Bugfix-01 (#5)

- Updated Zenodo metadata
- Fixed deprecated `np.int`
- Resolved dependencies
- Added analysis section in inference notebook
- Added hyper-parameter tuning in inference notebook
- Accept multiple input formats

.zenodo.json

@@ -0,0 +1,61 @@
+{
+  "upload_type": "software",
+  "title": "DT_SegNet",
+  "creators": [
+    {
+      "name": "Xia, Zeyu",
+      "affiliation": "Queensland University of Technology",
+      "orcid": "0000-0003-0234-5857"
+    },
+    {
+      "name": "Ma, Kan",
+      "affiliation": "University of Birmingham",
+      "orcid": "0000-0001-5729-5477"
+    },
+    {
+      "name": "Cheng, Sibo",
+      "affiliation": "Imperial College London",
+      "orcid": "0000-0002-8707-2589"
+    },
+    {
+      "name": "Blackburn, Thomas",
+      "affiliation": "University of Birmingham",
+      "orcid": "0000-0001-9160-3285"
+    },
+    {
+      "name": "Peng, Ziling",
+      "affiliation": "Institute of Advanced Science Facilities"
+    },
+    {
+      "name": "Zhu, Kewei",
+      "affiliation": "University of York"
+    },
+    {
+      "name": "Zhang, Weihang",
+      "affiliation": "Imperial College London"
+    },
+    {
+      "name": "Xiao, Dunhui",
+      "affiliation": "Tongji University"
+    },
+    {
+      "name": "Knowles, Alexander J",
+      "affiliation": "University of Birmingham"
+    },
+    {
+      "name": "Arcucci, Rossella",
+      "affiliation": "Imperial College London"
+    }
+  ],
+  "description": "A comprehensive, two-tiered deep learning approach designed for precise object detection and segmentation in electron microscopy (EM) images.",
+  "access_right": "open",
+  "license": "mit",
+  "related_identifiers": [
+    {
+      "scheme": "doi",
+      "identifier": "10.1039/D3CP00402C",
+      "relation": "isSupplementTo",
+      "resource_type": "article"
+    }
+  ]
+}

0_Labelling_Tools/0_EISeg/eiseg/inference/predictor/base.py

@@ -243,7 +243,7 @@ def split_points_by_order(tpoints, groups):
             (bs, 2 * x, 3), -1, dtype=np.float32) for x in groups
     ]
 
-    last_point_indx_group = np.zeros((bs, num_groups, 2), dtype=np.int)
+    last_point_indx_group = np.zeros((bs, num_groups, 2), dtype=np.int32)
     for group_indx, group_size in enumerate(groups):
         last_point_indx_group[:, group_indx, 1] = group_size
 

1_Detection_Model/requirements.txt

@@ -11,8 +11,8 @@ requests>=2.23.0
 scipy>=1.4.1
 torch>=1.7.0
 torchvision>=0.8.1
-tqdm
-protobuf<=3.20.1  # https://github.com/ultralytics/yolov5/issues/8012
+tqdm>=4.64.0
+# protobuf<=3.20.1  # https://github.com/ultralytics/yolov5/issues/8012
 
 # Logging -------------------------------------
 tensorboard>=2.4.1

1_Detection_Model/utils/dataloaders.py

@@ -484,7 +484,7 @@ def __init__(self,
         self.im_files = list(cache.keys())  # update
         self.label_files = img2label_paths(cache.keys())  # update
         n = len(shapes)  # number of images
-        bi = np.floor(np.arange(n) / batch_size).astype(np.int)  # batch index
+        bi = np.floor(np.arange(n) / batch_size).astype(np.int32)  # batch index
         nb = bi[-1] + 1  # number of batches
         self.batch = bi  # batch index of image
         self.n = n
@@ -526,7 +526,7 @@ def __init__(self,
                 elif mini > 1:
                     shapes[i] = [1, 1 / mini]
 
-            self.batch_shapes = np.ceil(np.array(shapes) * img_size / stride + pad).astype(np.int) * stride
+            self.batch_shapes = np.ceil(np.array(shapes) * img_size / stride + pad).astype(np.int32) * stride
 
         # Cache images into RAM/disk for faster training (WARNING: large datasets may exceed system resources)
         self.ims = [None] * n
@@ -896,7 +896,7 @@ def extract_boxes(path=DATASETS_DIR / 'coco128'):  # from utils.dataloaders impo
                     b = x[1:] * [w, h, w, h]  # box
                     # b[2:] = b[2:].max()  # rectangle to square
                     b[2:] = b[2:] * 1.2 + 3  # pad
-                    b = xywh2xyxy(b.reshape(-1, 4)).ravel().astype(np.int)
+                    b = xywh2xyxy(b.reshape(-1, 4)).ravel().astype(np.int32)
 
                     b[[0, 2]] = np.clip(b[[0, 2]], 0, w)  # clip boxes outside of image
                     b[[1, 3]] = np.clip(b[[1, 3]], 0, h)

3_Segmentation_Model/requirements.txt

@@ -5,4 +5,4 @@ tqdm
 filelock
 scipy
 prettytable
-sklearn == 0.0
+scikit-learn