labelox integration perforamce improvement and fixes

deeplake/integrations/labelbox/labelbox_converter.py

@@ -15,6 +15,7 @@ def __init__(
         self.labelbox_feature_id_to_type_mapping = dict()
         self.regsistered_actions = dict()
         self.label_mappings = dict()
+        self.values_cache = dict()
 
         self.project = project
         self.project_id = project_id
@@ -44,6 +45,7 @@ def dataset_with_applied_annotations(self):
                 continue
             print('parsing annotations for project with index: ', p_idx)
             for lbl_idx, labels in enumerate(p["projects"][self.project_id]["labels"]):
+                self.values_cache = dict()
                 if "frames" not in labels["annotations"]:
                     continue
                 frames = labels["annotations"]["frames"]
@@ -72,6 +74,8 @@ def dataset_with_applied_annotations(self):
                 # iterate over segments and assign same value to all frames in the segment
                 self.parse_segments_(segments, frames, idx_offset)
 
+                self.apply_cached_values_(self.values_cache)
+
             idx_offset += p["media_attributes"]["frame_count"]
 
         return self.dataset
@@ -214,6 +218,20 @@ def parse_segments_(self, segments, frames, offset):
                         offset + i - 1, obj
                     )
 
+    def apply_cached_values_(self, cache):
+        print('applying cached values')
+        for tensor_name, row_map in cache.items():
+            print('applying cached values for tensor: ', tensor_name)
+            max_val = max(row_map.keys())
+            values = []
+            for i in tqdm.tqdm(range(max_val + 1)):
+                if i in row_map:
+                    values.append(row_map[i])
+                else:
+                    values.append(None)
+
+            self.dataset[tensor_name].extend(values)
+
     def yield_projects_(self, project_j, ds):
         raise NotImplementedError("fixed_project_order_ is not implemented")
 

deeplake/integrations/labelbox/v3_converters.py

@@ -21,13 +21,12 @@ def bbox_converter_(obj, converter, tensor_name, context, generate_labels):
     converter.register_feature_id_for_kind("tool", "bounding_box", obj, tensor_name)
 
     def bbox_converter(row, obj):
-        vals = []
-        try:
-            vals = ds[tensor_name][row].numpy(aslist=True).tolist()
-        except (KeyError, IndexError):
-            pass
+        if tensor_name not in converter.values_cache:
+            converter.values_cache[tensor_name] = dict()
+        if row not in converter.values_cache[tensor_name]:
+            converter.values_cache[tensor_name][row] = []
 
-        vals.append(
+        converter.values_cache[tensor_name][row].append(
             [
                 int(v)
                 for v in [
@@ -38,8 +37,6 @@ def bbox_converter(row, obj):
                 ]
             ]
         )
-        ds[tensor_name][row] = vals
-
     converter.regsistered_actions[obj.feature_schema_id] = bbox_converter
 
 
@@ -68,7 +65,11 @@ def radio_converter_(obj, converter, tensor_name, context, generate_labels):
     )
 
     def radio_converter(row, o):
-        ds[tensor_name][row] = converter.label_mappings[tensor_name][o["value"]]
+        if tensor_name not in converter.values_cache:
+            converter.values_cache[tensor_name] = dict()
+        if row not in converter.values_cache[tensor_name]:
+            converter.values_cache[tensor_name][row] = []
+        converter.values_cache[tensor_name][row] = [converter.label_mappings[tensor_name][o["value"]]]
 
     for option in obj.options:
         converter.regsistered_actions[option.feature_schema_id] = radio_converter
@@ -104,14 +105,12 @@ def checkbox_converter_(obj, converter, tensor_name, context, generate_labels):
     )
 
     def checkbox_converter(row, obj):
-        vals = []
-        try:
-            vals = ds[tensor_name][row].numpy(aslist=True).tolist()
-        except (KeyError, IndexError):
-            pass
-        vals.append(converter.label_mappings[tensor_name][obj["value"]])
+        if tensor_name not in converter.values_cache:
+            converter.values_cache[tensor_name] = dict()
+        if row not in converter.values_cache[tensor_name]:
+            converter.values_cache[tensor_name][row] = []
 
-        ds[tensor_name][row] = vals
+        converter.values_cache[tensor_name][row].append(converter.label_mappings[tensor_name][obj["value"]])
 
     for option in obj.options:
         converter.regsistered_actions[option.feature_schema_id] = checkbox_converter
@@ -136,13 +135,12 @@ def point_converter_(obj, converter, tensor_name, context, generate_labels):
         print("point converter does not support generating labels")
 
     def point_converter(row, obj):
-        vals = []
-        try:
-            vals = ds[tensor_name][row].numpy(aslist=True).tolist()
-        except (KeyError, IndexError):
-            pass
-        vals.append([int(obj["point"]["x"]), int(obj["point"]["y"])])
-        ds[tensor_name][row] = vals
+        if tensor_name not in converter.values_cache:
+            converter.values_cache[tensor_name] = dict()
+        if row not in converter.values_cache[tensor_name]:
+            converter.values_cache[tensor_name][row] = []
+
+        converter.values_cache[tensor_name][row].append([int(obj["point"]["x"]), int(obj["point"]["y"])])
 
     converter.regsistered_actions[obj.feature_schema_id] = point_converter
 
@@ -160,13 +158,12 @@ def line_converter_(obj, converter, tensor_name, context, generate_labels):
         print("line converter does not support generating labels")
 
     def polygon_converter(row, obj):
-        vals = []
-        try:
-            vals = ds[tensor_name][row].numpy(aslist=True)
-        except (KeyError, IndexError):
-            pass
-        vals.append([[int(l["x"]), int(l["y"])] for l in obj["line"]])
-        ds[tensor_name][row] = vals
+        if tensor_name not in converter.values_cache:
+            converter.values_cache[tensor_name] = dict()
+        if row not in converter.values_cache[tensor_name]:
+            converter.values_cache[tensor_name][row] = []
+
+        converter.values_cache[tensor_name][row].append([[int(l["x"]), int(l["y"])] for l in obj["line"]])
 
     converter.regsistered_actions[obj.feature_schema_id] = polygon_converter
 
@@ -230,22 +227,22 @@ def mask_converter(row, obj):
                     )
                     ds[f"{tensor_name}_labels"][row] = val
 
-                    mask = np.array(Image.open(response)).astype(np.bool_)
-                    mask = mask[..., np.newaxis]
-                    try:
-                        if generate_labels:
-                            arr = ds[tensor_name][row].numpy()
-                            labels = ds[f"{tensor_name}_labels"].info['class_names']
-                            if labels != arr.shape[-1]:
-                                val = np.concatenate([ds[tensor_name][row].numpy(), np.zeros_like(mask)], axis=-1)
-                            idx = labels.index(tool_name)
-                            val[:,:,idx] = np.logical_or(val[:,:,idx], mask[:,:,0])
-                        else:
-                            val = np.logical_or(ds[tensor_name][row].numpy(), mask)
-                    except (KeyError, IndexError):
-                        val = mask
-
-                    ds[tensor_name][row] = val
+                mask = np.array(Image.open(response)).astype(np.bool_)
+                mask = mask[..., np.newaxis]
+                try:
+                    if generate_labels:
+                        arr = ds[tensor_name][row].numpy()
+                        labels = ds[f"{tensor_name}_labels"].info['class_names']
+                        if len(labels) != arr.shape[-1]:
+                            val = np.concatenate([ds[tensor_name][row].numpy(), np.zeros_like(mask)], axis=-1)
+                        idx = labels.index(tool_name)
+                        val[:,:,idx] = np.logical_or(val[:,:,idx], mask[:,:,0])
+                    else:
+                        val = np.logical_or(ds[tensor_name][row].numpy(), mask)
+                except (KeyError, IndexError):
+                    val = mask
+
+                ds[tensor_name][row] = val
         except Exception as e:
             print(f"Error downloading mask: {e}")
 
@@ -265,6 +262,10 @@ def text_converter_(obj, converter, tensor_name, context, generate_labels):
         print("text converter does not support generating labels")
 
     def text_converter(row, obj):
-        ds[tensor_name][row] = obj["text_answer"]["content"]
+        if tensor_name not in converter.values_cache:
+            converter.values_cache[tensor_name] = dict()
+        if row not in converter.values_cache[tensor_name]:
+            converter.values_cache[tensor_name][row] = []
+        converter.values_cache[tensor_name][row] = obj["text_answer"]["content"]
 
     converter.regsistered_actions[obj.feature_schema_id] = text_converter