wip: slice class and segmentation mask

Author: PJEstrada

pytorch_test.py

@@ -0,0 +1,33 @@
+import diffgram
+from diffgram.pytorch_diffgram.diffgram_pytorch_dataset import DiffgramPytorchDataset
+
+project = diffgram.Project(project_string_id = "voc-test",
+                  client_id = "LIVE__p0blrrm6p5fnan5sh8ec",
+                  client_secret = "d14sl5vtg672ms8rg97yp1vc9do1ao3ee2xlzktk29kbk49t8mklpt7bvnmh",
+                  debug = True)
+
+file = project.file.get_by_id(1554, with_instances = True)
+
+diffgram_dataset = DiffgramPytorchDataset(
+    project = project,
+    diffgram_file_id_list = [1554]
+)
+
+
+
+
+
+# Draw
+import matplotlib.pyplot as plt
+from PIL import Image, ImageDraw
+img = Image.new("L", [diffgram_dataset[0]['diffgram_file'].image['width'], diffgram_dataset[0]['diffgram_file'].image['height']], 0)
+mask1 = diffgram_dataset[0]['polygon_mask_list'][0]
+mask2 = diffgram_dataset[0]['polygon_mask_list'][1]
+print(mask1)
+for x in mask1:
+    print(x)
+plt.figure()
+plt.subplot(1,2,1)
+# plt.imshow(img, 'gray', interpolation='none')
+plt.imshow(mask1, 'jet', interpolation='none', alpha=0.7)
+plt.imshow(mask2, 'Oranges', interpolation='none', alpha=0.7)

sdk/diffgram/core/directory.py

@@ -92,13 +92,35 @@ def all_files(self):
 			result = result + diffgram_files
 		return result
 
+	def all_file_ids(self):
+		page_num = 1
+		result = []
+		while page_num is not None:
+			diffgram_files = self.list_files(limit = 1000, page_num = page_num, file_view_mode = 'ids_only')
+			page_num = self.file_list_metadata['next_page']
+			result = result + diffgram_files
+		return result
+
+	def slice(self, query):
+		from diffgram.core.sliced_directory import SlicedDirectory
+		result = self.list_files(
+			limit = 25,
+			page_num = 1,
+			file_view_mode = 'ids_only'
+		)
+		sliced_dataset = SlicedDirectory(
+			query = query,
+			original_directory = self
+		)
+		return sliced_dataset
+
 	def to_pytorch(self, transform = None):
 		"""
 			Transforms the file list inside the dataset into a pytorch dataset.
 		:return:
 		"""
-		dataset_files = self.all_files()
-		file_id_list = [file.id for file in dataset_files]
+		from diffgram.core.sliced_directory import SlicedDirectory
+		file_id_list = self.all_file_ids()
 		pytorch_dataset = DiffgramPytorchDataset(
 			project = self.client,
 			diffgram_file_id_list = file_id_list,
@@ -162,7 +184,8 @@ def list_files(
 			page_num=1,
 			limit=100,
 			search_term: str =None,
-			file_view_mode: str = 'annotation'):
+			file_view_mode: str = 'annotation',
+			query: str = None):
 		"""
 		Get a list of files in directory (from Diffgram service). 
 	

sdk/diffgram/core/sliced_directory.py

@@ -0,0 +1,37 @@
+from diffgram.core.directory import Directory
+from diffgram.pytorch_diffgram.diffgram_pytorch_dataset import DiffgramPytorchDataset
+
+class SlicedDirectory(Directory):
+
+    def __init__(self, client, original_directory: Directory, query: str):
+        self.original_directory = original_directory
+        self.query = query
+        self.client = client
+
+    def all_file_ids(self):
+        page_num = 1
+        result = []
+        while page_num is not None:
+            diffgram_files = self.list_files(limit = 1000,
+                                             page_num = page_num,
+                                             file_view_mode = 'ids_only',
+                                             query = self.query)
+            page_num = self.file_list_metadata['next_page']
+            result = result + diffgram_files
+        return result
+
+
+    def to_pytorch(self, transform = None):
+        """
+            Transforms the file list inside the dataset into a pytorch dataset.
+        :return:
+        """
+        file_id_list = self.all_file_ids()
+        pytorch_dataset = DiffgramPytorchDataset(
+            project = self.client,
+            diffgram_file_id_list = file_id_list,
+            transform = transform
+
+        )
+        return pytorch_dataset
+

sdk/diffgram/file/file_constructor.py

@@ -383,7 +383,7 @@ def format_assumptions(
 
         return instance_list
 
-    def import_bulk:
+    def import_bulk(self):
         """
         Import multiple packets
         FUTURE
@@ -392,7 +392,7 @@ def import_bulk:
             { packet_id : { packet }}
 
         """
-        pass
+        raise NotImplementedError
 
     def get_file_list(self, id_list: list, with_instances: bool = False):
         """

sdk/diffgram/pytorch_diffgram/__pycache__/__init__.cpython-38.pyc

@@ -3,6 +3,8 @@
 import os
 from imageio import imread
 import numpy as np
+import scipy as sp
+from PIL import Image, ImageDraw
 
 
 class DiffgramPytorchDataset(Dataset):
@@ -15,17 +17,37 @@ def __init__(self, project, diffgram_file_id_list = None, transform = None):
         :param transform (callable, optional): Optional transforms to be applied on a sample
         """
         self.diffgram_file_id_list = diffgram_file_id_list
-        self.__validate_file_ids()
+
         self.project = project
         self.transform = transform
         self._internal_file_list = []
-
+        self.__validate_file_ids()
 
     def __validate_file_ids(self):
-        url = '/api/'
-        raise NotImplementedError
-
-    def __extract_bbox_values(self, instance_list):
+        result = self.project.file.file_list_exists(self.diffgram_file_id_list)
+        if not result:
+            raise Exception(
+                'Some file IDs do not belong to the project. Please provide only files from the same project.')
+
+    def __extract_masks_from_polygon(self, instance_list, diffgram_file, empty_value = 0):
+        nx, ny = diffgram_file.image['width'], diffgram_file.image['height']
+        mask_list = []
+        for instance in instance_list:
+            if instance['type'] != 'polygon':
+                continue
+            poly = [(p['x'], p['y']) for p in instance['points']]
+
+            img = Image.new(mode = 'L', size = (nx, ny), color = 0)  # mode L = 8-bit pixels, black and white
+            draw = ImageDraw.Draw(img)
+            print()
+            draw.polygon(poly, outline = 1, fill = 1)
+            mask = np.array(img).astype('float32')
+            # mask[np.where(mask == 0)] = empty_value
+            print('mask', len(mask))
+            mask_list.append(mask)
+        return mask_list
+
+    def __extract_bbox_values(self, instance_list, diffgram_file):
         """
             Creates a pytorch tensor based on the instance type.
             For now we are assuming shapes here, but we can extend it
@@ -41,10 +63,10 @@ def __extract_bbox_values(self, instance_list):
         for inst in instance_list:
             if inst['type'] != 'box':
                 continue
-            x_min_list.append(inst['x_min'])
-            x_max_list.append(inst['x_max'])
-            y_min_list.append(inst['y_min'])
-            y_max_list.append(inst['y_max'])
+            x_min_list.append(inst['x_min'] / diffgram_file.image['width'])
+            x_max_list.append(inst['x_max'] / diffgram_file.image['width'])
+            y_min_list.append(inst['y_min'] / diffgram_file.image['width'])
+            y_max_list.append(inst['y_max'] / diffgram_file.image['width'])
 
         return x_min_list, x_max_list, y_min_list, y_max_list
 
@@ -58,7 +80,7 @@ def __getitem__(self, idx):
         if torch.is_tensor(idx):
             idx = idx.tolist()
 
-        diffgram_file = self.project.file.get_by_id(idx, with_instances = True)
+        diffgram_file = self.project.file.get_by_id(self.diffgram_file_id_list[idx], with_instances = True)
         if hasattr(diffgram_file, 'image'):
             image = imread(diffgram_file.image.get('url_signed'))
         else:
@@ -68,17 +90,17 @@ def __getitem__(self, idx):
         instance_types_in_file = set([x['type'] for x in instance_list])
         # Process the instances of each file
         processed_instance_list = []
-
-        sample = {'image': image}
+        sample = {'image': image, 'diffgram_file': diffgram_file}
         if 'box' in instance_types_in_file:
-            x_min_list, x_max_list, y_min_list, y_max_list = self.__extract_bbox_values(instance_list)
+            x_min_list, x_max_list, y_min_list, y_max_list = self.__extract_bbox_values(instance_list, diffgram_file)
             sample['x_min_list'] = torch.Tensor(x_min_list)
             sample['x_max_list'] = torch.Tensor(x_max_list)
             sample['y_min_list'] = torch.Tensor(y_min_list)
             sample['y_max_list'] = torch.Tensor(y_max_list)
         if 'polygon' in instance_types_in_file:
-
+            mask_list = self.__extract_masks_from_polygon(instance_list, diffgram_file)
+            sample['polygon_mask_list'] = mask_list
         if self.transform:
             sample = self.transform(sample)
 
-        return sample
+        return sample