS3*CEN_M>Q<>+g)vXW2=X`N=C;wA$Ycz!)n9IgB>AbWlPa|ZKH@X)KC0tEYW*uEn*v^>)nE- zi2)Z`@nk5%G+sZ!QhfSNMyKXqpv$O`pY6F8lqBtd4*v^M?i_fW#IE{lTSeS|blTLp z&u?$&)a|p#k@Q*QC=z}WD$Tx1)mql8bbgqu?K9ed*_Zl(2)k0G+(@%!N})cZY(eAv z#>G#d_yu>9QM`y*A#O@~a@l!|-C3059~`OV0HEid|MhcgGw12^$yqRf9hk8HhGo5rRzKdVGB9gpDn_Tw5G!a>q#ty_54c!;+-uN6(kJZt zc<>H9pmb(VP6Yc%bOY_TW{*Avk=$(&+FzSy z@4ZQQ-n^K3KAs9H?EW!~y14g7B<5fDmNI;E&oVy1=U+udL(F`J3gUi3K!rI{;DnA% zIFh#4O|w_4zIuDPSYTfeo#SDdYHk=XWz{F}aa=foi-~zIv|>O`Q)X7P)I z!?%VHvk#eMBTSU!Ek=$zq{y2Z*!TyVvR8oXz|{9AWR;E^9XhJ2OZ3>u`SxL0(`O5^ z_eqgzq25ew0f~jg3M3d|GShk7f2`4<_naUmBREK)Yh+!}qBZ(ml!KcHiITph%hZ{z zuBWu+475t>Gp!Qbvg1{Hl_}F3C;lEnAjmpUb3|IjS5{4Vkbh5>1^H+*p)LV$*$cmO zV}sL%vAn!zGTIc~U8eiJE6eEBO-#o4OJ-vsYSnZ~sQ!hR0qM=7I<>MbW9M>-k5)-^ zd{;httv_6u10jRhK+n?=43FkH_d(8z1$`wA+8Jk%$I5Auuv zWsuVu)ueeqEHBzH-ZJn_Yu2fMD_7}Os|1SV8Q#Y$_aKhR#7d(-n6J0?s+`+yDZlQF z-Hf7#?)-|9;^z@>f7*^k;;Ta8- zyFlnCs~=QPoS4AepLyw>-kx<(OkC-y5;|EI=`&?(JlK=*d^| zQD^$Ue9YY106{@$cx}zzwQx|>cHLs(*F{{~`|DHJ+mdWi_r!d3$t*dbsrO8Y`oI2< zJ#tI`?XDuh*f3{xh?TvG?KWlEh8AhC@(ZGuZ@1!aD_-0hxUpXcvlJ!=P%utX`)&hlUReM!x4NN1<6@lQ=q_J!?mVQURXXk+Wn~y6@b^6M|GW$5@iy$M}^nkBduDrEFBWG2) zDRoHurOWr>kRf@fH*?g(5$zYBHGhFLc|dd+s^$h&M(KYr_O0yy0)O`UcldMBl;CP9 zjv=Jl_jyTaDU7Akkc?)$TNsbcfPQ;v-t7C4n(YAFfSkQX)~fEI8TKDFu9eao9DqK& zoeKydlcRyMMIN{AYAyhod?=76L{F<&K6)@SODGOGShK?ir2pAiK}5?wItg{n>|`{P z8`~xTSYBP`|EZbm|Et?=+*_5HlAf!lFV>%$9BzWoL#u0CWt+ai+T*o8i4ZOLtGS|I z185COpl#sd0VDqCI7hdi&E{rcO09;rH#-~Z)y zWA=BOrqVb!Uk{7 z!*@_y7VT2H2x^H$`)7wfHsZd=NM>R;MYHg&I$!vfndVKp^}Z^< zqnoo~MeB(&wItF&uYEA#m72XB0W#UaUD7^cxn+guxCY?K3Zp>^Knn-x3*g&?v}_xV zv>an{qn}k?ye_qcUi_Eui?QmMWp4Ue%Z-wk-~@VTlhHtEk_8|I0@|rpoUR{LIo*OH z{{n?)&+uG)IkGJwMBg7G1Uezv0n6Ctw`z>Xp%7zq2#oeG5D4flt=e9^mi#F`%N_(g zzWCdh&NywCP#BbdfgS=$SNfwr-4#V(zK`F6N|sC@{lBw+f-$Q@_fU>gZS@;nEn#N` zd662mkacLEpw&Ct`?6S%gBO!3-yM+c0uY+_w*Oyc`Twim>yj>XJXs2}_c1;E%3(6w z;X)qnA>Nh2`}1*7!4)FWUo2t^sJ4iZVYk-+a*8hncnI!|Z^L$0?Z;v`%tCdt?=|!? z82(uiQPNvef96GBmu_SJTmqn6&oKQn1JVgNk@QKU2rw3_ zWvD=Ra!6P3-`XJCR+96E66;;?YQ{C`_7aK}D5v9X(QQbLJG?We+lT!C{6%n5OG%<3 z?1d^j2i5rdj!&3sv`eu}xEBqGt#yiNcX(m7J&93{<+tGs8akTt0Kggvtm z@2u%1FwOfSi4TrdLjrU6J8@1a!B$?YWyNUPZBLr#f6 zy^EpOE}C|XL{;?n7$n8~_3&qfWBMsiK47#F$!&p7ymv_jFydBPsO?+sb9kf}=y4Kr z9H56>AS@45D2Bu%h}PbMR>H5{1yfdF@`e?aNU745y`m;#d3cCa_iR%?Pko}q)evpX z6K4@=_l&V3^H4LY&{;6p*boMYcNkNTE_%uWenyT6y}eV9$8n4CODpNY4mtC7szL>m z!eqr7SW7+qQ8hs*d-FzZ|bL>U>()fUDETi56wv0}{z3KsnT zpbq)py-xbC_ph;G^)dp!L}Ie-Vq2eJNcgqRE~0}raoS?;a-#flF86iL&$=Hl0{&7h z?a=*m8utF?`=}yQB>iOFBY2_U?%dN#IU}m!l2fju+M(x<|wh{yqtQMm8)<$SfQ z<4+FSw8_yj3FuhR(H`Fyv>rcNZ;GGNJh4nvUQ?V|mqUqDDCs+nA9!{7q)I*2Ac1(P zhW-W8ezXLx^-YO+!Dh(`ca>o&mu#WYj*Ln*m+i_cf!0TCayDAy!PkiyZgadDf?Zg?{$!{C4h72y?#2KK7(`+xtM9Kl5kVzJ%q< z!E1xNm^Y#3H7YE6C~PNpYU_1pnJqlrN{=P#0Vj(oDaOj0@t>h~KO5@Ow~gCASYSE7 zf4m;=%)^_0hB2jO=J^+hOc1=|CyE=MsleG?X>pluy3rfiM;H|qBD&YM#!NAcQtIMd zHTK;;EH>)U9xG9UZ4|~TlePA|XeBDd3zng$|IegMitgQ0`~rj?NN9R*Ylws_`k7md$lMBTf93zZut{Qwi_# z@TKRZ=-FFjAJ0T`n9&|zBoU+Y9Ga)ZO*A$Xu#KyuV()QiygM&E@Ux`I4;_1hR4X5T z`jFNu$ZM&tt@Wgb+Swf=*U_vz%~> z6#3fbgA?LGt=&G!h=4Ek@T=qBMqHoBlf)KMljPX@6DYqvq^2ftC4q=gQ=Mm$D*UW# z!AIrZDpA~0CgN#~pUnk3c!4+XVOH}|8(@56&ZijtCyUg(Ikym8!|;6Sad0PPqTXZ@ zZwC>NyyF`zC!)_AtM6IeYJ7HX$pX9!Ax6)S4vjW9938Xl(kKi&y&fp$gkZSAm1Nld znEJ+$qw=#Ag2lj=v^{Ag(==ZxiV`1{u4Y??k zLL2#gNz=W2?hwySlY2R4*G z(5pJd^@wO@vrKJS0HPL92DtX37zUm|-pmTFVC1qXSZ=}Q+v z=70*%2vM}_JgwnI(^qCfCsugaYGXS@En!6?s6hv@QAlko*b^Fl#cb#qufjIdL4+08 zwizf!z|I26ccMbDn07NVXI@~UmXh3d#X{o-Ib;w28FQUg@dKi9R@^-C#-m;$SNtMa6KG)sB;Slt$DWV^&@syorM=FzPAmqU)K zJ5Mj&N4&2`w|wt^cgPk+YA`{6bXpNmfINq!>@u!WorxlR3mV7^*54T<`(Q=C?zNQxUm@7Zet2?(A%cG7J zU)@ASfVK;YIk~$((oN>qhN({i(o8im17F$K?7cpu{}PXD&9iG3Iqojw8VIeibwAMs9i^-89n9dg1UV ziKdL25eF7;E3kXmlzVzan3=PV0Pl?d3jhj=)g=b$^(QN(*Bjoq%cXXuk5E8g;t45n zz!dw?7LKJk>cg39#^vnCtYb#B9K%yUJwY#E(C56r$*8;iq;GI{gQ>R36FNiU#*7h< zw_>LBx3}#{IOd>ETUjt=nlo7GrQK}-<7(=fn*$ACJb%^6g^h8{_E|zUKGAl+BM*bRnh1Tj?rHH;=Bccy&&e zLA8!`JqO*0RNp&Uu(yLxg(R#~r`VNhA^UBciHEk5%e;QXj@k33^r;RSiR9``4})73 z7O@~|t3@VGuJO1g-WpK)yn9K?GSIKDf8Su0q|awMx0_IWJj7oY<>Zi0JY0sIutY6@ z@kc)QZ?F`KjdXsXfC>Y8RBwwk?r7c2^34p_q_5b#q%R1n{_uTkn^nZ;KlSnwM5B9QmK?7{25zkf16 z{8pu4`<|NzLyFVWkQ_|AXX$%=cHa|jYTrDcN70k;8OrP`_q8CccmD$MzcR=6Cn9Vi zL^a#qO2SOd?5-0Va!obF=cFt?#p}+M8*sS>Q&~v5=?VZo>Gi+$liz1~PRcE#f=n$= z@Kdjx!_=}g<4(x}Ro0S^Sdfpt5yb-!RD=MZ!o$wiFjDSo_*_)HucahuR&fhT1&%Kv zYh&q~Cr{jT@H_FCt!}~Pvp$nv6c;9aSS*4zNz$2*W3xnxaSu;!3a5B-c=urSxghL? zY3DDi5-j9AF1D*Eg%_I4)am81XUxwX&t&x?Y%RWB&*V_K`JN?S=V?in-Zr`@S6)+5 zq)?h*L4tb1!R3Rwe#xdP>5p2}9V8!*212gNhbYYDrjEGJeHYj@tKbp4lT@4UtLpff zjyQyPU(~u(A?NSgVue>8D@9h_^JDk0xc->$k!z^oBt7fMAoxsa9y2ZHo7PFx1!nyc z4#W|d=b$IY1-)VG6N2DMH?cXJzAqrmp9#IKhyL~lOX09$!M#Y@pK-=>vkDTGm2^k( zips-mbP46Pbm=0%=;R;zT5#Nyi-{YpScMv}!vnI|5M-5<=&ZVb0iapc7C?>i6n&~R z!G!z#HJ4}gL@cj%N_q4QUes30tFMsKmqLU-+-?tFJ52cHpAu#N!bi4vigxou0X+Hu zMHnzi=WOcE`q`HMwLVPyp>#4VI-KSfLABTTE7m0{eT|~--(2D8{Qo1HsTd-SmcG>i zI1~UH3^2Pdbph1fNrAu1bYsxb4MhwcP@^sq3JgyhB3Js}}G%9Mo@-_6kc)6^1;F~xrW3O*Pc>`4>-cF3&TIQN=$Jz^P zM?UjqJF*JZ>0?gCB$4`1)`~C_S=W0X#%O(fX=vOetknMDm6djeuqaztY0#WMsw$1WnJ;MK>4tOME6Xe!B-vRmFrJQlE}sq zA&*a4YG(vC<+6E|s~w^`$T}& zAGXhKhAl5JD~)lw!}LT(THu_5L?QKYXp5DLbf^ljQr){OXCaJTvxE@!-L9%T#UmAX zU8qN0EoAHx7F*yMud9S571Tg4-B}-1f0vh10Eo!mRa3tYf3WpCnL@5*Pv4d>wj)*< z59%{fwDi!}?)q#}Sb?cJXn@N)?fbewJb5(cr(OqGpePFn2c&gu9(qnpqh_fy`LxDg zN5B5&Cm_|X}Ovx)L8ywC6-r-GPbMx|GFn(vof8V zEIh%aLzG8NLTlecl_fXby&v4|wv-EU95LwV@_91WwcT_#k~$Z1lJT1l`U)S}SpAQs z_$J^ues11;yUcEb>*si~g*A~Dt@JKn3D9OV8OOkjl#nbk z0bfgpyJOj3Lz62yC#PO(i>}5JMYqxy%wYty7+sXTV3i(kT}%h8+TMsbr_ zvWeYT=a_g97O4^N+YXdhi2kC)K(4d z&*>l`>jeD$yfcF$^jE8;`CkMISq&as@bjC%Go!E?2DIvusD`jo&$>zFW#~eaRsj8h zW)hDtJ4y8i=X(pgk8w#i`A**(|Y$ah;6 zb90LQp9qBJmpzs-db_-Qgu~7Ckqkw70XFl-3f4q?N}TQccEMG@ly!Pc5Z%riCYID0 zx=3s>y>d$%;l$Vh>`VS3=7alP_f5(HQl3}@2yxsJ=JBOb?H44YllXnoWf{VO0-nO3 z!^{h5i zH`%h+M)cC0?OpAske+u@=K~-IErStMseJtC?zTNX6Wm%wwdgrc^C1Sc!%|1o;X!iH zT9;n5gC$p5WM2I^b2(dEw(a!oP@Qtmi%%)oB#S`${)Zv1KfrjZ)^bP&2pxiVe6F(5 zb~ShH(ihRL7U-eEA@i)WCUQONmM>@GO=@TT{x%xPclU@L==%f$q1m>gYfT<$xM>tP zu!G!_p2;N5c@Av>Z7M4gvZciFJMBe^^2r8{WacieXBqvtmlxCo7k zjdcyv6jgf~H)u19Gf*ZpHg|;ATN@6*z^o`41x&e~5(8d7z`xm!?1@EMY7EgtEH8Al z9nt~N$lp}soYO%_g{>e5m_v!4ihEu(&OrU`fFp`FCx%|};|a7$NtF19>|;!oR`M&Y z;j%ZQ;JPzw*hJHLj=o+c%jD@}$a-`|aefwVUMVk1$*u^2voK@H@JVC=OcV@E-UQ+R zrA<<_gg4dhVwz$4%eDSzrOu@SF&&G-$4eL8KUOmK5f$SAQlqA+_G>stR2T<=!xQA( z`B|xIiIK*`UC;WEtfrB*VD*M(A)fNuXKP*q@ZY{JRR^r0ESRJP7_5>oh-&BX9?it$ z4~Dc~#BOR4-K)mEM4VW^$hpX$Wai5SFzF#P5_pue%#dfumg8MXzA4y$n9p%owpp{rMHh~ zC|fTh`j&q+;J*5`Lt1s*1?Dgu`qKd6l1K5lP7`BZeNpJ!nc#%^UPU6<^udfE?Rc%R z_nRJVc2~`2E9ZPFs%1>qK9NB?H2q}Gq=RF0SB3HX%z}!Y7=rbC$QbpDX=*Cw0&F_y zI7U}*CS*!|>Q?tzoK0@O6RamxJf$d4=MnQvd_=zFkWoz`ieOhd%!e|xYQ{2AO=@>m zKVH5}L}`BAW7qAIDMQL?ifV6OgeMgD#-R&damZn9pg!ei8A!#$n%TOQoVe7LOfgQW zBVYJ(#HiizsVhE%=AacdBcJm}>z+)aoa2paK;m6+{p1mQ8ke}LH|?f%EH5kj&gIA2 z5_XMa(nzywLfC7$(Yz#YYv$_A(XjZJ?G?|4%d@20v-f73h;N-%I!zI8qHc>@Ux;#K zQ(J?q!z2}z%;ojCZgsP!tC-fbL;SWAt(iHn9WG73V82NJ?JT*C6&2QMl#EzWo@IY# zDlT7J)}TW5)s7|n4H40UoZ!^ArzY#^50MO?R=6Mb`)p3VG9|FZi5sUm@N4CxVd^8Y z#r>$QVRT-^X;>yG?#D8s2gY@0Y*~ zodoSTI1X_Js(*0KfDNp$W92L*9w=icgIbTJV;qkf4GU1_n@tU-+7~MC>oRO>s#P$5 zWwvqolUrh?j#Y6p*sJ70d=MIA@5f68!_RRPTgF9WdIDBm7}hz zKMFRPwyVlmMczFbdhU>0gJhr>;Sv;2(Xkdx9h3FOy;~J!5_wBS`rT{CU{)!{!+3#h z?{bmuMmQ{e5ZqZ0J+O)+!+5Mhjg^QMom;*%S?_??ksJ`_u9Z7e9D`xwZRWXnDxO*? zdZ-S5_!?!GwUa=kR$yv`AS#o!_HZ@BXHE)xV}cb#BFY*88mL6Y6ckrW&!QPs+0Kor zOWh}-M+02UF~6Tbo)vmfD@wf=@jd~`uGyHOB0P(&SAE@_A$X9IyY+o09E-1Zng?Iv zzJ^og#8@-U_N;r>z+}GHhEcDbeT|ot#q7&9Fo!&yOSif@?u+@4*rUv96ci;c*|VxP z4nOdoGhsW-yE@~FY2u0I`e`+!oq?nE0vsOKiqta+nLb*%M!0fb5c0%WFPA>V($dFM%* zR(n+$NW$9<_L*8qjCQj21`BQ$dD9QXjZGcs8{dABu44odx%Bp~=0~n9MxOeYsP^Ao zn_uXp6v|Tud%MJgu0DxiyMz2~v6#%iT&I6e8Ud#>LPt*|U#MwMij+!Be0x~@VNP+9 z&Xf6i>3A!;^xTQ8EoawAz*G9e7^Myza@aU($w3K@&L%#^>Q-u!T3X4ACZ`;kH?vO~ zHJzcC9%`&RWuSGNU>0e;UQV6@Sjg@$U*?5 z@4rdM=tufN{sP5@g|_T{+d2(u%4;-UL)?Cr5CjOlrP8Ss(E%%!BYgo&cGqF80ay5d zmsTAxNf+IlgqwoZhY!{=v>Q|*>j`~wt!V>6&Y%nJ0D=?qKp-i;4;XQ|R-)aeyw;@m zC&mR-;F1=5DiP10*&b%I#f9-6bo4BLX5qJ-ic4k-AjWk--{_3*eBeHSi?bbZY&@9j zO@m@kw>>_;#H|$QSEFr9)(wpzLce^VJ~GErT9*?yllp0|*K(gduzAu?Rw_DQ<(+?l zS^=5oQ`-bwQ$f?fx)YQZF9yMVzRq;sRCk{PXp%TeL$6ZHb@O7D&51q=`fhgJZy;gN z6`;-f=pjc#?f z->;$Ba@b|Sd6A~2E~T53Vlb;f1&~4_rr3z9p05t@p|4CLJoWaiFJMi@(wXVVf;cqV za}}?|DAw3AC}SIc3yb}H+tR6WzN;aCF>VYgm8-9+uaDIJEN>g6GW)vQQhwyiTM`ly zQhDf$$sff#*T1z@-nmxdXIKT}7}A1Pz6L5`GbqWMS^5_mp<-T6!V`RL1kXG5hW!&~ zHH2@(rqxN4Z`4i4%VXDKzLLnWG@8{TR;jBF>`m2A4EF~!3TI<+CI+cE>+XG7AZO!* z3e^pe3JBjQc6%?5JQZF6NlqY9rb7-dlqn%*n7~ zF|$%pQ*;eEkid_#_5BT0y$k1Ly`J`&K3G(kj;^3^mC+sQh4R9~SOuNTk&>Z7^}5Bb z2D5v@p0e5(rv!<^$VAJDy6xaXx@AmTtlTD~Sx7#@>FHphWBI(o*2~?wPcbozRyOx~ zu3{L@ONy!GpJM|I2Os3YD`LLhj|iJpmeI5G<7;E1g)&3wsx~{}$E-nbE?$eCLJrJw zv-n%WswNYMXL=kgI6tV|^-lApo+32b0|nUoB>-^LQ<5#fJul;0XtPPZu@Vqu(L4l< z=I#C@yWk?6TgWW{3?b4u{!w|{kCyZUkSXuX@8C5>i|nK4X-47=R8@C;=)P4@nPC4{ zuuAh=s~cGD+aQAFAkq>A#!z$Q-YVk_0RU-*HGxsVrW0qS)kAi{fwwEX%HVh1B|JjwbU>JJscwtQeOS&Ud#;`FqU{RZ}#tFsV$E(^ z9n0!D-Z)Vtzd#Hj_*y-?a5wS9ZpXK{{m7ZXn$rgRd(d0VozN$hK?aVDkciRYDhCA# z4L6wSw4q%jXF;f{uZrJLm1>L2x4{W&xRfvW@1c-#mB7HGYvpced}n69MX}KlZ;} z-f)I24`1VvKY< z#BRShJM(4NL_n6 z*b&TrB{L%1?0jo&$rTH;_U|68Za8vpB;uE@!`spocq@8>S?!q$@y4&+M)M=9+UTYp znznQLZG@6?TG^R4mTkQD^>W0h1Q_bI>jW1JqN9f&)*F+pRJ|IK3HkBul2@2p?Jx^i zL0c?DB%+&v{2wHMnTf0WICFK-q3wq#6>_&%e*NKx61{_4^V2INtKM7U3SPPt@ZXfx z<)+535@h-Cpf5%BwfXR+zLBI98t+yI*jF!XTvc?O648(JjMShd@&Y;k^$~=Dm=EZc zpOD_Cu>2DfyNF-_lnAh#07pKJJLc+Nmw&UkEeFOZC_sKp>E5#ohn3@g1qg-UE=i)w{wk*iTi=Fsl={v#~ z#MMvZ(I;$AgK*NVMOUc4Vf$=^ifoC{cfGfDyG{&;*+?-(}JXM*~0Gi*H4IR zUl!E_b>J@v;rCwL-}y*2#c?b93tW6b=>KcBRz$ns=EbriWuWP+kVObf=0OzL9Fo6} zW*rj{_m|o=s8HaO7p1bxO1!mz+-JIf?dRX^6t z#?2{PP^>US6si;Sv@ z5)dKM7jy%rGCd2cG1q_n!Jw*e?&YAo0k6skKAl^0>q1!< zj}J&ilNDDbYQadLJ4s}3#x%Zv=rE82&c$%E*^FNF?_Ty)L6EH|sk_ReG19oHAm47e zGlv2mWw`jiP#i3zSs`cZwfPu+D|&>6Jak+gzaiZOlCO zfP2qC>iMz%MN^5|FNGHz)t<3`f!=qI#zu?%wszvd=u0av{kTPq2|sgs&|JDIss9>1 zR9hXw$Z@+#Y1M%z13k%8 z4N5!5v3dUroU%~1e(`&cOb_>eH4vPpemu!4KES@qgTCHW z0%sXI-)YGILvha|h5S$tda=zZy=1C%WtO^@lZWVV*?27t3pWK@*Z!_&&)~hhtduF%S0J zAX#h%S?v3=>tr&daPQp8P}MJ=`K#Bp3T;%vIjfB|hK-DtOl9 zH}V%qU=F>)c%xXy8(=fORXTjFtvC0uT<^nk5%v6?BXdbx{v98+rO-$1drDL87ZbXi zu$-`-2~PU&w*zIjZQg@uQikbkrb0xn-SCni;9>d8wAo9&tRSv(iZg1;a2T@wk!Ade zLGi|c}?=>$2npHsm{h`8o3o37eSK+DOQ$KzZJDh(;J~b(^i)5O+$P- zTXRcUeG9o+>gsg4#-tZr0x_?h@yIbYeWV&z4Om9Fxg-?4++nMR0{TG)YcniZ0l%j; zRd1Wzb`V7ZN7Q@T^L9pYNJ7=3MQ~{QR!55E91-n|G6C+fu8hO_S~_?JH_Za`lZ_*0 z{}vk=_k#dGm&>}n$&j()_^Z;y!FkiN*|9H4znO=t0Z|Hmn^jJkXW#Psi_;j@o2TGx zCFl5hB7OLSmTJEn-jolqSgqE#1jpZN6oh6v#7m2Bk;220gq)XbsGj@HwZeO(E2Mk4Tp^3R#P=6fOI%2TqAt~x*kSbI7erpArTm-yOIZby8I^mkX0B9I%g&yg z4MWY@VXQr87&M7#>|8&iqSJRWsY+)ZKky)w_svG&rg`~&cxMUZsYfZ`??9e9t3jlU z&8P|D6Br^vManf#2&Sm|OZncJQO4U6eo(b*dnTxd#qpVUX}7iPraNcMJ>z~`w!O1j zj*(Phz@wR?l2Yt;8)@#}pKzIVIpj!05U${#30-2~nR~@7!l%7XoLx=u+Gi@C{dlm- z7ecn^QuX|#IFk{YwUx-Ha9FT{ZB4pEQ->N&mj8EGtyi(rTU5-f%(#KPS;t4=x8f_I z#}}WUB;wTfnf>V0fj3xZu-ajoE^+NM;7clae58dZcDz*88?_kw%u?)$H1TeQHbgEz zP+=AA9b!3s-7;m(Fe#I~Or>~sfi}{r zrmnI+e$)f1!AqFOr(4Fyqoj#1N?!oXo>AVvW^Jdc!(B6MC(gnjH^-}};+I77N2gjT zzOKNEa>v7tx9>plYqeC?KQx4uC*tov5QRT4`SI3HbKVtUi*p|Y>P4{FP)rI(y0g@! zFTk-M$gpq;36UmUwx_;}SYGE+Zu@0q4m~rKR~swI%h*pa@2u!@U@G^|jR;Z{vvTzz zHF1-vUPq9jiLzmSXNyh7o-J4URG1?U`v7R z`Q_r|;yBc3yQpbVddzWYTP&98K~8i#LNpv+$srpR>K7Scvi9HoWkH&UFN8w7ahLfZ z6!vpn$&Pbu+=mv`ydt9}yRtri;b~M!g>1E)-3TmTvw0pA{H#(s4i<3V@(M zyGG_;AOK-u7yZ7iSa|>D)qGZrSKhKqhCP>rM%JuBo3Bfep~4@}Et@ns74*&^F$k<> z==G6#q!@5x+S9HMKa{c6T?h4$%Hg>jsVA9_W==eOoruP6_Hzhq8FU+q0FC~vw3s?8 zJQp6bXbL&997O3>^h4_aO+84=KkTi*)9b#@Mw2s1)V;^7ISB@%Y81 zJMgb^!fP#Ha4(bq7#ZbWl%A>GNDOh-Ut3b&{c&09sYMDU{c&KbY?(Qfo~UR)$Y{k> zT2lwSty2DYTRrH3Ib$h(p|2nfBGztI*QN!y@ij{4b}p0TL>~y&B;Ipah!iVT zdJV0^j$**`89Q$IVvT(Cy(YnqvflFHCyHrJ#5~9qZyf93-)F%82c!`>>Wag85oNdG zp>T49Yp`s?qo;eGHzfbUUS!tclKyW%qKjmtMxZM~@F?8uImJ|eCACgm>X6x#0`c?A z0i#5rY}57#HkgLQKLhXnllDR+Vm2C@fh5rJ^>+FI2>N?^kN$G(tfB7K5b3$yE5aTv zee|c@KquEEY?=LcAkH}Y*;u1|;a>TRok_1*a)OfT3Uexnolk?-491SSYlLGGsmO>7 zf|AD#V;(yP007Ti_sS|dnB8J_o{kKmsbFsgnOsssMqk8yjGLP=+yNK#LIxqYYTwzR zX{Uzzlf_Mj?r5BC?Qsd=Vsli*ROh1US1+1?B#Nd&hjxP=DLb3F*beU*Xo~vBtM;7> zhNn_5{B;1`_7df|_(|CUIwKj#74yk4aLewh@UbFc*Be8aQ8lY>)l9-nIV?^~sNyk| z?)KyQ0EF7W1ez7)rExr%T|7H~*RXF)0UH)@eH2iO{}vxFA=NT?vVx;958~8mM1N6O zY>3DkD6l_nOlU_~d7cKnmNnoq|)*w#jnb? z;`qS{GHpQ4cvS1jiCe3rAso@Q^^=)j=3gc}Tb{;85d3yI2}V!8P*X*uQ;r;MmX8aj zK)Bvnt5`bB$hA@|d}@6X#tOB?;x{-(!+lrfy&Wj3So-medk(3+jGyDIravI)3z&3} zuyo#VcC|C38c#&Gt0{^*an0#8kJIJk-Xdbl!UJ2oA1E@+DV5S(Cq?@SXYYpR?b0+= zQ96=eebDHm4Vjl|->O#5S6O2uRVXDUJ_mb@M}E1D*H~hH;$?~PWxH+w^wtpKWHNd# zZmP$|ad`Nn^mcna^=J3k?-FRf-5?nSez(hL<=bPan*6*p33j?m)qGm-&qh(zk>m&L zuMP=_>K}bn!do{tV{}5=Q-?)4EVfxTA79nlTJWWsDm|p&AQ2J8Vor6W7&Sp%cFdYq zr+Y-JFZQjTEO23Ss8Z-*C42%j<5{sHK*LqX=bz1}gXQYsL7M|l4Aw(ma?`FQbR>!l zNhF;AYNk5&j9XD=F(za;*h*BP-ipA9{$Zr(A*ZC76%=DlzBfbzaqZ9vQ>&ZERPDW> z%A&SdIjxGqYD1WPQ~3A{#f#`)CDgfT;dv|kTg=IBe9_<1 z886=dT(6svbn#f3{5)R9G?AZcox!;Z817cN7lAZhSG~! ze#sr3yvPESaFMjeBCqh9=bSjZ*P^=Ug5>Z@UVS=yRt|3=-c4?abU(0#WhCUj{<*kt z#h-0C;3cwLNYK+<81GG0?z^TeutdGEJ=IJ1NUvUIe527un(b7G;!iBD;a*&1&`doh^NXypz6ZZKazrI>XWB61%tH0|(VKMyK6^xHZvA=L**PI_6f{8oCU zNxh?Ms{RgDX(uS6kscS&U-VovB8-v83d>uhp9vWSTF_th3ghq-1kkfJ<_ZtMJy$COECnSnLaQzTJU(}5!1_ZywCSb zormfIdwuGu6Vig(8u=|Jk&yl<3_X0@ZM}LZ5{?}VC6Vdbq74D@F=Fj`-piZsjMMO} zCdKZ5l+F3GOCr9iq&2hu$@_JWD}9CqSWH;)+vDFf^b=K-LPhsgEIhbr)APePTpH3< zt|$G_QpXo5)_J|f_Iek>l-0e?Pb>_6mC>b=P_j~s(CHTaOQ_V0m3Y`bI$2kI_r*Hu zQ!l(KjHW$?-vJ-R2F8r)HZ{MZ`9Hf_Zo|Q9#2g zq2BM07Rx|5y@OoENV{9z=*#ErdKz`7;b!HVyBab-Jga4V|G*kNou=(QK~*}#*)7AGCwp5|hH~L>LvcO-=@I{&+bNBnMEC2$ zZ&|!rra*yP!`ykW*~1C*>y7qM_ADFTWO*P)AP&hL`m!drN3UhO4fVS}_ zx^+JR%G?f3T@Sv+5~sN+EVSE{E}JoGloL+$Et)H3q6%6gP}URuaQznKy-gg=ZwQcj zI}+>X`#vluPdKKk%U!+FcbWGO^8vvMcWR76i&o|)42BKn8!z$AS#WC-CAjHs2PH*{ z!2v|OagdG1H|;~e(`F}C-`xt;(pnKm{cz1p?@~6o&72%)iL9b6U`>{km6IXMIVBqo zaSQptgIy?riA$`sO8RIA%^0NkwQ5X3u*y}Qrri2rZ@IxTgNWr(4ZU*aJ=p(ACuEUp z7@nqgrkbyS*_XJ%ar9s(GjUd1{a}r#qSHCdj3qi*8LUZezh36|tAy*m zC#3bCV!_1BIAQ?s6<-tDl1CEf-Xw}xH&I?D>C71|sW?jeP>?0+QuqC26UBOxg53Ax z&ykAEe%iQbK5ONb>_!)_oTn%%Vt3E-T+(OR22;G0(pJbQc-q9CH@DN3#g%K4Nys@@ zn1{;bkcuVHO}ls1freB6M(`-s)H{-l*4VzM;{||%FvPab z2+;yitVaLX!u=bb+c}*Zd$s5It>gVaktvnd8OHN z^Az7kM<#C_ssfQ;V{EvT7A;(Th)CFLeg;bX1$upa^redZK*zAEW^AivV#WJeV0CA8 z>H|yNw|qVF^yp$j(nfnVVHG41jZi-JtU~r~64D|4diVJCGq;o5p8FPp+MSQrqP(6p z^s!+LJ^62yx>bpYm#T`RD`VI%i=7cJTR~!noPjDV@ov2mTglVFZp}2Zui8|P$K%5| z)@#e_44IAd;ulrxaIuGrp7U&RX-IbspMce!i0z%MOI4h(@0*$0Y3@|Nuv%FE)jiX1 zS=qB>`m@0v*dM3cr$vztPKp&(l?fUX&uRvKSKc3)`=#&@QR&3;f|qV>^L#g;ow7ix z(wBTV{DYGqvoB}c_31!=I94 z==rjIe-}oM;!MN{0sI;N1vYO(Jer+w3Ud-5ALLgZ)bR)y^ zb;VH!b{ioyibVFnp#S)?OW%wP$(6u497?;*q{=RyCR=h_qez4pU?V8(#ym9(he298 z&kMd^IbFC;a((KI8aqX^AO&gEKby;)beP!a9lxa15o zOo1ZYlo0;z<`<@lq6$CdeQgN32<-Wk=&K_8!*n&4tj;)Fhh7a&NZ=GL8KDkz_JhX;FqSu8y4{9q-8sox1Y}7+#1NJc#w(o~c zq7}K>7X1V1=t^1eGwQ`s&G#eWU#!e=oha{mPD+5&+@h z6QvSNUm`yLhLqfVEli`kSRtsxztop*{v_gxfX9UFqQL32zwpD4AgNJZ0*`f<T&kM%GBgy+2u54I=Q&JCi zjq=0bao88fV7}eUBv03(G?VZJk8m^#o63d9(hGw6nArlF(_SV^pk~qeia!*zbo|-hLdEceH2*Rjmp$?@D|& z@Wxh-uXNL9aDgl#4k~8Z1hPW8wI=zH)ChFj=GT}KEWM7h{7m*P-&f2e@s7L6lepqIht30R7Q&%VCaEXzpu}t_Z05IMR}O)oeCU zd+U~*VTAPjv@-X_;oL!+&@k6@HzwY5ix!G8TerrF87Ohc9oGD!_P8`$pwLl|iW_Db zF*vxU>03@0F&BPHfcZ>4 z5HHs&IUg#)|8p@nJ*(KbP5f6giL$O`2uJJv%nO!?TvF)1PN&%7k3y=02kpCylX&xn zC1wVff#gJ6)qvVFCLl9n1m9?Z%H74Plo%OyfPat@lN%emAh5jDt6(eAwq*KLate=9T@u=1>lI%w@x(?pVvo!Pa) zZL=)?|#UWOw?A|5>&wm>gLr)~C!%JqW5D z89M)=pj);4dDH$}h1^AA9i?EkI!ZS_{~sd+jwa9NKhwpvJ*#TuT~=)&-TPw(@iTpq z6PK7zR7Jc$>nW7CJ)L*Re!2o_xv+<3A5;5TFr4_vP$A7i!!&Io;Y-cahRxrZsw4tS`9ki`11_rRn-BeAUTylqvu z>jQt@^AGxi$*{LF* z&pvaxcw0-l;#S6|!5DLXdr&YhXl)WaFZ zQsBr^&QfwTP)>Bud(|~0oOquzNd6m7DB`!;dc z#@{B-!oq6j2w4g6N4W*Lq1yINiGIlC)iAijgO`f=c6#x4oP06geEPdbMH!zezlMMW3$K zW&eAVqSwv+mOJ78-VJ@R@4PKt?OTXO?iLo%*(5_Sq{>=3NjvF5!{ryt~h`cjJ#xP8CcEF@-8GBw&n`)Q9wJTS-Xww|QW7)edE!e&7;+qBg=c@1NV)15{*THn(PgtsxnFuGT)cZ{i;JLUk(4(}OOzPQAP8RnTD>AI>LJV@^s2pZZW7`X~Iu>!Vg%w|<+NX}f1u z_RXk+>AQY;gvo0z_r_zj**7#6@N4b(QMzV%-}^twsj54!uznG`by#Kj z=?a^LGAHiqpE^=v^F=h=>$2iCJJT#?{~f|FqH2IshT(#%=1zR2>f`G9@590RwN-zC zYn0!9XsdtIz52$Y%a1>rL~Ysrt(?#NxRUs(;~L5nd%ivRveNW?d)2>{>t0>Ai`_D9 z*Mn`@EA9tQGufVTg+0tgexFRua_#20oxpg#BBkH6MP9t3JMfZ@$?Vg2DAL3 zU78=Bx0dXcT`IJ9?ye)37#q&8s{q%{ZA4mj32g5%9(=t^zw^P53G>u0tSx!HqHlu9 z_r7;e7b)&}tfVT`Ej-%lT>J}t=xBmB literal 0 HcmV?d00001 diff --git a/PyTorch/Segmentation/nnUNet/main.py b/PyTorch/Segmentation/nnUNet/main.py index be4e2efd..be131d4b 100755 --- a/PyTorch/Segmentation/nnUNet/main.py +++ b/PyTorch/Segmentation/nnUNet/main.py @@ -14,9 +14,10 @@ import os +import nvidia_dlprof_pytorch_nvtx import torch from pytorch_lightning import Trainer, seed_everything -from pytorch_lightning.callbacks import EarlyStopping, ModelCheckpoint +from pytorch_lightning.callbacks import EarlyStopping, ModelCheckpoint, early_stopping from data_loading.data_module import DataModule from models.nn_unet import NNUnet @@ -28,8 +29,6 @@ if __name__ == "__main__": args = get_main_args() if args.profile: - import nvidia_dlprof_pytorch_nvtx - nvidia_dlprof_pytorch_nvtx.init() print("Profiling enabled") @@ -61,9 +60,13 @@ if __name__ == "__main__": ] elif args.exec_mode == "train": model = NNUnet(args) + early_stopping = EarlyStopping(monitor="dice_mean", patience=args.patience, verbose=True, mode="max") + callbacks = [early_stopping] if args.save_ckpt: - model_ckpt = ModelCheckpoint(monitor="dice_sum", mode="max", save_last=True) - callbacks = [EarlyStopping(monitor="dice_sum", patience=args.patience, verbose=True, mode="max")] + model_ckpt = ModelCheckpoint( + filename="{epoch}-{dice_mean:.2f}", monitor="dice_mean", mode="max", save_last=True + ) + callbacks.append(model_ckpt) else: # Evaluation or inference if ckpt_path is not None: model = NNUnet.load_from_checkpoint(ckpt_path) @@ -76,8 +79,8 @@ if __name__ == "__main__": precision=16 if args.amp else 32, benchmark=True, deterministic=False, - min_epochs=args.min_epochs, - max_epochs=args.max_epochs, + min_epochs=args.epochs, + max_epochs=args.epochs, sync_batchnorm=args.sync_batchnorm, gradient_clip_val=args.gradient_clip_val, callbacks=callbacks, @@ -85,7 +88,6 @@ if __name__ == "__main__": default_root_dir=args.results, resume_from_checkpoint=ckpt_path, accelerator="ddp" if args.gpus > 1 else None, - checkpoint_callback=model_ckpt, limit_train_batches=1.0 if args.train_batches == 0 else args.train_batches, limit_val_batches=1.0 if args.test_batches == 0 else args.test_batches, limit_test_batches=1.0 if args.test_batches == 0 else args.test_batches, @@ -106,6 +108,9 @@ if __name__ == "__main__": trainer.test(model, test_dataloaders=data_module.test_dataloader()) elif args.exec_mode == "train": trainer.fit(model, data_module) + if is_main_process(): + logname = args.logname if args.logname is not None else "train_log.json" + log(logname, torch.tensor(model.best_mean_dice), results=args.results) elif args.exec_mode == "evaluate": model.args = args trainer.test(model, test_dataloaders=data_module.val_dataloader()) @@ -113,13 +118,14 @@ if __name__ == "__main__": logname = args.logname if args.logname is not None else "eval_log.json" log(logname, model.eval_dice, results=args.results) elif args.exec_mode == "predict": - model.args = args if args.save_preds: - prec = "amp" if args.amp else "fp32" - dir_name = f"preds_task_{args.task}_dim_{args.dim}_fold_{args.fold}_{prec}" + ckpt_name = "_".join(args.ckpt_path.split("/")[-1].split(".")[:-1]) + dir_name = f"predictions_{ckpt_name}" + dir_name += f"_task={model.args.task}_fold={model.args.fold}" if args.tta: dir_name += "_tta" save_dir = os.path.join(args.results, dir_name) model.save_dir = save_dir make_empty_dir(save_dir) + model.args = args trainer.test(model, test_dataloaders=data_module.test_dataloader()) diff --git a/PyTorch/Segmentation/nnUNet/models/layers.py b/PyTorch/Segmentation/nnUNet/models/layers.py index 92c19683..d8f6981a 100644 --- a/PyTorch/Segmentation/nnUNet/models/layers.py +++ b/PyTorch/Segmentation/nnUNet/models/layers.py @@ -67,7 +67,6 @@ def get_output_padding(kernel_size, stride, padding): return out_padding if len(out_padding) > 1 else out_padding[0] - class ConvLayer(nn.Module): def __init__(self, in_channels, out_channels, kernel_size, stride, **kwargs): super(ConvLayer, self).__init__() @@ -94,30 +93,6 @@ class ConvBlock(nn.Module): return out -class ResidBlock(nn.Module): - def __init__(self, in_channels, out_channels, kernel_size, stride, **kwargs): - super(ResidBlock, self).__init__() - self.conv1 = ConvLayer(in_channels, out_channels, kernel_size, stride, **kwargs) - self.conv2 = get_conv(out_channels, out_channels, kernel_size, 1, kwargs["dim"]) - self.norm = get_norm(kwargs["norm"], out_channels) - self.lrelu = nn.LeakyReLU(negative_slope=kwargs["negative_slope"], inplace=True) - self.downsample = None - if max(stride) > 1 or in_channels != out_channels: - self.downsample = get_conv(in_channels, out_channels, kernel_size, stride, kwargs["dim"]) - self.norm_res = get_norm(kwargs["norm"], out_channels) - - def forward(self, input_data): - residual = input_data - out = self.conv1(input_data) - out = self.conv2(out) - out = self.norm(out) - if self.downsample is not None: - residual = self.downsample(residual) - residual = self.norm_res(residual) - out = self.lrelu(out + residual) - return out - - class UpsampleBlock(nn.Module): def __init__(self, in_channels, out_channels, kernel_size, stride, **kwargs): super(UpsampleBlock, self).__init__() diff --git a/PyTorch/Segmentation/nnUNet/models/loss.py b/PyTorch/Segmentation/nnUNet/models/loss.py index 88ebe167..a31e7ee6 100644 --- a/PyTorch/Segmentation/nnUNet/models/loss.py +++ b/PyTorch/Segmentation/nnUNet/models/loss.py @@ -13,21 +13,32 @@ # limitations under the License. import torch.nn as nn -from monai.losses import DiceLoss, FocalLoss +from monai.losses import DiceCELoss, DiceFocalLoss, DiceLoss, FocalLoss class Loss(nn.Module): def __init__(self, focal): super(Loss, self).__init__() - self.dice = DiceLoss(include_background=False, softmax=True, to_onehot_y=True, batch=True) - self.focal = FocalLoss(gamma=2.0) - self.cross_entropy = nn.CrossEntropyLoss() - self.use_focal = focal + if focal: + self.loss = DiceFocalLoss(gamma=2.0, softmax=True, to_onehot_y=True, batch=True) + else: + self.loss = DiceCELoss(softmax=True, to_onehot_y=True, batch=True) def forward(self, y_pred, y_true): - loss = self.dice(y_pred, y_true) - if self.use_focal: - loss += self.focal(y_pred, y_true) - else: - loss += self.cross_entropy(y_pred, y_true[:, 0].long()) - return loss + return self.loss(y_pred, y_true) + + +class LossBraTS(nn.Module): + def __init__(self, focal): + super(LossBraTS, self).__init__() + self.dice = DiceLoss(sigmoid=True, batch=True) + self.ce = FocalLoss(gamma=2.0, to_onehot_y=False) if focal else nn.BCEWithLogitsLoss() + + def _loss(self, p, y): + return self.dice(p, y) + self.ce(p, y.float()) + + def forward(self, p, y): + y_wt, y_tc, y_et = y > 0, ((y == 1) + (y == 3)) > 0, y == 3 + p_wt, p_tc, p_et = p[:, 0].unsqueeze(1), p[:, 1].unsqueeze(1), p[:, 2].unsqueeze(1) + l_wt, l_tc, l_et = self._loss(p_wt, y_wt), self._loss(p_tc, y_tc), self._loss(p_et, y_et) + return l_wt + l_tc + l_et diff --git a/PyTorch/Segmentation/nnUNet/models/metrics.py b/PyTorch/Segmentation/nnUNet/models/metrics.py index 7d9392ac..91d0a8d7 100644 --- a/PyTorch/Segmentation/nnUNet/models/metrics.py +++ b/PyTorch/Segmentation/nnUNet/models/metrics.py @@ -13,35 +13,61 @@ # limitations under the License. import torch -from pytorch_lightning.metrics.functional import stat_scores -from pytorch_lightning.metrics.metric import Metric +from torchmetrics import Metric class Dice(Metric): - def __init__(self, nclass): - super().__init__(dist_sync_on_step=True) - self.add_state("n_updates", default=torch.zeros(1), dist_reduce_fx="sum") - self.add_state("dice", default=torch.zeros((nclass,)), dist_reduce_fx="sum") + def __init__(self, n_class, brats): + super().__init__(dist_sync_on_step=False) + self.n_class = n_class + self.brats = brats + self.add_state("steps", default=torch.zeros(1), dist_reduce_fx="sum") + self.add_state("dice", default=torch.zeros((n_class,)), dist_reduce_fx="sum") + self.add_state("loss", default=torch.zeros(1), dist_reduce_fx="sum") - def update(self, pred, target): - self.n_updates += 1 - self.dice += self.compute_stats(pred, target) + def update(self, preds, target, loss): + self.steps += 1 + self.dice += self.compute_stats_brats(preds, target) if self.brats else self.compute_stats(preds, target) + self.loss += loss def compute(self): - return 100 * self.dice / self.n_updates + return 100 * self.dice / self.steps, self.loss / self.steps - @staticmethod - def compute_stats(pred, target): - num_classes = pred.shape[1] - scores = torch.zeros(num_classes - 1, device=pred.device, dtype=torch.float32) - for i in range(1, num_classes): - if (target != i).all(): + def compute_stats_brats(self, p, y): + scores = torch.zeros(self.n_class, device=p.device, dtype=torch.float32) + p = (torch.sigmoid(p) > 0.5).int() + y_wt, y_tc, y_et = y > 0, ((y == 1) + (y == 3)) > 0, y == 3 + y = torch.stack([y_wt, y_tc, y_et], dim=1) + + for i in range(self.n_class): + p_i, y_i = p[:, i], y[:, i] + if (y_i != 1).all(): # no foreground class - _, _pred = torch.max(pred, 1) - scores[i - 1] += 1 if (_pred != i).all() else 0 + scores[i - 1] += 1 if (p_i != 1).all() else 0 continue - _tp, _fp, _tn, _fn, _ = stat_scores(pred=pred, target=target, class_index=i) - denom = (2 * _tp + _fp + _fn).to(torch.float) - score_cls = (2 * _tp).to(torch.float) / denom if torch.is_nonzero(denom) else 0.0 + tp, fn, fp = self.get_stats(p_i, y_i, 1) + denom = (2 * tp + fp + fn).to(torch.float) + score_cls = (2 * tp).to(torch.float) / denom if torch.is_nonzero(denom) else 0.0 scores[i - 1] += score_cls return scores + + def compute_stats(self, preds, target): + scores = torch.zeros(self.n_class, device=preds.device, dtype=torch.float32) + preds = torch.argmax(preds, dim=1) + for i in range(1, self.n_class + 1): + if (target != i).all(): + # no foreground class + scores[i - 1] += 1 if (preds != i).all() else 0 + continue + tp, fn, fp = self.get_stats(preds, target, i) + denom = (2 * tp + fp + fn).to(torch.float) + score_cls = (2 * tp).to(torch.float) / denom if torch.is_nonzero(denom) else 0.0 + scores[i - 1] += score_cls + return scores + + @staticmethod + def get_stats(preds, target, class_idx): + tp = torch.logical_and(preds == class_idx, target == class_idx).sum() + fn = torch.logical_and(preds != class_idx, target == class_idx).sum() + fp = torch.logical_and(preds == class_idx, target != class_idx).sum() + return tp, fn, fp diff --git a/PyTorch/Segmentation/nnUNet/models/nn_unet.py b/PyTorch/Segmentation/nnUNet/models/nn_unet.py index 34b576f0..892b7afb 100644 --- a/PyTorch/Segmentation/nnUNet/models/nn_unet.py +++ b/PyTorch/Segmentation/nnUNet/models/nn_unet.py @@ -20,8 +20,9 @@ import torch import torch.nn as nn from apex.optimizers import FusedAdam, FusedSGD from monai.inferers import sliding_window_inference +from scipy.special import expit, softmax from skimage.transform import resize -from torch_optimizer import RAdam +from utils.scheduler import WarmupCosineSchedule from utils.utils import ( flip, get_dllogger, @@ -33,7 +34,7 @@ from utils.utils import ( layout_2d, ) -from models.loss import Loss +from models.loss import Loss, LossBraTS from models.metrics import Dice from models.unet import UNet @@ -41,24 +42,25 @@ from models.unet import UNet class NNUnet(pl.LightningModule): def __init__(self, args, bermuda=False, data_dir=None): super(NNUnet, self).__init__() + self.save_hyperparameters() self.args = args self.bermuda = bermuda if data_dir is not None: self.args.data = data_dir - self.save_hyperparameters() self.build_nnunet() - self.best_sum = 0 - self.best_sum_epoch = 0 + self.best_mean = 0 + self.best_mean_epoch = 0 self.best_dice = self.n_class * [0] self.best_epoch = self.n_class * [0] - self.best_sum_dice = self.n_class * [0] + self.best_mean_dice = self.n_class * [0] self.test_idx = 0 self.test_imgs = [] if not self.bermuda: self.learning_rate = args.learning_rate - self.loss = Loss(self.args.focal) + loss = LossBraTS if self.args.brats else Loss + self.loss = loss(self.args.focal) self.tta_flips = get_tta_flips(args.dim) - self.dice = Dice(self.n_class) + self.dice = Dice(self.n_class, self.args.brats) if self.args.exec_mode in ["train", "evaluate"]: self.dllogger = get_dllogger(args.results) @@ -72,10 +74,17 @@ class NNUnet(pl.LightningModule): return self.model(img) return self.tta_inference(img) if self.args.tta else self.do_inference(img) + def compute_loss(self, preds, label): + if self.args.deep_supervision: + pred0, pred1, pred2 = preds + loss = self.loss(pred0, label) + 0.5 * self.loss(pred1, label) + 0.25 * self.loss(pred2, label) + return loss / 1.75 + return self.loss(preds, label) + def training_step(self, batch, batch_idx): img, lbl = self.get_train_data(batch) pred = self.model(img) - loss = self.loss(pred, lbl) + loss = self.compute_loss(pred, lbl) return loss def validation_step(self, batch, batch_idx): @@ -84,49 +93,50 @@ class NNUnet(pl.LightningModule): img, lbl = batch["image"], batch["label"] pred = self._forward(img) loss = self.loss(pred, lbl) - self.dice.update(pred, lbl[:, 0]) - return {"val_loss": loss} + self.dice.update(pred, lbl[:, 0], loss) def test_step(self, batch, batch_idx): if self.args.exec_mode == "evaluate": return self.validation_step(batch, batch_idx) img = batch["image"] - pred = self._forward(img) + pred = self._forward(img).squeeze(0).cpu().detach().numpy() if self.args.save_preds: meta = batch["meta"][0].cpu().detach().numpy() - original_shape = meta[2] min_d, max_d = meta[0, 0], meta[1, 0] min_h, max_h = meta[0, 1], meta[1, 1] min_w, max_w = meta[0, 2], meta[1, 2] - - final_pred = torch.zeros((1, pred.shape[1], *original_shape), device=img.device) - final_pred[:, :, min_d:max_d, min_h:max_h, min_w:max_w] = pred - final_pred = nn.functional.softmax(final_pred, dim=1) - final_pred = final_pred.squeeze(0).cpu().detach().numpy() - - if not all(original_shape == final_pred.shape[1:]): - class_ = final_pred.shape[0] - resized_pred = np.zeros((class_, *original_shape)) - for i in range(class_): + n_class, original_shape, cropped_shape = pred.shape[0], meta[2], meta[3] + if not all(cropped_shape == pred.shape[1:]): + resized_pred = np.zeros((n_class, *cropped_shape)) + for i in range(n_class): resized_pred[i] = resize( - final_pred[i], original_shape, order=3, mode="edge", cval=0, clip=True, anti_aliasing=False + pred[i], cropped_shape, order=3, mode="edge", cval=0, clip=True, anti_aliasing=False ) - final_pred = resized_pred + pred = resized_pred + final_pred = np.zeros((n_class, *original_shape)) + final_pred[:, min_d:max_d, min_h:max_h, min_w:max_w] = pred + if self.args.brats: + final_pred = expit(final_pred) + else: + final_pred = softmax(final_pred, axis=0) self.save_mask(final_pred) def build_nnunet(self): in_channels, n_class, kernels, strides, self.patch_size = get_unet_params(self.args) self.n_class = n_class - 1 + if self.args.brats: + n_class = 3 self.model = UNet( in_channels=in_channels, n_class=n_class, kernels=kernels, strides=strides, dimension=self.args.dim, - residual=self.args.residual, normalization_layer=self.args.norm, negative_slope=self.args.negative_slope, + deep_supervision=self.args.deep_supervision, + more_chn=self.args.more_chn, ) if is_main_process(): print(f"Filters: {self.model.filters},\nKernels: {kernels}\nStrides: {strides}") @@ -180,39 +190,30 @@ class NNUnet(pl.LightningModule): mode=self.args.blend, ) - @staticmethod - def metric_mean(name, outputs): - return torch.stack([out[name] for out in outputs]).mean(dim=0) - def validation_epoch_end(self, outputs): - if self.current_epoch < self.args.skip_first_n_eval: - self.log("dice_sum", 0.001 * self.current_epoch) - self.dice.reset() - return None - loss = self.metric_mean("val_loss", outputs) - dice = self.dice.compute() - dice_sum = torch.sum(dice) - if dice_sum >= self.best_sum: - self.best_sum = dice_sum - self.best_sum_dice = dice[:] - self.best_sum_epoch = self.current_epoch + dice, loss = self.dice.compute() + self.dice.reset() + dice_mean = torch.mean(dice) + if dice_mean >= self.best_mean: + self.best_mean = dice_mean + self.best_mean_dice = dice[:] + self.best_mean_epoch = self.current_epoch for i, dice_i in enumerate(dice): if dice_i > self.best_dice[i]: self.best_dice[i], self.best_epoch[i] = dice_i, self.current_epoch if is_main_process(): metrics = {} - metrics.update({"mean dice": round(torch.mean(dice).item(), 2)}) - metrics.update({"TOP_mean": round(torch.mean(self.best_sum_dice).item(), 2)}) + metrics.update({"Mean dice": round(torch.mean(dice).item(), 2)}) + metrics.update({"Highest": round(torch.mean(self.best_mean_dice).item(), 2)}) if self.n_class > 1: metrics.update({f"L{i+1}": round(m.item(), 2) for i, m in enumerate(dice)}) - metrics.update({f"TOP_L{i+1}": round(m.item(), 2) for i, m in enumerate(self.best_sum_dice)}) metrics.update({"val_loss": round(loss.item(), 4)}) self.dllogger.log(step=self.current_epoch, data=metrics) self.dllogger.flush() self.log("val_loss", loss) - self.log("dice_sum", dice_sum) + self.log("dice_mean", dice_mean) def test_epoch_end(self, outputs): if self.args.exec_mode == "evaluate": @@ -222,22 +223,20 @@ class NNUnet(pl.LightningModule): optimizer = { "sgd": FusedSGD(self.parameters(), lr=self.learning_rate, momentum=self.args.momentum), "adam": FusedAdam(self.parameters(), lr=self.learning_rate, weight_decay=self.args.weight_decay), - "radam": RAdam(self.parameters(), lr=self.learning_rate, weight_decay=self.args.weight_decay), }[self.args.optimizer.lower()] - scheduler = { - "none": None, - "multistep": torch.optim.lr_scheduler.MultiStepLR(optimizer, self.args.steps, gamma=self.args.factor), - "cosine": torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, self.args.max_epochs), - "plateau": torch.optim.lr_scheduler.ReduceLROnPlateau( - optimizer, factor=self.args.factor, patience=self.args.lr_patience - ), - }[self.args.scheduler.lower()] - - opt_dict = {"optimizer": optimizer, "monitor": "val_loss"} - if scheduler is not None: - opt_dict.update({"lr_scheduler": scheduler}) - return opt_dict + if self.args.scheduler: + scheduler = { + "scheduler": WarmupCosineSchedule( + optimizer=optimizer, + warmup_steps=250, + t_total=self.args.epochs * len(self.trainer.datamodule.train_dataloader()), + ), + "interval": "step", + "frequency": 1, + } + return {"optimizer": optimizer, "monitor": "val_loss", "lr_scheduler": scheduler} + return {"optimizer": optimizer, "monitor": "val_loss"} def save_mask(self, pred): if self.test_idx == 0: diff --git a/PyTorch/Segmentation/nnUNet/models/unet.py b/PyTorch/Segmentation/nnUNet/models/unet.py index 25cf8f12..282e94aa 100644 --- a/PyTorch/Segmentation/nnUNet/models/unet.py +++ b/PyTorch/Segmentation/nnUNet/models/unet.py @@ -11,9 +11,11 @@ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. + +import torch import torch.nn as nn -from models.layers import ConvBlock, OutputBlock, ResidBlock, UpsampleBlock +from models.layers import ConvBlock, OutputBlock, UpsampleBlock class UNet(nn.Module): @@ -25,34 +27,39 @@ class UNet(nn.Module): strides, normalization_layer, negative_slope, - residual, dimension, + deep_supervision, + more_chn, ): super(UNet, self).__init__() + self.more_chn = more_chn self.dim = dimension self.n_class = n_class - self.residual = residual self.negative_slope = negative_slope self.norm = normalization_layer + f"norm{dimension}d" - self.filters = [min(2 ** (5 + i), 320 if dimension == 3 else 512) for i in range(len(strides))] + self.deep_supervision = deep_supervision + self.depth = len(strides) + if self.more_chn: + self.filters = [64, 96, 128, 192, 256, 384, 512, 768, 1024][: self.depth] + else: + self.filters = [min(2 ** (5 + i), 320 if dimension == 3 else 512) for i in range(self.depth)] - down_block = ResidBlock if self.residual else ConvBlock self.input_block = self.get_conv_block( - conv_block=down_block, + conv_block=ConvBlock, in_channels=in_channels, out_channels=self.filters[0], kernel_size=kernels[0], stride=strides[0], ) self.downsamples = self.get_module_list( - conv_block=down_block, + conv_block=ConvBlock, in_channels=self.filters[:-1], out_channels=self.filters[1:], kernels=kernels[1:-1], strides=strides[1:-1], ) self.bottleneck = self.get_conv_block( - conv_block=down_block, + conv_block=ConvBlock, in_channels=self.filters[-2], out_channels=self.filters[-1], kernel_size=kernels[-1], @@ -65,6 +72,8 @@ class UNet(nn.Module): kernels=kernels[1:][::-1], strides=strides[1:][::-1], ) + self.deep_supervision_head1 = self.get_output_block(1) + self.deep_supervision_head2 = self.get_output_block(2) self.output_block = self.get_output_block(decoder_level=0) self.apply(self.initialize_weights) self.n_layers = len(self.upsamples) - 1 @@ -76,9 +85,17 @@ class UNet(nn.Module): out = downsample(out) encoder_outputs.append(out) out = self.bottleneck(out) - for idx, upsample in enumerate(self.upsamples): - out = upsample(out, encoder_outputs[self.n_layers - idx]) + decoder_outputs = [] + for i, upsample in enumerate(self.upsamples): + out = upsample(out, encoder_outputs[self.depth - i - 2]) + decoder_outputs.append(out) out = self.output_block(out) + if self.training and self.deep_supervision: + out1 = self.deep_supervision_head1(decoder_outputs[-2]) + out2 = self.deep_supervision_head2(decoder_outputs[-3]) + out1 = nn.functional.interpolate(out1, out.shape[2:], mode="trilinear", align_corners=True) + out2 = nn.functional.interpolate(out2, out.shape[2:], mode="trilinear", align_corners=True) + return torch.stack([out, out1, out2]) return out def get_conv_block(self, conv_block, in_channels, out_channels, kernel_size, stride): diff --git a/PyTorch/Segmentation/nnUNet/notebooks/BraTS21.ipynb b/PyTorch/Segmentation/nnUNet/notebooks/BraTS21.ipynb new file mode 100644 index 00000000..d1ea956f --- /dev/null +++ b/PyTorch/Segmentation/nnUNet/notebooks/BraTS21.ipynb @@ -0,0 +1,746 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# nnU-Net for BraTS21\n", + "\n", + "# Table of contents\n", + "- [Introduction](#introduction)\n", + "- [Dataset](#dataset)\n", + "- [Data pre-processing](#preprocessing)\n", + "- [Data augmentations](#augmentations)\n", + "- [Loss function](#loss)\n", + "- [Model](#model)\n", + "- [Training](#training)\n", + "- [Inference](#inference)\n", + "- [Post-processing](#postprocessing)\n", + "\n", + "# Introduction \n", + "\n", + "The goal of [BraTS 2021 challenge](https://www.med.upenn.edu/cbica/brats2021) was to create a model for segmenting the brain glioblastoma subregions in mpMRI scans. By using our nnU-Net implementation, NVIDIA data scientists have [won the BraTS21 validation phase](https://developer.nvidia.com/blog/nvidia-data-scientists-take-top-spots-in-miccai-2021-brain-tumor-segmentation-challenge). In this notebook, we will share with you the recipe we used for training our nnU-Net for BraTS21 challenge, so that you can reproduce our results. In particular, we will walk you through the following steps: data pre-processing, designing the loss function, building and training the model, running inference and finally post-processing the predictions.\n", + "\n", + "# Dataset \n", + "\n", + "The training dataset provided for the BraTS21 challenge consists of 1,251 brain mpMRI scans along with segmentation annotations of tumorous regions. The 3D volumes were skull-stripped and resampled to 1 mm isotropic resolution, with dimensions of (240, 240, 155) voxels. For each example, four modalities were given: Fluid Attenuated Inversion Recovery (FLAIR), native (T1), post-contrast T1-weighted (T1Gd), and T2-weighted (T2). See image below with each modality. Annotations consist of four classes: 1 for necrotic tumor core (NCR), 2 for peritumoral edematous tissue (ED), 4 for enhancing tumor (ET), and 0 for background (voxels that are not part of the tumor).\n", + "\n", + "To download the training and validation dataset, you need to have an account on https://www.synapse.org platform and be registered for BraTS21 challenge. We will assume that after downloading and unzipping, the dataset is organized as follows:\n", + "\n", + "```\n", + "/data \n", + " │\n", + " ├───BraTS2021_train\n", + " │ ├──BraTS2021_00000 \n", + " │ │ └──BraTS2021_00000_flair.nii.gz\n", + " │ │ └──BraTS2021_00000_t1.nii.gz\n", + " │ │ └──BraTS2021_00000_t1ce.nii.gz\n", + " │ │ └──BraTS2021_00000_t2.nii.gz\n", + " │ │ └──BraTS2021_00000_seg.nii.gz\n", + " │ ├──BraTS2021_00002\n", + " │ │ └──BraTS2021_00002_flair.nii.gz\n", + " │ ... └──...\n", + " │\n", + " └────BraTS2021_val\n", + " ├──BraTS2021_00001 \n", + " │ └──BraTS2021_00001_flair.nii.gz\n", + " │ └──BraTS2021_00001_t1.nii.gz\n", + " │ └──BraTS2021_00001_t1ce.nii.gz\n", + " │ └──BraTS2021_00001_t2.nii.gz\n", + " ├──BraTS2021_00002\n", + " │ └──BraTS2021_00002_flair.nii.gz\n", + " ... └──...\n", + "```\n", + "\n", + "Let's visualize a BraTS2021_00000 example from the training dataset. Each plot presents a different modality (from left to right: FLAIR, T1, T1ce, T2), and an annotation." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import nibabel as nib\n", + "from glob import glob\n", + "\n", + "imgs = [nib.load(f\"/data/BraTS2021_train/BraTS2021_00000/BraTS2021_00000_{m}.nii.gz\").get_fdata().astype(np.float32)[:, :, 75] for m in [\"flair\", \"t1\", \"t1ce\", \"t2\"]]\n", + "lbl = nib.load(\"/data/BraTS2021_train/BraTS2021_00000/BraTS2021_00000_seg.nii.gz\").get_fdata().astype(np.uint8)[:, :, 75]\n", + "fig, ax = plt.subplots(nrows=1, ncols=5, figsize=(15, 15))\n", + "for i, img in enumerate(imgs):\n", + " ax[i].imshow(img, cmap='gray')\n", + " ax[i].axis('off')\n", + "ax[-1].imshow(lbl, vmin=0, vmax=4)\n", + "ax[-1].axis('off')\n", + "plt.tight_layout() \n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Data pre-processing \n", + "\n", + "Each example of the BraTS21 dataset consists of four [NIfTI](https://nifti.nimh.nih.gov/) files with different MRI modalities (filenames with suffixes flair, t1, t1ce, t2). Additionally, examples in the training dataset have a NIfTI with annotation (filename with suffix seg). As a first step of data pre-processing, all four modalities are stacked such that each example has shape (4, 240, 240, 155) (input tensor is in the (C, H, W, D) layout, where C-channels, H-height, W-width and D-depth). Then redundant background voxels (with voxel value zero) on the borders of each volume are [cropped](https://docs.monai.io/en/latest/transforms.html#cropforeground), as they do not provide any useful information and can be ignored by the neural network. Subsequently, for each example, the mean and the standard deviation are computed within the non-zero region for each channel separately. All volumes are [normalized](https://docs.monai.io/en/latest/transforms.html#normalizeintensityd) by first subtracting the mean and then divided by the standard deviation. The background voxels are not normalized so that their value remained at zero. To distinguish between background voxels and normalized voxels which have values close to zero, we add an input channel with one-hot encoding for foreground voxels and stacked with the input data. As a result, each example has 5 channels.\n", + "\n", + "Let's start by preparing the raw training and validation datasets into stacked NIfTI files." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Preparing BraTS21 dataset from: /data/BraTS2021_train\n", + "Preparing time: 259.65\n", + "Preparing BraTS21 dataset from: /data/BraTS2021_val\n", + "Preparing time: 38.62\n", + "Finished!\n" + ] + } + ], + "source": [ + "import json\n", + "import os\n", + "from glob import glob\n", + "from subprocess import call\n", + "import time\n", + "\n", + "import nibabel\n", + "import numpy as np\n", + "from joblib import Parallel, delayed\n", + "\n", + "\n", + "def load_nifty(directory, example_id, suffix):\n", + " return nibabel.load(os.path.join(directory, example_id + \"_\" + suffix + \".nii.gz\"))\n", + "\n", + "\n", + "def load_channels(d, example_id):\n", + " return [load_nifty(d, example_id, suffix) for suffix in [\"flair\", \"t1\", \"t1ce\", \"t2\"]]\n", + "\n", + "\n", + "def get_data(nifty, dtype=\"int16\"):\n", + " if dtype == \"int16\":\n", + " data = np.abs(nifty.get_fdata().astype(np.int16))\n", + " data[data == -32768] = 0\n", + " return data\n", + " return nifty.get_fdata().astype(np.uint8)\n", + "\n", + "\n", + "def prepare_nifty(d):\n", + " example_id = d.split(\"/\")[-1]\n", + " flair, t1, t1ce, t2 = load_channels(d, example_id)\n", + " affine, header = flair.affine, flair.header\n", + " vol = np.stack([get_data(flair), get_data(t1), get_data(t1ce), get_data(t2)], axis=-1)\n", + " vol = nibabel.nifti1.Nifti1Image(vol, affine, header=header)\n", + " nibabel.save(vol, os.path.join(d, example_id + \".nii.gz\"))\n", + "\n", + " if os.path.exists(os.path.join(d, example_id + \"_seg.nii.gz\")):\n", + " seg = load_nifty(d, example_id, \"seg\")\n", + " affine, header = seg.affine, seg.header\n", + " vol = get_data(seg, \"unit8\")\n", + " vol[vol == 4] = 3\n", + " seg = nibabel.nifti1.Nifti1Image(vol, affine, header=header)\n", + " nibabel.save(seg, os.path.join(d, example_id + \"_seg.nii.gz\"))\n", + "\n", + "\n", + "def prepare_dirs(data, train):\n", + " img_path, lbl_path = os.path.join(data, \"images\"), os.path.join(data, \"labels\")\n", + " call(f\"mkdir {img_path}\", shell=True)\n", + " if train:\n", + " call(f\"mkdir {lbl_path}\", shell=True)\n", + " dirs = glob(os.path.join(data, \"BraTS*\"))\n", + " for d in dirs:\n", + " if \"_\" in d.split(\"/\")[-1]:\n", + " files = glob(os.path.join(d, \"*.nii.gz\"))\n", + " for f in files:\n", + " if \"flair\" in f or \"t1\" in f or \"t1ce\" in f or \"t2\" in f:\n", + " continue\n", + " if \"_seg\" in f:\n", + " call(f\"mv {f} {lbl_path}\", shell=True)\n", + " else:\n", + " call(f\"mv {f} {img_path}\", shell=True)\n", + " call(f\"rm -rf {d}\", shell=True)\n", + "\n", + "\n", + "def prepare_dataset_json(data, train):\n", + " images, labels = glob(os.path.join(data, \"images\", \"*\")), glob(os.path.join(data, \"labels\", \"*\"))\n", + " images = sorted([img.replace(data + \"/\", \"\") for img in images])\n", + " labels = sorted([lbl.replace(data + \"/\", \"\") for lbl in labels])\n", + "\n", + " modality = {\"0\": \"FLAIR\", \"1\": \"T1\", \"2\": \"T1CE\", \"3\": \"T2\"}\n", + " labels_dict = {\"0\": \"background\", \"1\": \"edema\", \"2\": \"non-enhancing tumor\", \"3\": \"enhancing tumour\"}\n", + " if train:\n", + " key = \"training\"\n", + " data_pairs = [{\"image\": img, \"label\": lbl} for (img, lbl) in zip(images, labels)]\n", + " else:\n", + " key = \"test\"\n", + " data_pairs = [{\"image\": img} for img in images]\n", + "\n", + " dataset = {\n", + " \"labels\": labels_dict,\n", + " \"modality\": modality,\n", + " key: data_pairs,\n", + " }\n", + "\n", + " with open(os.path.join(data, \"dataset.json\"), \"w\") as outfile:\n", + " json.dump(dataset, outfile)\n", + "\n", + "\n", + "def run_parallel(func, args):\n", + " return Parallel(n_jobs=os.cpu_count())(delayed(func)(arg) for arg in args)\n", + "\n", + "\n", + "def prepare_dataset(data, train):\n", + " print(f\"Preparing BraTS21 dataset from: {data}\")\n", + " start = time.time()\n", + " run_parallel(prepare_nifty, sorted(glob(os.path.join(data, \"BraTS*\"))))\n", + " prepare_dirs(data, train)\n", + " prepare_dataset_json(data, train)\n", + " end = time.time()\n", + " print(f\"Preparing time: {(end - start):.2f}\")\n", + "\n", + "prepare_dataset(\"/data/BraTS2021_train\", True)\n", + "prepare_dataset(\"/data/BraTS2021_val\", False)\n", + "print(\"Finished!\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now, lets preprocesses the datasets by cropping and normalizing the volumes. We will store the pre-processed volumes as NumPy arrays." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Preprocessing /data/BraTS2021_train\n", + "Pre-processing time: 408.09\n", + "Preprocessing /data/BraTS2021_val\n", + "Pre-processing time: 68.28\n", + "Finished!\n" + ] + } + ], + "source": [ + "!python3 ../preprocess.py --task 11 --ohe --exec_mode training\n", + "!python3 ../preprocess.py --task 12 --ohe --exec_mode test\n", + "print(\"Finished!\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Data Augmentations \n", + "\n", + "Data augmentation is a technique that alleviates the overfitting problem by artificially extending a dataset during the training phase. To make our method more robust, the following data augmentations are used during training phase:\n", + "\n", + "1. **Biased crop**: From the input volume, a patch of dimensions (5, 128, 128, 128) was randomly cropped. Additionally, with probability of 0.4 the patch selected via random biased crop is guaranteed that some foreground voxels (with positive class in the ground truth) are present in the cropped region.\n", + "2. **Zoom**: With probability of 0.15, a zoom factor is sampled uniformly from (1.0, 1.4) and then input volume is zoomed by a sampled factor with cubic interpolation, while label with nearest neighbor interpolation.\n", + "3. **Flips**: With probability of 0.5, for each x, y, z axis independently, volume was flipped along that axis.\n", + "4. **Gaussian Noise**: With probability of 0.15, random Gaussian noise with mean zero and standard deviation sampled uniformly from (0, 0.33) is sampled for each voxel and added to the input volume. \n", + "5. **Gaussian Blur**: With probability of 0.15, Gaussian blurring with standard deviation of the Gaussian Kernel sampled uniformly from (0.5, 1.5) is applied to the input volume.\n", + "6. **Brightness**: With probability of 0.15, a random value is sampled uniformly from (0.7, 1.3) and then input volume voxels are multiplied by it.\n", + "7. **Contrast**: With probability of 0.15, a random value is sampled uniformly from (0.65, 1.5) and then input volume voxels are multiplied by it and clipped to its original min and max values.\n", + "\n", + "The data loading pipeline is implemented with [NVIDIA Data Loading Library (DALI)](https://docs.nvidia.com/deeplearning/dali/user-guide/docs/index.html), which addresses the problem of the CPU bottleneck by offloading data augmentations to the GPU. We encourage you to check out the implementation details of our [DALI pipeline](https://github.com/NVIDIA/DeepLearningExamples/blob/master/PyTorch/Segmentation/nnUNet/data_loading/dali_loader.py)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Loss function \n", + "\n", + "The BraTS leaderboard is computed based on three partially overlapping regions: whole tumor (1, 2, 4), tumor core (1, 4) and enhancing tumor (4), instead of classes present in the labels. Thus, it is beneficial to construct the loss function based on classes used for ranking calculation. Therefore, we optimize each region separately with a sum of binary Cross-Entropy with the Dice loss." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [], + "source": [ + "import torch.nn as nn\n", + "from monai.losses import DiceLoss\n", + "\n", + "class Loss(nn.Module):\n", + " def __init__(self):\n", + " super(Loss, self).__init__()\n", + " self.dice = DiceLoss(sigmoid=True, batch=True)\n", + " self.ce = nn.BCEWithLogitsLoss()\n", + "\n", + " def _loss(self, p, y):\n", + " return self.dice(p, y) + self.ce(p, y.float())\n", + "\n", + " def forward(self, p, y):\n", + " y_wt, y_tc, y_et = y > 0, ((y == 1) + (y == 3)) > 0, y == 3\n", + " p_wt, p_tc, p_et = p[:, 0].unsqueeze(1), p[:, 1].unsqueeze(1), p[:, 2].unsqueeze(1)\n", + " l_wt, l_tc, l_et = self._loss(p_wt, y_wt), self._loss(p_tc, y_tc), self._loss(p_et, y_et)\n", + " return l_wt + l_tc + l_et" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Model \n", + "\n", + "We have made some modifications to the U-Net architecture for the BraTS challenge with respect to the original nnU-Net template. In particular, the U-Net template in the nnU-Net has the encoder depth of 6, and the convolution channels at each encoder level are: 32, 64, 128, 256, 320, 320. Based on the experiments we run, increasing the depth of the encoder to 7, modifying the number of channels to: 64, 96, 128, 192, 256, 384, 512, and using deep supervision improves the final score.\n", + "\n", + "For deep supervision, we used two additional output heads at the decoder levels with feature map sizes (64, 64, 64) and (32, 32, 32). To match the shape of the additional predictions with the label shape of (128, 128, 128) we downsampled the label using the nearest neighbor interpolation to the (64, 64, 64) and (32, 32, 32) shapes, so that loss can be computed for additional outputs." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "data": { + "image/jpeg": "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\n", + "text/plain": [ + "" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from IPython.display import Image\n", + "Image(filename=\"../images/unet-brats.jpg\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Figure 1: *The final U-Net architecture used for BraTS21 challenge.*\n", + "\n", + "# Training \n", + "\n", + "Now, let's start training the model. For that, we will call the training script from our nnUNet repo with some additional command line arguments for BraTS challenge: \n", + "\n", + "- `--brats` - use loss function with partially overlapping regions (WT, TC, ET) and BraTS specific inference;\n", + "- `--deep_supervision` - use deep supervision loss with two additional output heads;\n", + "- `--more_chn` - create encoder with more channels than regular U-Net;\n", + "- `--min_fmap 2` - create deeper encoder, with feature map size in the bottleneck 2x2x2;\n", + "\n", + "and the regular command line arguments:\n", + "\n", + "- `--scheduler` - use cosine decay learning rate scheduler with warm up 250 steps of warm up;\n", + "- `--learning_rate 0.0003` - initial learning rate after warm up will be set to 0.0003;\n", + "- `--epochs 30` - training will be done for 30 epochs;\n", + "- `--fold 0` - training will be done for fold 0 (by default, 5-fold cross validation is used);\n", + "- `--amp` - training with automatic mixed precision, for faster training and memory reduction;\n", + "- `--gpus 1` - one GPU will be used during training;\n", + "- `--task 11` - task number for BraTS21 training dataset. See file `data_preprocessing/configs.py` for more details;\n", + "- `--save_ckpt` - save checkpoint with highest dice score acheived during training.\n", + "\n", + "We will run training on 1xA100 GPU. To train the model with [AMP](https://developer.nvidia.com/automatic-mixed-precision), you will need a GPU with at least 15G memory.\n", + "\n", + "Here, we will train the model on just 1-fold (fold with index 0) and 30 epochs. For the challenge submission, we have trained 5 models on each fold for 150 epochs, and averaged their predictions." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Global seed set to 1\n", + "Number of examples: Train 1000 - Val 251\n", + "Filters: [64, 96, 128, 192, 256, 384, 512],\n", + "Kernels: [[3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]]\n", + "Strides: [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2]]\n", + "GPU available: True, used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "Using native 16bit precision.\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\n", + " | Name | Type | Params\n", + "------------------------------------\n", + "0 | model | UNet | 50.8 M\n", + "1 | loss | LossBraTS | 0 \n", + "2 | dice | Dice | 0 \n", + "------------------------------------\n", + "50.8 M Trainable params\n", + "0 Non-trainable params\n", + "50.8 M Total params\n", + "203.356 Total estimated model params size (MB)\n", + "Epoch 0: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 09:47:10.369319 - Epoch: 0 Mean dice : 76.92 Highest : 76.92 L1 : 72.81 L2 : 72.71 L3 : 85.23 val_loss : 0.9542\n", + "Epoch 1: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 09:50:06.658588 - Epoch: 1 Mean dice : 78.11 Highest : 78.11 L1 : 74.22 L2 : 76.79 L3 : 83.33 val_loss : 0.7793\n", + "Epoch 2: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 09:53:03.036486 - Epoch: 2 Mean dice : 82.41 Highest : 82.41 L1 : 81.34 L2 : 77.48 L3 : 88.41 val_loss : 0.6169\n", + "Epoch 3: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 09:55:59.582026 - Epoch: 3 Mean dice : 83.33 Highest : 83.33 L1 : 83.86 L2 : 78.26 L3 : 87.87 val_loss : 0.5845\n", + "Epoch 4: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 09:58:55.668410 - Epoch: 4 Mean dice : 85.02 Highest : 85.02 L1 : 84.49 L2 : 80.92 L3 : 89.63 val_loss : 0.5207\n", + "Epoch 5: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:01:51.775193 - Epoch: 5 Mean dice : 86.09 Highest : 86.09 L1 : 86.6 L2 : 81.56 L3 : 90.11 val_loss : 0.4856\n", + "Epoch 6: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:04:47.909984 - Epoch: 6 Mean dice : 86.75 Highest : 86.75 L1 : 87.31 L2 : 81.84 L3 : 91.1 val_loss : 0.4545\n", + "Epoch 7: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:07:44.297625 - Epoch: 7 Mean dice : 86.24 Highest : 86.75 L1 : 87.2 L2 : 80.88 L3 : 90.63 val_loss : 0.4771\n", + "Epoch 8: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:10:39.488046 - Epoch: 8 Mean dice : 87.65 Highest : 87.65 L1 : 88.53 L2 : 83.03 L3 : 91.38 val_loss : 0.4257\n", + "Epoch 9: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:13:35.567349 - Epoch: 9 Mean dice : 81.52 Highest : 87.65 L1 : 78.71 L2 : 78.49 L3 : 87.37 val_loss : 0.6421\n", + "Epoch 10: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:16:30.845535 - Epoch: 10 Mean dice : 85.77 Highest : 87.65 L1 : 84.83 L2 : 82.32 L3 : 90.16 val_loss : 0.4978\n", + "Epoch 11: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:19:26.081598 - Epoch: 11 Mean dice : 86.78 Highest : 87.65 L1 : 87.52 L2 : 82.28 L3 : 90.53 val_loss : 0.4516\n", + "Epoch 12: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:22:21.393119 - Epoch: 12 Mean dice : 87.57 Highest : 87.65 L1 : 87.97 L2 : 83.67 L3 : 91.08 val_loss : 0.4225\n", + "Epoch 13: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:25:16.783225 - Epoch: 13 Mean dice : 86.75 Highest : 87.65 L1 : 88.16 L2 : 82.88 L3 : 89.2 val_loss : 0.4548\n", + "Epoch 14: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:28:12.184295 - Epoch: 14 Mean dice : 86.88 Highest : 87.65 L1 : 86.13 L2 : 82.73 L3 : 91.77 val_loss : 0.4468\n", + "Epoch 15: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:31:07.374641 - Epoch: 15 Mean dice : 86.45 Highest : 87.65 L1 : 86.58 L2 : 83.46 L3 : 89.31 val_loss : 0.4728\n", + "Epoch 16: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:34:03.013703 - Epoch: 16 Mean dice : 87.56 Highest : 87.65 L1 : 88.26 L2 : 82.44 L3 : 91.99 val_loss : 0.4162\n", + "Epoch 17: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:36:58.202911 - Epoch: 17 Mean dice : 86.81 Highest : 87.65 L1 : 86.95 L2 : 83.3 L3 : 90.19 val_loss : 0.4453\n", + "Epoch 18: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:39:53.265586 - Epoch: 18 Mean dice : 88.21 Highest : 88.21 L1 : 88.7 L2 : 84.11 L3 : 91.81 val_loss : 0.4033\n", + "Epoch 19: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:42:49.002307 - Epoch: 19 Mean dice : 87.69 Highest : 88.21 L1 : 87.31 L2 : 83.86 L3 : 91.9 val_loss : 0.4166\n", + "Epoch 20: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:45:44.182336 - Epoch: 20 Mean dice : 88.9 Highest : 88.9 L1 : 89.62 L2 : 84.57 L3 : 92.52 val_loss : 0.3775\n", + "Epoch 21: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:48:40.095388 - Epoch: 21 Mean dice : 88.38 Highest : 88.9 L1 : 89.7 L2 : 83.1 L3 : 92.35 val_loss : 0.3971\n", + "Epoch 22: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:51:35.371210 - Epoch: 22 Mean dice : 88.05 Highest : 88.9 L1 : 88.21 L2 : 84.05 L3 : 91.88 val_loss : 0.4091\n", + "Epoch 23: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:54:30.622294 - Epoch: 23 Mean dice : 88.54 Highest : 88.9 L1 : 89.35 L2 : 83.88 L3 : 92.4 val_loss : 0.3882\n", + "Epoch 24: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 10:57:25.706283 - Epoch: 24 Mean dice : 88.86 Highest : 88.9 L1 : 90.19 L2 : 84.43 L3 : 91.97 val_loss : 0.3735\n", + "Epoch 25: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 11:00:20.871930 - Epoch: 25 Mean dice : 88.72 Highest : 88.9 L1 : 89.98 L2 : 84.97 L3 : 91.22 val_loss : 0.3824\n", + "Epoch 26: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 11:03:15.888703 - Epoch: 26 Mean dice : 89.26 Highest : 89.26 L1 : 90.43 L2 : 84.83 L3 : 92.53 val_loss : 0.3639\n", + "Epoch 27: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 11:06:12.074362 - Epoch: 27 Mean dice : 88.67 Highest : 89.26 L1 : 89.51 L2 : 84.29 L3 : 92.21 val_loss : 0.3861\n", + "Epoch 28: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 11:09:07.148004 - Epoch: 28 Mean dice : 89.4 Highest : 89.4 L1 : 90.79 L2 : 85.35 L3 : 92.06 val_loss : 0.36\n", + "Epoch 29: 100%|█████████████████████████████████████████████████████████████████████████████████████████▉| 751/751 [02:53<00:00]\n", + "DLL 2021-10-07 11:12:03.077000 - Epoch: 29 Mean dice : 89.69 Highest : 89.69 L1 : 90.69 L2 : 85.5 L3 : 92.88 val_loss : 0.3489" + ] + } + ], + "source": [ + "!python ../main.py --brats --deep_supervision --more_chn --min_fmap 2 --scheduler --learning_rate 0.0003 --epochs 30 --fold 0 --amp --gpus 1 --task 11 --save_ckpt" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Inference \n", + "\n", + "During inference, the input volume can have arbitrary size, instead of the fixed patch size (128, 128, 128) as during the training phase. Thus, we use a [sliding window inference](https://docs.monai.io/en/latest/inferers.html) from [MONAI](https://monai.io/) library, where the window has the same size as the training patch, i.e., (128, 128, 128) and adjacent windows overlap by half the size of a patch. The predictions on the overlapping regions are then averaged with Gaussian importance weighting, such that the weights of the center voxels have higher importance.\n", + "\n", + "One of the known tricks to improve predictions robustness is to apply test time augmentations (TTA). During inference, we are creating eight versions of the input volume, such that each version corresponds to one of eight possible flips along the x, y, z axis combination. Then we run inference for each version of the input volume and transform the predictions back to the original input volume orientation by applying the same flips to predictions as was used for the input volume. Finally, the probabilities from all predictions were averaged.\n", + "\n", + "Let's run inference with TTA on the challenge validation dataset.\n", + "\n", + "Note: You will have to modify the `--ckpt_path` argument, such that the path to checkpoint is valid." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Global seed set to 1\n", + "Number of test examples: 219\n", + "Filters: [64, 96, 128, 192, 256, 384, 512],\n", + "Kernels: [[3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]]\n", + "Strides: [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2]]\n", + "GPU available: True, used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "Using native 16bit precision.\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "Testing: 100%|███████████████████████████████████████████████████████████████████████████████████| 219/219 [10:39<00:00]\n" + ] + } + ], + "source": [ + "!python ../main.py --gpus 1 --amp --save_preds --exec_mode predict --brats --data /data/12_3d/test --ckpt_path /results/checkpoints/epoch=29-dice_mean=89.69.ckpt --tta" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Post-processing \n", + "\n", + "By optimizing the three overlapping regions (ET, TC, WT) we need to convert them back to the original classes (NCR, ED, ET). The strategy for transforming classes back to the original one is the following: if the WT probability for a given voxel is less than 0.45 then its class is set to 0 (background), otherwise if the probability for TC is less than 0.4 the voxel class is 2 (ED), and finally if probability for ET is less than 0.4 voxel has class 1 (NCR), or otherwise 4 (ET).\n", + "\n", + "Furthermore, we applied the following post-processing strategy: find ET connected components, for components smaller than 16 voxels with mean probability smaller than 0.9, replace their class to NCR (such that voxels are still considered part of the tumor core), next if there is overall less than 73 voxels with ET and their mean probability is smaller than 0.9 replace all ET voxels to NCR. With such post-processing we avoided the edge case where the model predicted a few voxels with enhancing tumor (ET) but there were not any in the ground truth. Such post-processing was beneficial to the final score as if there were no enhancing tumor voxels in the label, then the Dice score for zero false positive prediction was 1, and 0 otherwise." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Preparing final predictions\n", + "Finished!\n" + ] + } + ], + "source": [ + "import os\n", + "from glob import glob\n", + "from subprocess import call\n", + "\n", + "import nibabel as nib\n", + "import numpy as np\n", + "from scipy.ndimage.measurements import label\n", + "\n", + "\n", + "def to_lbl(pred):\n", + " enh = pred[2]\n", + " c1, c2, c3 = pred[0] > 0.5, pred[1] > 0.5, pred[2] > 0.5\n", + " pred = (c1 > 0).astype(np.uint8)\n", + " pred[(c2 == False) * (c1 == True)] = 2\n", + " pred[(c3 == True) * (c1 == True)] = 4\n", + "\n", + " components, n = label(pred == 4)\n", + " for et_idx in range(1, n + 1):\n", + " _, counts = np.unique(pred[components == et_idx], return_counts=True)\n", + " if 1 < counts[0] and counts[0] < 8 and np.mean(enh[components == et_idx]) < 0.9:\n", + " pred[components == et_idx] = 1\n", + "\n", + " et = pred == 4\n", + " if 0 < et.sum() and et.sum() < 73 and np.mean(enh[et]) < 0.9:\n", + " pred[et] = 1\n", + "\n", + " pred = np.transpose(pred, (2, 1, 0)).astype(np.uint8)\n", + " return pred\n", + "\n", + "\n", + "def prepare_preditions(e):\n", + " fname = e[0].split(\"/\")[-1].split(\".\")[0]\n", + " preds = [np.load(f) for f in e]\n", + " p = to_lbl(np.mean(preds, 0))\n", + "\n", + " img = nib.load(f\"/data/BraTS2021_val/images/{fname}.nii.gz\")\n", + " nib.save(\n", + " nib.Nifti1Image(p, img.affine, header=img.header),\n", + " os.path.join(\"/results/final_preds\", fname + \".nii.gz\"),\n", + " )\n", + "\n", + "os.makedirs(\"/results/final_preds\")\n", + "preds = sorted(glob(f\"/results/predictions*\"))\n", + "examples = list(zip(*[sorted(glob(f\"{p}/*.npy\")) for p in preds]))\n", + "print(\"Preparing final predictions\")\n", + "for e in examples:\n", + " prepare_preditions(e)\n", + "print(\"Finished!\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Visualization\n", + "\n", + "Let's visualize the final prediction made on the challenge validation dataset." + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "BraTS2021_00001\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "BraTS2021_00013\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "BraTS2021_00015\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "BraTS2021_00027\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "BraTS2021_00037\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import nibabel as nib\n", + "from glob import glob\n", + "\n", + "n, z = 5, 75\n", + "data = sorted(glob(\"/results/final_preds/*.nii.gz\"))\n", + "for i in range(n):\n", + " fname = data[i].split(\"/\")[-1].split(\".\")[0]\n", + " print(fname)\n", + " img = nib.load(f\"/data/BraTS2021_val/images/{fname}.nii.gz\").get_fdata().astype(np.float32)\n", + " pred = nib.load(data[i]).get_fdata().astype(np.uint8)[:, :, z]\n", + " imgs = [img[:, :, z, i] for i in [0, 3]] + [pred]\n", + " \n", + " fig, ax = plt.subplots(nrows=1, ncols=3, figsize=(12, 12))\n", + " for i in range(3):\n", + " if i < 2:\n", + " ax[i].imshow(imgs[i], cmap='gray')\n", + " else:\n", + " ax[i].imshow(imgs[i]);\n", + " ax[i].axis('off') \n", + " plt.tight_layout() \n", + " plt.show()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.6.9" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/PyTorch/Segmentation/nnUNet/preprocess.py b/PyTorch/Segmentation/nnUNet/preprocess.py index ee0c57bf..7f0d7743 100755 --- a/PyTorch/Segmentation/nnUNet/preprocess.py +++ b/PyTorch/Segmentation/nnUNet/preprocess.py @@ -29,7 +29,8 @@ parser.add_argument( choices=["training", "val", "test"], help="Mode for data preprocessing", ) -parser.add_argument("--dilation", action="store_true", help="Perform morphological label dilation") +parser.add_argument("--ohe", action="store_true", help="Add one-hot-encoding for foreground voxels (voxels > 0)") +parser.add_argument("--verbose", action="store_true") parser.add_argument("--task", type=str, help="Number of task to be run. MSD uses numbers 01-10") parser.add_argument("--dim", type=int, default=3, choices=[2, 3], help="Data dimension to prepare") parser.add_argument("--n_jobs", type=int, default=-1, help="Number of parallel jobs for data preprocessing") @@ -44,4 +45,4 @@ if __name__ == "__main__": if args.exec_mode == "test": path = os.path.join(path, "test") end = time.time() - print(f"Preprocessing time: {(end - start):.2f}") + print(f"Pre-processing time: {(end - start):.2f}") diff --git a/PyTorch/Segmentation/nnUNet/requirements.txt b/PyTorch/Segmentation/nnUNet/requirements.txt index b13baea4..f3f0d709 100755 --- a/PyTorch/Segmentation/nnUNet/requirements.txt +++ b/PyTorch/Segmentation/nnUNet/requirements.txt @@ -1,8 +1,7 @@ git+https://github.com/NVIDIA/dllogger -nibabel==3.1.1 -joblib==0.16.0 -scikit-learn==0.23.2 -pynvml==8.0.4 -pillow==6.2.0 -fsspec==0.8.0 -pytorch_ranger==0.1.1 \ No newline at end of file +nibabel==3.2.1 +joblib==1.0.1 +pytorch-lightning==1.3.8 +scikit-learn==1.0 +scikit-image==0.18.3 +pynvml==11.0.0 diff --git a/PyTorch/Segmentation/nnUNet/scripts/benchmark.py b/PyTorch/Segmentation/nnUNet/scripts/benchmark.py index 5c3d550a..47304da8 100644 --- a/PyTorch/Segmentation/nnUNet/scripts/benchmark.py +++ b/PyTorch/Segmentation/nnUNet/scripts/benchmark.py @@ -35,7 +35,7 @@ if __name__ == "__main__": args = parser.parse_args() path_to_main = os.path.join(dirname(dirname(os.path.realpath(__file__))), "main.py") cmd = "" - cmd += f"python main.py --task {args.task} --benchmark --max_epochs 2 --min_epochs 1 --optimizer adam " + cmd += f"python main.py --task {args.task} --benchmark --epochs 2 " cmd += f"--results {args.results} " cmd += f"--logname {args.logname} " cmd += f"--exec_mode {args.mode} " diff --git a/PyTorch/Segmentation/nnUNet/scripts/train.py b/PyTorch/Segmentation/nnUNet/scripts/train.py index 3e68b16b..e836740e 100644 --- a/PyTorch/Segmentation/nnUNet/scripts/train.py +++ b/PyTorch/Segmentation/nnUNet/scripts/train.py @@ -24,6 +24,7 @@ parser.add_argument("--fold", type=int, required=True, choices=[0, 1, 2, 3, 4], parser.add_argument("--dim", type=int, required=True, choices=[2, 3], help="Dimension of UNet") parser.add_argument("--amp", action="store_true", help="Enable automatic mixed precision") parser.add_argument("--tta", action="store_true", help="Enable test time augmentation") +parser.add_argument("--resume_training", action="store_true", help="Resume training from checkpoint") parser.add_argument("--results", type=str, default="/results", help="Path to results directory") parser.add_argument("--logname", type=str, default="log", help="Name of dlloger output") @@ -40,4 +41,5 @@ if __name__ == "__main__": cmd += f"--gpus {args.gpus} " cmd += "--amp " if args.amp else "" cmd += "--tta " if args.tta else "" + cmd += "--resume_training " if args.resume_training else "" call(cmd, shell=True) diff --git a/PyTorch/Segmentation/nnUNet/utils/scheduler.py b/PyTorch/Segmentation/nnUNet/utils/scheduler.py new file mode 100644 index 00000000..c2105e26 --- /dev/null +++ b/PyTorch/Segmentation/nnUNet/utils/scheduler.py @@ -0,0 +1,17 @@ +import math + +from torch.optim.lr_scheduler import LambdaLR + + +class WarmupCosineSchedule(LambdaLR): + def __init__(self, optimizer, warmup_steps, t_total, cycles=0.5, last_epoch=-1): + self.warmup_steps = warmup_steps + self.t_total = t_total + self.cycles = cycles + super(WarmupCosineSchedule, self).__init__(optimizer, self.lr_lambda, last_epoch=last_epoch) + + def lr_lambda(self, step): + if step < self.warmup_steps: + return float(step) / float(max(1.0, self.warmup_steps)) + progress = float(step - self.warmup_steps) / float(max(1, self.t_total - self.warmup_steps)) + return max(0.0, 0.5 * (1.0 + math.cos(math.pi * float(self.cycles) * 2.0 * progress))) diff --git a/PyTorch/Segmentation/nnUNet/utils/utils.py b/PyTorch/Segmentation/nnUNet/utils/utils.py index 3ef36118..4b8c6dfc 100755 --- a/PyTorch/Segmentation/nnUNet/utils/utils.py +++ b/PyTorch/Segmentation/nnUNet/utils/utils.py @@ -101,7 +101,7 @@ def get_unet_params(args): strides, kernels, sizes = [], [], patch_size[:] while True: spacing_ratio = [spacing / min(spacings) for spacing in spacings] - stride = [2 if ratio <= 2 and size >= 8 else 1 for (ratio, size) in zip(spacing_ratio, sizes)] + stride = [2 if ratio <= 2 and size >= 2 * args.min_fmap else 1 for (ratio, size) in zip(spacing_ratio, sizes)] kernel = [3 if ratio <= 2 else 1 for ratio in spacing_ratio] if all(s == 1 for s in stride): break @@ -109,7 +109,7 @@ def get_unet_params(args): spacings = [i * j for i, j in zip(spacings, stride)] kernels.append(kernel) strides.append(stride) - if len(strides) == 5: + if len(strides) == 6: break strides.insert(0, len(spacings) * [1]) kernels.append(len(spacings) * [3]) @@ -184,13 +184,15 @@ def get_main_args(strings=None): arg("--logname", type=str, default=None, help="Name of dlloger output") arg("--task", type=str, help="Task number. MSD uses numbers 01-10") arg("--gpus", type=non_negative_int, default=1, help="Number of gpus") - arg("--learning_rate", type=float, default=0.001, help="Learning rate") + arg("--learning_rate", type=float, default=0.0008, help="Learning rate") arg("--gradient_clip_val", type=float, default=0, help="Gradient clipping norm value") arg("--negative_slope", type=float, default=0.01, help="Negative slope for LeakyReLU") arg("--tta", action="store_true", help="Enable test time augmentation") + arg("--brats", action="store_true", help="Enable BraTS specific training and inference") + arg("--deep_supervision", action="store_true", help="Enable deep supervision") + arg("--more_chn", action="store_true", help="Create encoder with more channels") arg("--amp", action="store_true", help="Enable automatic mixed precision") arg("--benchmark", action="store_true", help="Run model benchmarking") - arg("--residual", action="store_true", help="Enable residual block in encoder") arg("--focal", action="store_true", help="Use focal loss instead of cross entropy") arg("--sync_batchnorm", action="store_true", help="Enable synchronized batchnorm") arg("--save_ckpt", action="store_true", help="Enable saving checkpoint") @@ -200,20 +202,16 @@ def get_main_args(strings=None): arg("--ckpt_path", type=str, default=None, help="Path to checkpoint") arg("--fold", type=non_negative_int, default=0, help="Fold number") arg("--patience", type=positive_int, default=100, help="Early stopping patience") - arg("--lr_patience", type=positive_int, default=70, help="Patience for ReduceLROnPlateau scheduler") arg("--batch_size", type=positive_int, default=2, help="Batch size") arg("--val_batch_size", type=positive_int, default=4, help="Validation batch size") - arg("--steps", nargs="+", type=positive_int, required=False, help="Steps for multistep scheduler") arg("--profile", action="store_true", help="Run dlprof profiling") arg("--momentum", type=float, default=0.99, help="Momentum factor") arg("--weight_decay", type=float, default=0.0001, help="Weight decay (L2 penalty)") arg("--save_preds", action="store_true", help="Enable prediction saving") arg("--dim", type=int, choices=[2, 3], default=3, help="UNet dimension") arg("--resume_training", action="store_true", help="Resume training from the last checkpoint") - arg("--factor", type=float, default=0.3, help="Scheduler factor") arg("--num_workers", type=non_negative_int, default=8, help="Number of subprocesses to use for data loading") - arg("--min_epochs", type=non_negative_int, default=30, help="Force training for at least these many epochs") - arg("--max_epochs", type=non_negative_int, default=10000, help="Stop training after this number of epochs") + arg("--epochs", type=non_negative_int, default=1000, help="Number of training epochs") arg("--warmup", type=non_negative_int, default=5, help="Warmup iterations before collecting statistics") arg("--norm", type=str, choices=["instance", "batch", "group"], default="instance", help="Normalization layer") arg("--nvol", type=positive_int, default=1, help="Number of volumes which come into single batch size for 2D model") @@ -252,18 +250,22 @@ def get_main_args(strings=None): ) arg( "--scheduler", - type=str, - default="none", - choices=["none", "multistep", "cosine", "plateau"], - help="Learning rate scheduler", + action="store_true", + help="Enable cosine rate scheduler with warmup", ) arg( "--optimizer", type=str, - default="radam", - choices=["sgd", "radam", "adam"], + default="adam", + choices=["sgd", "adam"], help="Optimizer", ) + arg( + "--min_fmap", + type=non_negative_int, + default=4, + help="The minimal size that feature map can be reduced in bottleneck", + ) arg( "--blend", type=str,