spacr 0.2.81__py3-none-any.whl → 0.3.1__py3-none-any.whl

spacr/resources/MEDIAR/config/baseline.json

@@ -0,0 +1,60 @@
+{
+    "data_setups":{
+        "labeled":{
+            "root": "/home/gihun/data/CellSeg/",
+            "mapping_file": "./train_tools/data_utils/mapping_labeled.json",
+            "tuning_mapping_file": "/home/gihun/CellSeg/train_tools/data_utils/mapping_tuning.json",
+            "batch_size": 8, 
+            "valid_portion": 0.1
+        },
+        "unlabeled":{
+            "enabled": false
+        },
+        "public":{
+            "enabled": false
+        }        
+    },
+    "train_setups":{
+        "model":{
+            "name": "swinunetr",
+            "params": {
+                "img_size": 512,
+                "in_channels": 3,
+                "out_channels": 3,
+                "spatial_dims": 2
+            },
+            "pretrained":{
+                "enabled": false
+            }
+        },
+        "trainer": {
+            "name": "baseline",
+            "params": {            
+                "num_epochs": 200,
+                "valid_frequency": 1,
+                "device": "cuda:0",
+                "algo_params": {}
+            }
+        },
+        "optimizer":{
+            "name": "adamw",
+            "params": {"lr": 5e-5}
+        },
+        "scheduler":{
+            "enabled": false
+        },
+        "seed": 19940817
+    },
+    "pred_setups":{
+        "input_path":"/home/gihun/data/CellSeg/Official/TuningSet",
+        "output_path": "./results/baseline",
+        "make_submission": true,
+        "exp_name": "baseline",
+        "algo_params": {}
+    },
+    "wandb_setups":{
+        "project": "CellSeg",
+        "group": "Baseline",
+        "name": "baseline"
+    }
+}

spacr/resources/MEDIAR/config/mediar_example.json

@@ -0,0 +1,72 @@
+{
+    "data_setups":{
+        "labeled":{
+            "root": "/home/gihun/data/CellSeg/",
+            "mapping_file": "./train_tools/data_utils/mapping_labeled.json",
+            "amplified": false,
+            "batch_size": 8, 
+            "valid_portion": 0.1
+        },
+        "public":{
+            "enabled": true,
+            "params":{
+                "root": "/home/gihun/data/CellSeg/",
+                "mapping_file": "./train_tools/data_utils/mapping_public.json",
+                "batch_size": 1
+            }
+        },
+        "unlabeled":{
+            "enabled": false
+        }        
+    },
+    "train_setups":{
+        "model":{
+            "name": "mediar-former",
+            "params": {
+                "encoder_name": "mit_b5",
+                "encoder_weights": "imagenet",
+                "decoder_channels": [1024, 512, 256, 128, 64],
+                "decoder_pab_channels": 256,
+                "in_channels": 3,
+                "classes": 3
+            },
+            "pretrained":{
+                "enabled": false,
+                "weights": "./weights/pretrained/phase2.pth",
+                "strict": false
+            }
+        },
+        "trainer": {
+            "name": "mediar",
+            "params": {            
+                "num_epochs": 200,
+                "valid_frequency": 1,
+                "device": "cuda:0",
+                "amp": true,
+                "algo_params": {"with_public": false}
+            }
+        },
+        "optimizer":{
+            "name": "adamw",
+            "params": {"lr": 5e-5}
+        },
+        "scheduler":{
+            "enabled": true,
+            "name": "cosine",
+            "params": {"T_max": 100, "eta_min": 1e-7}
+        },
+        "seed": 19940817
+    },
+    "pred_setups":{
+        "input_path":"/home/gihun/data/CellSeg/Official/TuningSet",
+        "output_path": "./mediar_example",
+        "make_submission": true,
+        "exp_name": "mediar_example",
+        "algo_params": {"use_tta": false}
+    },
+    "wandb_setups":{
+        "project": "CellSeg",
+        "group": "MEDIAR",
+        "name": "mediar_example"
+    }
+}

spacr/resources/MEDIAR/config/pred/pred_mediar.json

@@ -0,0 +1,17 @@
+{
+    "pred_setups":{
+        "name": "medair",
+        "input_path":"input_path",
+        "output_path": "./test",
+        "make_submission": true,
+        "model_path": "model_path",
+        "device": "cuda:0",
+        "model":{
+            "name": "mediar-former",
+            "params": {},
+            "pretrained":{
+                "enabled": false
+            }
+        }
+    }
+}

spacr/resources/MEDIAR/config/step1_pretraining/phase1.json

@@ -0,0 +1,55 @@
+{
+    "data_setups":{
+        "labeled":{
+            "root": "/home/gihun/MEDIAR/",
+            "mapping_file": "./train_tools/data_utils/mapping_public.json",
+            "mapping_file_tuning": "/home/gihun/MEDIAR/train_tools/data_utils/mapping_tuning.json",
+            "batch_size": 9, 
+            "valid_portion": 0
+        },
+        "public":{
+            "enabled": false,
+            "params":{}
+        }     
+    },
+    "train_setups":{
+        "model":{
+            "name": "mediar-former",
+            "params": {},
+            "pretrained":{
+                "enabled": false
+            }
+        },
+        "trainer": {
+            "name": "mediar",
+            "params": {            
+                "num_epochs": 80,
+                "valid_frequency": 10,
+                "device": "cuda:0",
+                "amp": true,
+                "algo_params": {"with_public": false}
+            }
+        },
+        "optimizer":{
+            "name": "adamw",
+            "ft_rate": 1.0,
+            "params": {"lr": 5e-5}
+        },
+        "scheduler":{
+            "enabled": true,
+            "name": "cosine",
+            "params": {"T_max": 80, "eta_min": 1e-6}
+        },
+        "seed": 19940817
+    },
+    "pred_setups":{
+        "input_path":"/home/gihun/MEDIAR/data/Official/Tuning/images",
+        "output_path": "./mediar_pretrained_phase1",
+        "make_submission": false
+    },
+    "wandb_setups":{
+        "project": "CellSeg",
+        "group": "Pretraining",
+        "name": "phase1"
+    }
+}

spacr/resources/MEDIAR/config/step1_pretraining/phase2.json

@@ -0,0 +1,58 @@
+{
+    "data_setups":{
+        "labeled":{
+            "root": "/home/gihun/MEDIAR/",
+            "mapping_file": "./train_tools/data_utils/mapping_labeled.json",
+            "mapping_file_tuning": "/home/gihun/MEDIAR/train_tools/data_utils/mapping_tuning.json",
+            "join_mapping_file": "./train_tools/data_utils/mapping_public.json",
+            "batch_size": 9, 
+            "valid_portion": 0
+        },
+        "unlabeled":{
+            "enabled": false
+        },
+        "public":{
+            "enabled": false
+        }        
+    },
+    "train_setups":{
+        "model":{
+            "name": "mediar-former",
+            "params": {},
+            "pretrained":{
+                "enabled": false
+            }
+        },
+        "trainer": {
+            "name": "mediar",
+            "params": {            
+                "num_epochs": 60,
+                "valid_frequency": 10,
+                "device": "cuda:0",
+                "amp": true,
+                "algo_params": {"with_public": false}
+            }
+        },
+        "optimizer":{
+            "name": "adamw",
+            "ft_rate": 1.0,
+            "params": {"lr": 5e-5}
+        },
+        "scheduler":{
+            "enabled": true,
+            "name": "cosine",
+            "params": {"T_max": 60, "eta_min": 1e-6}
+        },
+        "seed": 19940817
+    },
+    "pred_setups":{
+        "input_path":"/home/gihun/MEDIAR/data/Official/Tuning/images",
+        "output_path": "./mediar_pretrain_phase2",
+        "make_submission": false
+    },
+    "wandb_setups":{
+        "project": "CellSeg",
+        "group": "Pretraining",
+        "name": "phase2"
+    }
+}

spacr/resources/MEDIAR/config/step2_finetuning/finetuning1.json

@@ -0,0 +1,66 @@
+{
+    "data_setups":{
+        "labeled":{
+            "root": "/home/gihun/MEDIAR/",
+            "mapping_file": "./train_tools/data_utils/mapping_labeled.json",
+            "mapping_file_tuning": "/home/gihun/MEDIAR/train_tools/data_utils/mapping_tuning.json",
+            "amplified": true,
+            "batch_size": 8, 
+            "valid_portion": 0.0
+        },
+        "public":{
+            "enabled": false,
+            "params":{
+                "root": "/home/gihun/MEDIAR/",
+                "mapping_file": "./train_tools/data_utils/mapping_public.json",
+                "batch_size": 1
+            }
+        },
+        "unlabeled":{
+            "enabled": false
+        }
+    },
+    "train_setups":{
+        "model":{
+            "name": "mediar-former",
+            "params": {},
+            "pretrained":{
+                "enabled": true,
+                "weights": "./weights/pretrained/phase1.pth",
+                "strict": false
+            }
+        },
+        "trainer": {
+            "name": "mediar",
+            "params": {            
+                "num_epochs": 200,
+                "valid_frequency": 1,
+                "device": "cuda:7",
+                "amp": true,
+                "algo_params": {"with_public": false}
+            }
+        },
+        "optimizer":{
+            "name": "adamw",
+            "params": {"lr": 2e-5}
+        },
+        "scheduler":{
+            "enabled": true,
+            "name": "cosine",
+            "params": {"T_max": 100, "eta_min": 1e-7}
+        },
+        "seed": 19940817
+    },
+    "pred_setups":{
+        "input_path":"/home/gihun/MEDIAR/data/Official/Tuning/images",
+        "output_path": "./results/",
+        "make_submission": true,
+        "exp_name": "mediar_from_phase1",
+        "algo_params": {"use_tta": false}
+    },
+    "wandb_setups":{
+        "project": "CellSeg",
+        "group": "Fine-tuning",
+        "name": "from_phase1"
+    }
+}

spacr/resources/MEDIAR/config/step2_finetuning/finetuning2.json

@@ -0,0 +1,66 @@
+{
+    "data_setups":{
+        "labeled":{
+            "root": "/home/gihun/MEDIAR/",
+            "mapping_file": "./train_tools/data_utils/mapping_labeled.json",
+            "mapping_file_tuning": "/home/gihun/MEDIAR/train_tools/data_utils/mapping_tuning.json",
+            "amplified": true,
+            "batch_size": 8, 
+            "valid_portion": 0.0
+        },
+        "public":{
+            "enabled": true,
+            "params":{
+                "root": "/home/gihun/MEDIAR/",
+                "mapping_file": "./train_tools/data_utils/mapping_public.json",
+                "batch_size": 1
+            }
+        },
+        "unlabeled":{
+            "enabled": false
+        }     
+    },
+    "train_setups":{
+        "model":{
+            "name": "mediar-former",
+            "params": {},
+            "pretrained":{
+                "enabled": true,
+                "weights": "./weights/pretrained/phase2.pth",
+                "strict": false
+            }
+        },
+        "trainer": {
+            "name": "mediar",
+            "params": {            
+                "num_epochs": 50,
+                "valid_frequency": 1,
+                "device": "cuda:0",
+                "amp": true,
+                "algo_params": {"with_public": true}
+            }
+        },
+        "optimizer":{
+            "name": "adamw",
+            "params": {"lr": 2e-5}
+        },
+        "scheduler":{
+            "enabled": true,
+            "name": "cosine",
+            "params": {"T_max": 100, "eta_min": 1e-7}
+        },
+        "seed": 19940817
+    },
+    "pred_setups":{
+        "input_path":"/home/gihun/MEDIAR/data/Official/Tuning/images",
+        "output_path": "./results/from_phase2",
+        "make_submission": true,
+        "exp_name": "mediar_from_phase2",
+        "algo_params": {"use_tta": false}
+    },
+    "wandb_setups":{
+        "project": "CellSeg",
+        "group": "Fine-tuning",
+        "name": "from_phase2"
+    }
+}

spacr/resources/MEDIAR/config/step3_prediction/base_prediction.json

@@ -0,0 +1,16 @@
+{
+    "pred_setups":{
+        "name": "mediar",
+        "input_path":"/home/gihun/MEDIAR/data/Official/Tuning/images",
+        "output_path": "./results/mediar_base_prediction",
+        "make_submission": true,
+        "model_path": "./weights/finetuned/from_phase1.pth",
+        "device": "cuda:7",
+        "model":{
+            "name": "mediar-former",
+            "params": {}
+        },
+        "exp_name": "mediar_p1_base",
+        "algo_params": {"use_tta": false}
+    }
+}

spacr/resources/MEDIAR/config/step3_prediction/ensemble_tta.json

@@ -0,0 +1,23 @@
+{
+    "pred_setups":{
+        "name": "ensemble_mediar",
+        "input_path":"/home/gihun/MEDIAR/data/Official/Tuning/images",
+        "output_path": "./results/mediar_ensemble_tta",
+        "make_submission": true,
+        "model_path1": "./weights/finetuned/from_phase1.pth",
+        "model_path2": "./weights/finetuned/from_phase2.pth",
+        "device": "cuda:0",
+        "model":{
+            "name": "mediar-former",
+            "params": {
+                    "encoder_name":"mit_b5",        
+                    "decoder_channels": [1024, 512, 256, 128, 64],
+                    "decoder_pab_channels": 256,
+                    "in_channels":3, 
+                    "classes":3
+            }
+        },
+        "exp_name": "mediar_ensemble_tta",
+        "algo_params": {"use_tta": true}
+    }
+}

spacr/resources/MEDIAR/core/BasePredictor.py

@@ -0,0 +1,120 @@
+import torch
+import numpy as np
+import time, os
+import tifffile as tif
+
+from datetime import datetime
+from zipfile import ZipFile
+from pytz import timezone
+
+from train_tools.data_utils.transforms import get_pred_transforms
+
+
+class BasePredictor:
+    def __init__(
+        self,
+        model,
+        device,
+        input_path,
+        output_path,
+        make_submission=False,
+        exp_name=None,
+        algo_params=None,
+    ):
+        self.model = model
+        self.device = device
+        self.input_path = input_path
+        self.output_path = output_path
+        self.make_submission = make_submission
+        self.exp_name = exp_name
+
+        # Assign algoritm-specific arguments
+        if algo_params:
+            self.__dict__.update((k, v) for k, v in algo_params.items())
+
+        # Prepare inference environments
+        self._setups()
+
+    @torch.no_grad()
+    def conduct_prediction(self):
+        self.model.to(self.device)
+        self.model.eval()
+        total_time = 0
+        total_times = []
+
+        for img_name in self.img_names:
+            img_data = self._get_img_data(img_name)
+            img_data = img_data.to(self.device)
+
+            start = time.time()
+
+            pred_mask = self._inference(img_data)
+            pred_mask = self._post_process(pred_mask.squeeze(0).cpu().numpy())
+            self.write_pred_mask(
+                pred_mask, self.output_path, img_name, self.make_submission
+            )
+
+            end = time.time()
+
+            time_cost = end - start
+            total_times.append(time_cost)
+            total_time += time_cost
+            print(
+                f"Prediction finished: {img_name}; img size = {img_data.shape}; costing: {time_cost:.2f}s"
+            )
+
+        print(f"\n Total Time Cost: {total_time:.2f}s")
+
+        if self.make_submission:
+            fname = "%s.zip" % self.exp_name
+
+            os.makedirs("./submissions", exist_ok=True)
+            submission_path = os.path.join("./submissions", fname)
+
+            with ZipFile(submission_path, "w") as zipObj2:
+                pred_names = sorted(os.listdir(self.output_path))
+                for pred_name in pred_names:
+                    pred_path = os.path.join(self.output_path, pred_name)
+                    zipObj2.write(pred_path)
+
+            print("\n>>>>> Submission file is saved at: %s\n" % submission_path)
+
+        return time_cost
+
+    def write_pred_mask(self, pred_mask, output_dir, image_name, submission=False):
+
+        # All images should contain at least 5 cells
+        if submission:
+            if not (np.max(pred_mask) > 5):
+                print("[!Caution] Only %d Cells Detected!!!\n" % np.max(pred_mask))
+
+        file_name = image_name.split(".")[0]
+        file_name = file_name + "_label.tiff"
+        file_path = os.path.join(output_dir, file_name)
+
+        tif.imwrite(file_path, pred_mask, compression="zlib")
+
+    def _setups(self):
+        self.pred_transforms = get_pred_transforms()
+        os.makedirs(self.output_path, exist_ok=True)
+
+        now = datetime.now(timezone("Asia/Seoul"))
+        dt_string = now.strftime("%m%d_%H%M")
+        self.exp_name = (
+            self.exp_name + dt_string if self.exp_name is not None else dt_string
+        )
+
+        self.img_names = sorted(os.listdir(self.input_path))
+
+    def _get_img_data(self, img_name):
+        img_path = os.path.join(self.input_path, img_name)
+        img_data = self.pred_transforms(img_path)
+        img_data = img_data.unsqueeze(0)
+
+        return img_data
+
+    def _inference(self, img_data):
+        raise NotImplementedError
+
+    def _post_process(self, pred_mask):
+        raise NotImplementedError