Refactor model structure to support KonfAI inference from Hugging Face repositories

M298.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:8f04169298561a6e1fcad186ef1a24623b48edf5f776710793de13a5a9c6d40e
-size 66225317

M730.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:462c3c7ca2eb1a40a0984d5b1f004cfbf617c3e9218af65f442006e132fd594e
-size 123170169

M731.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:f9998319e380221bc0e69dae2d0f120f7e9d7c38254dc982afd05aa9f44877b1
-size 123169913

M732.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ce4d777f8da2ebdeb2a4b19f976f8eac36bba252c5f63fccf76f9e3dd8bc296c
-size 66217253

M733.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:3340e70c92bb9c095f1db8e139319efeb7fb8e444f04ffbd8421c70e657fc444
-size 22435409

M853.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:024cadec3839c0b552fb07b4ba866073430c3856d06ef5a2e0bb2896d20e7596
-size 22434641

total-3mm/metadata.json

@@ -0,0 +1,7 @@
+{
+    "display_name": "Total 3mm",
+    "short_description": "",
+    "description": "",
+    "tta": 0,
+    "mc_dropout": 0
+}

total-3mm/requirements.txt

@@ -0,0 +1 @@
+segmentation_models_pytorch

total/Model.py

@@ -0,0 +1,78 @@
+import torch
+from konfai.network import network, blocks
+from konfai.predictor import Reduction
+
+class ConvBlock(network.ModuleArgsDict):
+    def __init__(self, in_channels : int, out_channels : int, stride: int = 1 ) -> None:
+        super().__init__()
+        self.add_module("Conv_0", torch.nn.Conv3d(in_channels=in_channels, out_channels=out_channels, kernel_size=3, stride=stride, padding=1, bias=True))
+        self.add_module("Norm_0", torch.nn.InstanceNorm3d(num_features=out_channels, affine=True))
+        self.add_module("Activation_0", torch.nn.LeakyReLU(negative_slope=0.01, inplace=True))
+        self.add_module("Conv_1", torch.nn.Conv3d(in_channels=out_channels, out_channels=out_channels, kernel_size=3, stride=1, padding=1, bias=True))
+        self.add_module("Norm_1", torch.nn.InstanceNorm3d(num_features=out_channels, affine=True))
+        self.add_module("Activation_1", torch.nn.LeakyReLU(negative_slope=0.01, inplace=True))
+        
+class UNetHead(network.ModuleArgsDict):
+    def __init__(self, in_channels: int, nb_class: int) -> None:
+        super().__init__()
+        self.add_module("Conv", torch.nn.Conv3d(in_channels = in_channels, out_channels = nb_class, kernel_size = 1, stride = 1, padding = 0))
+        self.add_module("Softmax", torch.nn.Softmax(dim=1))
+    
+class UNetBlock(network.ModuleArgsDict):
+
+    def __init__(self, channels, mri: bool, i : int = 0) -> None:
+        super().__init__()
+        self.add_module("DownConvBlock", ConvBlock(in_channels=channels[0], out_channels=channels[1], stride= ((1,2,2) if mri and i > 4 else 2) if i>0 else 1))
+     
+        if len(channels) > 2:
+            self.add_module("UNetBlock", UNetBlock(channels[1:], mri, i+1))
+            self.add_module("UpConvBlock", ConvBlock(in_channels=channels[1]*2, out_channels=channels[1]))
+            
+        if i > 0:
+            self.add_module("CONV_TRANSPOSE", torch.nn.ConvTranspose3d(in_channels = channels[1], out_channels = channels[0], kernel_size =  (1,2,2) if mri and i > 4 else 2, stride = (1,2,2) if mri and i > 4 else 2, padding = 0))
+            self.add_module("SkipConnection", blocks.Concat(), in_branch=[0, 1])
+
+class Unet_TS(network.Network):
+
+    def __init__(self,
+        optimizer: network.OptimizerLoader = network.OptimizerLoader(),
+        schedulers: dict[str, network.LRSchedulersLoader] = {
+            "default:ReduceLROnPlateau": network.LRSchedulersLoader(0)
+        },
+        outputs_criterions: dict[str, network.TargetCriterionsLoader] = {"default": network.TargetCriterionsLoader()},
+        channels = [1, 32, 64, 128, 320, 320], 
+        mri: bool = False) -> None:
+        super().__init__(
+            in_channels=channels[0],
+            optimizer=optimizer,
+            schedulers=schedulers,
+            outputs_criterions=outputs_criterions,
+            patch=None,
+            dim=3,
+        )
+        self.add_module("UNetBlock", UNetBlock(channels, mri))
+        self.add_module("Head", UNetHead(channels[1], 42))
+
+    def load(
+        self,
+        state_dict: dict[str, dict[str, torch.Tensor] | int],
+        init: bool = True,
+        ema: bool = False,
+    ):
+        nb_class, in_channels = state_dict["Model"]["Unet_TS"]["Head.Conv.weight"].shape[:2]
+        self["Head"].add_module("Conv", torch.nn.Conv3d(in_channels = in_channels, out_channels = nb_class, kernel_size = 1, stride = 1, padding = 0))
+        super().load(state_dict, init, ema)
+
+class Combine(Reduction):
+
+    def __init__(self):
+        pass
+
+    def __call__(self, tensors: list[torch.Tensor]) -> torch.Tensor:
+        fg_all = torch.cat([p[:, 1:, ...] for p in tensors], dim=1)
+        sum_fg = fg_all.sum(dim=1, keepdim=True)
+        bg = (1.0 - sum_fg).clamp(min=1e-6)
+
+        probs = torch.cat([bg, fg_all], dim=1)
+
+        return probs / probs.sum(dim=1, keepdim=True)

total/metadata.json

@@ -0,0 +1,7 @@
+{
+    "display_name": "total",
+    "short_description": "",
+    "description": "",
+    "tta": 0,
+    "mc_dropout": 0
+}

total/requirements.txt

@@ -0,0 +1 @@
+segmentation_models_pytorch

total_mr-3mm/Model.py

@@ -0,0 +1,78 @@
+import torch
+from konfai.network import network, blocks
+from konfai.predictor import Reduction
+
+class ConvBlock(network.ModuleArgsDict):
+    def __init__(self, in_channels : int, out_channels : int, stride: int = 1 ) -> None:
+        super().__init__()
+        self.add_module("Conv_0", torch.nn.Conv3d(in_channels=in_channels, out_channels=out_channels, kernel_size=3, stride=stride, padding=1, bias=True))
+        self.add_module("Norm_0", torch.nn.InstanceNorm3d(num_features=out_channels, affine=True))
+        self.add_module("Activation_0", torch.nn.LeakyReLU(negative_slope=0.01, inplace=True))
+        self.add_module("Conv_1", torch.nn.Conv3d(in_channels=out_channels, out_channels=out_channels, kernel_size=3, stride=1, padding=1, bias=True))
+        self.add_module("Norm_1", torch.nn.InstanceNorm3d(num_features=out_channels, affine=True))
+        self.add_module("Activation_1", torch.nn.LeakyReLU(negative_slope=0.01, inplace=True))
+        
+class UNetHead(network.ModuleArgsDict):
+    def __init__(self, in_channels: int, nb_class: int) -> None:
+        super().__init__()
+        self.add_module("Conv", torch.nn.Conv3d(in_channels = in_channels, out_channels = nb_class, kernel_size = 1, stride = 1, padding = 0))
+        self.add_module("Softmax", torch.nn.Softmax(dim=1))
+    
+class UNetBlock(network.ModuleArgsDict):
+
+    def __init__(self, channels, mri: bool, i : int = 0) -> None:
+        super().__init__()
+        self.add_module("DownConvBlock", ConvBlock(in_channels=channels[0], out_channels=channels[1], stride= ((1,2,2) if mri and i > 4 else 2) if i>0 else 1))
+     
+        if len(channels) > 2:
+            self.add_module("UNetBlock", UNetBlock(channels[1:], mri, i+1))
+            self.add_module("UpConvBlock", ConvBlock(in_channels=channels[1]*2, out_channels=channels[1]))
+            
+        if i > 0:
+            self.add_module("CONV_TRANSPOSE", torch.nn.ConvTranspose3d(in_channels = channels[1], out_channels = channels[0], kernel_size =  (1,2,2) if mri and i > 4 else 2, stride = (1,2,2) if mri and i > 4 else 2, padding = 0))
+            self.add_module("SkipConnection", blocks.Concat(), in_branch=[0, 1])
+
+class Unet_TS(network.Network):
+
+    def __init__(self,
+        optimizer: network.OptimizerLoader = network.OptimizerLoader(),
+        schedulers: dict[str, network.LRSchedulersLoader] = {
+            "default:ReduceLROnPlateau": network.LRSchedulersLoader(0)
+        },
+        outputs_criterions: dict[str, network.TargetCriterionsLoader] = {"default": network.TargetCriterionsLoader()},
+        channels = [1, 32, 64, 128, 320, 320], 
+        mri: bool = False) -> None:
+        super().__init__(
+            in_channels=channels[0],
+            optimizer=optimizer,
+            schedulers=schedulers,
+            outputs_criterions=outputs_criterions,
+            patch=None,
+            dim=3,
+        )
+        self.add_module("UNetBlock", UNetBlock(channels, mri))
+        self.add_module("Head", UNetHead(channels[1], 42))
+
+    def load(
+        self,
+        state_dict: dict[str, dict[str, torch.Tensor] | int],
+        init: bool = True,
+        ema: bool = False,
+    ):
+        nb_class, in_channels = state_dict["Model"]["Unet_TS"]["Head.Conv.weight"].shape[:2]
+        self["Head"].add_module("Conv", torch.nn.Conv3d(in_channels = in_channels, out_channels = nb_class, kernel_size = 1, stride = 1, padding = 0))
+        super().load(state_dict, init, ema)
+
+class Combine(Reduction):
+
+    def __init__(self):
+        pass
+
+    def __call__(self, tensors: list[torch.Tensor]) -> torch.Tensor:
+        fg_all = torch.cat([p[:, 1:, ...] for p in tensors], dim=1)
+        sum_fg = fg_all.sum(dim=1, keepdim=True)
+        bg = (1.0 - sum_fg).clamp(min=1e-6)
+
+        probs = torch.cat([bg, fg_all], dim=1)
+
+        return probs / probs.sum(dim=1, keepdim=True)

total_mr-3mm/metadata.json

@@ -0,0 +1,7 @@
+{
+    "display_name": "Total MR 3mm",
+    "short_description": "",
+    "description": "",
+    "tta": 0,
+    "mc_dropout": 0
+}

total_mr-3mm/requirements.txt

@@ -0,0 +1 @@
+segmentation_models_pytorch

total_mr/Model.py

@@ -0,0 +1,78 @@
+import torch
+from konfai.network import network, blocks
+from konfai.predictor import Reduction
+
+class ConvBlock(network.ModuleArgsDict):
+    def __init__(self, in_channels : int, out_channels : int, stride: int = 1 ) -> None:
+        super().__init__()
+        self.add_module("Conv_0", torch.nn.Conv3d(in_channels=in_channels, out_channels=out_channels, kernel_size=3, stride=stride, padding=1, bias=True))
+        self.add_module("Norm_0", torch.nn.InstanceNorm3d(num_features=out_channels, affine=True))
+        self.add_module("Activation_0", torch.nn.LeakyReLU(negative_slope=0.01, inplace=True))
+        self.add_module("Conv_1", torch.nn.Conv3d(in_channels=out_channels, out_channels=out_channels, kernel_size=3, stride=1, padding=1, bias=True))
+        self.add_module("Norm_1", torch.nn.InstanceNorm3d(num_features=out_channels, affine=True))
+        self.add_module("Activation_1", torch.nn.LeakyReLU(negative_slope=0.01, inplace=True))
+        
+class UNetHead(network.ModuleArgsDict):
+    def __init__(self, in_channels: int, nb_class: int) -> None:
+        super().__init__()
+        self.add_module("Conv", torch.nn.Conv3d(in_channels = in_channels, out_channels = nb_class, kernel_size = 1, stride = 1, padding = 0))
+        self.add_module("Softmax", torch.nn.Softmax(dim=1))
+    
+class UNetBlock(network.ModuleArgsDict):
+
+    def __init__(self, channels, mri: bool, i : int = 0) -> None:
+        super().__init__()
+        self.add_module("DownConvBlock", ConvBlock(in_channels=channels[0], out_channels=channels[1], stride= ((1,2,2) if mri and i > 4 else 2) if i>0 else 1))
+     
+        if len(channels) > 2:
+            self.add_module("UNetBlock", UNetBlock(channels[1:], mri, i+1))
+            self.add_module("UpConvBlock", ConvBlock(in_channels=channels[1]*2, out_channels=channels[1]))
+            
+        if i > 0:
+            self.add_module("CONV_TRANSPOSE", torch.nn.ConvTranspose3d(in_channels = channels[1], out_channels = channels[0], kernel_size =  (1,2,2) if mri and i > 4 else 2, stride = (1,2,2) if mri and i > 4 else 2, padding = 0))
+            self.add_module("SkipConnection", blocks.Concat(), in_branch=[0, 1])
+
+class Unet_TS(network.Network):
+
+    def __init__(self,
+        optimizer: network.OptimizerLoader = network.OptimizerLoader(),
+        schedulers: dict[str, network.LRSchedulersLoader] = {
+            "default:ReduceLROnPlateau": network.LRSchedulersLoader(0)
+        },
+        outputs_criterions: dict[str, network.TargetCriterionsLoader] = {"default": network.TargetCriterionsLoader()},
+        channels = [1, 32, 64, 128, 320, 320], 
+        mri: bool = False) -> None:
+        super().__init__(
+            in_channels=channels[0],
+            optimizer=optimizer,
+            schedulers=schedulers,
+            outputs_criterions=outputs_criterions,
+            patch=None,
+            dim=3,
+        )
+        self.add_module("UNetBlock", UNetBlock(channels, mri))
+        self.add_module("Head", UNetHead(channels[1], 42))
+
+    def load(
+        self,
+        state_dict: dict[str, dict[str, torch.Tensor] | int],
+        init: bool = True,
+        ema: bool = False,
+    ):
+        nb_class, in_channels = state_dict["Model"]["Unet_TS"]["Head.Conv.weight"].shape[:2]
+        self["Head"].add_module("Conv", torch.nn.Conv3d(in_channels = in_channels, out_channels = nb_class, kernel_size = 1, stride = 1, padding = 0))
+        super().load(state_dict, init, ema)
+
+class Combine(Reduction):
+
+    def __init__(self):
+        pass
+
+    def __call__(self, tensors: list[torch.Tensor]) -> torch.Tensor:
+        fg_all = torch.cat([p[:, 1:, ...] for p in tensors], dim=1)
+        sum_fg = fg_all.sum(dim=1, keepdim=True)
+        bg = (1.0 - sum_fg).clamp(min=1e-6)
+
+        probs = torch.cat([bg, fg_all], dim=1)
+
+        return probs / probs.sum(dim=1, keepdim=True)

total_mr/metadata.json

@@ -0,0 +1,7 @@
+{
+    "display_name": "Total MR",
+    "short_description": "",
+    "description": "",
+    "tta": 0,
+    "mc_dropout": 0
+}

total_mr/requirements.txt

@@ -0,0 +1 @@
+segmentation_models_pytorch