konfai 1.0.0__py3-none-any.whl

konfai/__init__.py

@@ -0,0 +1,16 @@
+import os
+import datetime
+
+MODELS_DIRECTORY = lambda : os.environ["DL_API_MODELS_DIRECTORY"]
+CHECKPOINTS_DIRECTORY =lambda : os.environ["DL_API_CHECKPOINTS_DIRECTORY"]
+MODEL = lambda : os.environ["DL_API_MODEL"]
+PREDICTIONS_DIRECTORY =lambda : os.environ["DL_API_PREDICTIONS_DIRECTORY"]
+EVALUATIONS_DIRECTORY =lambda : os.environ["DL_API_EVALUATIONS_DIRECTORY"]
+STATISTICS_DIRECTORY = lambda : os.environ["DL_API_STATISTICS_DIRECTORY"]
+SETUPS_DIRECTORY = lambda : os.environ["DL_API_SETUPS_DIRECTORY"]
+CONFIG_FILE = lambda : os.environ["DEEP_LEARNING_API_CONFIG_FILE"]
+DL_API_STATE = lambda : os.environ["DL_API_STATE"]
+DEEP_LEARNING_API_ROOT = lambda : os.environ["DEEP_LEARNING_API_ROOT"]
+CUDA_VISIBLE_DEVICES = lambda : os.environ["CUDA_VISIBLE_DEVICES"]
+
+DATE = lambda : datetime.datetime.now().strftime("%Y_%m_%d_%H_%M_%S")

konfai/data/HDF5.py

@@ -0,0 +1,326 @@
+from abc import ABC, abstractmethod
+import SimpleITK as sitk
+import numpy as np
+import torch
+import os
+import torch.nn.functional as F
+from typing import Any, Iterator
+from typing import Union
+import itertools
+import copy 
+from functools import partial
+from KonfAI.konfai.utils.config import config
+from KonfAI.konfai.utils.utils import get_patch_slices_from_shape
+from KonfAI.konfai.utils.dataset import Dataset, Attribute
+from KonfAI.konfai.data.transform import Transform, Save
+from KonfAI.konfai.data.augmentation import DataAugmentationsList
+
+
+class PathCombine(ABC):
+
+    def __init__(self) -> None:
+        self.data: torch.Tensor = None
+        self.overlap: int = None
+    
+    """
+    A = slice(0, overlap)
+    B = slice(-overlap, None)
+    C = slice(overlap, -overlap)
+    
+    1D 
+        A+B
+    2D : 
+        AA+AB+BA+BB
+        
+        AC+BC
+        CA+CB
+    3D : 
+        AAA+AAB+ABA+ABB+BAA+BAB+BBA+BBB
+        
+        AAC+ABC+BAC+BBC
+        ACA+ACB+BCA+BCB
+        CAA+CAB+CBA+CBB
+
+        ACC+BCC
+        CAC+CBC
+        CCA+CCB
+    
+    """
+    def setPatchConfig(self, patch_size: list[int], overlap: int):
+        self.data = F.pad(torch.ones([size-overlap*2 for size in patch_size]), [overlap]*2*len(patch_size), mode="constant", value=0)
+        self.data = self._setFunction(self.data, overlap)
+        dim = len(patch_size)
+
+        A = slice(0, overlap)
+        B = slice(-overlap, None)
+        C = slice(overlap, -overlap)
+        
+        for i in range(dim):
+            slices_badge = list(itertools.product(*[[A, B] for _ in range(dim-i)]))
+            for indexs in itertools.combinations([0,1,2], i):
+                result = []
+                for slices in slices_badge:
+                    slices = list(slices)
+                    for index in indexs:
+                        slices.insert(index, C)    
+                    result.append(tuple(slices))
+                for patch, s in zip(PathCombine._normalise([self.data[s] for s in result]), result):
+                    self.data[s] = patch
+
+
+    @staticmethod
+    def _normalise(patchs: list[torch.Tensor]) -> list[torch.Tensor]:
+        data_sum = torch.sum(torch.concat([patch.unsqueeze(0) for patch in patchs], dim=0), dim=0)
+        return [d/data_sum for d in patchs]
+            
+    def __call__(self, input: torch.Tensor) -> torch.Tensor:
+        return self.data.repeat([input.shape[0]]+[1]*(len(input.shape)-1)).to(input.device)*input
+    
+    @abstractmethod
+    def _setFunction(self, data: torch.Tensor, overlap: int) -> torch.Tensor:
+        pass
+
+class Mean(PathCombine):
+
+    @config("Mean")
+    def __init__(self) -> None:
+        super().__init__()
+
+    def _setFunction(self, data: torch.Tensor, overlap: int) -> torch.Tensor:
+        return torch.ones_like(self.data)
+    
+class Cosinus(PathCombine):
+
+    @config("Cosinus")
+    def __init__(self) -> None:
+        super().__init__()
+
+    def _function_sides(self, overlap: int, x: float):
+        return np.clip(np.cos(np.pi/(2*(overlap+1))*x), 0, 1)
+
+    def _setFunction(self, data: torch.Tensor, overlap: int) -> torch.Tensor:
+        image = sitk.GetImageFromArray(np.asarray(data, dtype=np.uint8))
+        danielssonDistanceMapImageFilter = sitk.DanielssonDistanceMapImageFilter()
+        distance = torch.tensor(sitk.GetArrayFromImage(danielssonDistanceMapImageFilter.Execute(image)))
+        return distance.apply_(partial(self._function_sides, overlap))
+        
+class Accumulator():
+
+    def __init__(self, patch_slices: list[tuple[slice]], patch_size: list[int], patchCombine: Union[PathCombine, None] = None, batch: bool = True) -> None:
+        self._layer_accumulator: list[Union[torch.Tensor, None]] = [None for i in range(len(patch_slices))]
+        self.patch_slices = []
+        for patch in patch_slices:
+            slices = []
+            for s, shape in zip(patch, patch_size):
+                slices.append(slice(s.start, s.start+shape))
+            self.patch_slices.append(tuple(slices))
+        self.shape = max([[v.stop for v in patch] for patch in patch_slices])
+        self.patch_size = patch_size
+        self.patchCombine = patchCombine
+        self.batch = batch
+
+    def addLayer(self, index: int, layer: torch.Tensor) -> None:
+        self._layer_accumulator[index] = layer
+    
+    def isFull(self) -> bool:
+        return len(self.patch_slices) == len([v for v in self._layer_accumulator if v is not None])
+
+    def assemble(self) -> torch.Tensor:
+        if all([self._layer_accumulator[0].shape[i-len(self.patch_size)] == size for i, size in enumerate(self.patch_size)]): 
+            N = 2 if self.batch else 1
+            
+            result = torch.zeros((list(self._layer_accumulator[0].shape[:N])+list(max([[v.stop for v in patch] for patch in self.patch_slices]))), dtype=self._layer_accumulator[0].dtype).to(self._layer_accumulator[0].device)
+            for patch_slice, data in zip(self.patch_slices, self._layer_accumulator):
+                slices_dest = tuple([slice(result.shape[i]) for i in range(N)] + list(patch_slice))
+
+                for dim, s in enumerate(patch_slice):
+                    if s.stop-s.start == 1:
+                        data = data.unsqueeze(dim=dim+N)
+                if self.patchCombine is not None:
+                    result[slices_dest] += self.patchCombine(data)
+                else:
+                    result[slices_dest] = data
+            result = result[tuple([slice(None, None)]+[slice(0, s) for s in self.shape])]
+        else:
+            result = torch.cat(tuple(self._layer_accumulator), dim=0)
+        self._layer_accumulator.clear()
+        return result
+
+class Patch(ABC):
+
+    def __init__(self, patch_size: list[int], overlap: Union[int, None], path_mask: Union[str, None] = None, padValue: float = 0, extend_slice: int = 0) -> None:
+        self.patch_size = patch_size
+        self.overlap = overlap
+        self._patch_slices : dict[int, list[tuple[slice]]] = {}
+        self._nb_patch_per_dim: dict[int, list[tuple[int, bool]]] = {}
+        self.path_mask = path_mask
+        self.mask = None
+        self.padValue = padValue
+        self.extend_slice = extend_slice
+        if self.path_mask is not None:
+            if os.path.exists(self.path_mask):
+                self.mask = torch.tensor(sitk.GetArrayFromImage(sitk.ReadImage(self.path_mask)))
+            else:
+                raise NameError('Mask file not found')
+            
+    def load(self, shape : dict[int, list[int]], a: int = 0) -> None:
+        self._patch_slices[a], self._nb_patch_per_dim[a] = get_patch_slices_from_shape(self.patch_size, shape, self.overlap)
+
+    def getPatch_slices(self, a: int = 0):
+        return self._patch_slices[a]
+    
+    @abstractmethod
+    def getData(self, data : torch.Tensor, index : int, a: int, isInput: bool) -> torch.Tensor:
+        pass
+    
+    def getData(self, data : torch.Tensor, index : int, a: int, isInput: bool) -> list[torch.Tensor]:
+        slices_pre = []
+        for max in data.shape[:-len(self.patch_size)]:
+            slices_pre.append(slice(max))
+        extend_slice = self.extend_slice if isInput else 0
+
+        bottom = extend_slice//2
+        top = int(np.ceil(extend_slice/2))
+        s = slice(self._patch_slices[a][index][0].start-bottom if self._patch_slices[a][index][0].start-bottom >= 0 else 0, self._patch_slices[a][index][0].stop+top if self._patch_slices[a][index][0].stop+top <= data.shape[len(slices_pre)] else data.shape[len(slices_pre)])
+        slices =  [s] + list(self._patch_slices[a][index][1:])
+        data_sliced = data[slices_pre+slices]
+        if data_sliced.shape[len(slices_pre)] < bottom+top+1:
+            pad_bottom = 0
+            pad_top = 0
+            if self._patch_slices[a][index][0].start-bottom < 0:
+                pad_bottom = bottom-s.start
+            if self._patch_slices[a][index][0].stop+top > data.shape[len(slices_pre)]:
+                pad_top = self._patch_slices[a][index][0].stop+top-data.shape[len(slices_pre)]
+            data_sliced = F.pad(data_sliced, [0 for _ in range((len(slices)-1)*2)]+[pad_bottom, pad_top], 'reflect')
+
+        padding = []
+        for dim_it, _slice in enumerate(reversed(slices)):
+            p = 0 if _slice.start+self.patch_size[-dim_it-1] <= data.shape[-dim_it-1] else self.patch_size[-dim_it-1]-(data.shape[-dim_it-1]-_slice.start)
+            padding.append(0)
+            padding.append(p)
+
+        data_sliced = F.pad(data_sliced, tuple(padding), "constant", 0 if data_sliced.dtype == torch.uint8 and self.padValue < 0 else self.padValue)
+        if self.mask is not None:
+            outside = torch.ones_like(data_sliced)*(0 if data_sliced.dtype == torch.uint8 and self.padValue < 0 else self.padValue)
+            data_sliced = torch.where(self.mask == 0, outside, data_sliced)
+        
+        for d in [i for i, v in enumerate(reversed(self.patch_size)) if v == 1]:
+            data_sliced = torch.squeeze(data_sliced, dim = len(data_sliced.shape)-d-1)
+        return torch.cat([data_sliced[:, i, ...] for i in range(data_sliced.shape[1])], dim=0) if extend_slice > 0 else data_sliced
+
+    def getSize(self, a: int = 0) -> int:
+        return len(self._patch_slices[a])
+
+class DatasetPatch(Patch):
+
+    @config("Patch")
+    def __init__(self, patch_size : list[int] = [128, 256, 256], overlap : Union[int, None] = None, mask: Union[str, None] = None, padValue: float = 0, extend_slice: int = 0) -> None:
+        super().__init__(patch_size, overlap, mask, padValue, extend_slice)
+
+class ModelPatch(Patch):
+
+    @config("Patch")
+    def __init__(self, patch_size : list[int] = [128, 256, 256], overlap : Union[int, None] = None, patchCombine: Union[str, None] = None, mask: Union[str, None] = None, padValue: float = 0, extend_slice: int = 0) -> None:
+        super().__init__(patch_size, overlap, mask, padValue, extend_slice)
+        self.patchCombine = patchCombine
+
+    def disassemble(self, *dataList: torch.Tensor) -> Iterator[list[torch.Tensor]]:
+        for i in range(self.getSize()):
+            yield [self.getData(data, i, 0, True) for data in dataList]
+
+class DatasetManager():
+
+    def __init__(self, index: int, group_src: str, group_dest : str, name: str, dataset : Dataset, patch : Union[DatasetPatch, None], pre_transforms : list[Transform], dataAugmentationsList : list[DataAugmentationsList]) -> None:
+        self.group_src = group_src
+        self.group_dest = group_dest
+        self.name = name
+        self.index = index
+        self.dataset = dataset
+        self.loaded = False
+        self.cache_attributes: list[Attribute] = []
+        _shape, cache_attribute =  self.dataset.getInfos(self.group_src, name)
+        self.cache_attributes.append(cache_attribute)
+        _shape = list(_shape[1:])
+        
+        self.data : list[torch.Tensor] = list()
+
+        for transformFunction in pre_transforms:
+            _shape = transformFunction.transformShape(_shape, cache_attribute)
+        
+        self.patch = DatasetPatch(patch_size=patch.patch_size, overlap=patch.overlap, mask=patch.path_mask, padValue=patch.padValue, extend_slice=patch.extend_slice) if patch else DatasetPatch(_shape)
+        self.patch.load(_shape, 0)
+        self.shape = _shape
+        self.dataAugmentationsList = dataAugmentationsList
+        self.resetAugmentation()
+        self.cache_attributes_bak = copy.deepcopy(self.cache_attributes)
+    
+    def resetAugmentation(self):
+        i = 1
+        for dataAugmentations in self.dataAugmentationsList:
+            shape = []
+            caches_attribute = []
+            for _ in range(dataAugmentations.nb):
+                shape.append(self.shape)
+                caches_attribute.append(copy.deepcopy(self.cache_attributes[0]))
+
+            for dataAugmentation in dataAugmentations.dataAugmentations:
+                shape = dataAugmentation.state_init(self.index, shape, caches_attribute)
+            for it, s in enumerate(shape):
+                self.cache_attributes.append(caches_attribute[it])
+                self.patch.load(s, i)
+                i+=1
+
+    def load(self, pre_transform : list[Transform], dataAugmentationsList : list[DataAugmentationsList], device: torch.device) -> None:
+        if self.loaded:
+            return
+        i = len(pre_transform)
+        data = None
+        for transformFunction in reversed(pre_transform):
+            if isinstance(transformFunction, Save):
+                filename, format = transformFunction.save.split(":")
+                dataset = Dataset(filename, format)
+                if dataset.isDatasetExist(self.group_dest, self.name):
+                    data, attrib = dataset.readData(self.group_dest, self.name)
+                    self.cache_attributes[0].update(attrib)
+                    break
+            i-=1
+        
+        if i==0:
+            data, _ = self.dataset.readData(self.group_src, self.name)
+
+
+        data = torch.from_numpy(data)
+
+        if len(pre_transform):
+            for transformFunction in pre_transform[i:]:
+                data = transformFunction(self.name, data, self.cache_attributes[0])
+                if isinstance(transformFunction, Save):
+                    filename, format = transformFunction.save.split(":")
+                    dataset = Dataset(filename, format)
+                    dataset.write(self.group_dest, self.name, data.numpy(), self.cache_attributes[0])
+        self.data : list[torch.Tensor] = list()
+        self.data.append(data)
+
+        for dataAugmentations in dataAugmentationsList:
+            a_data = [data.clone() for _ in range(dataAugmentations.nb)]
+            for dataAugmentation in dataAugmentations.dataAugmentations:
+                a_data = dataAugmentation(self.index, a_data, device)
+            
+            for d in a_data:
+                self.data.append(d)
+        self.loaded = True
+
+    def unload(self) -> None:
+        if hasattr(self, "data"):
+            del self.data
+        self.cache_attributes = copy.deepcopy(self.cache_attributes_bak)
+        self.loaded = False
+    
+    def getData(self, index : int, a : int, post_transforms : list[Transform], isInput: bool) -> torch.Tensor:
+        data = self.patch.getData(self.data[a], index, a, isInput)
+        for transformFunction in post_transforms:
+            data = transformFunction(self.name, data, self.cache_attributes[a])
+        return data
+
+    def getSize(self, a: int) -> int:
+        return self.patch.getSize(a)