p85971406
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!2184 CocoDataset add category_id check 

Merge pull request !2184 from xiefangqi/xfq_fix_coco_issue
This commit is contained in:
mindspore-ci-bot 2020-06-17 10:06:41 +08:00 committed by Gitee
parent 8711ab5028 aab190fcdb
commit ca94916ef9
5 changed files with 31 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
mindspore/ccsrc/dataset/engine/datasetops/source/coco_op.cc
View File

@ -407,6 +407,11 @@ Status CocoOp::ParseAnnotationIds() {
nlohmann::json image_list;
RETURN_IF_NOT_OK(SearchNodeInJson(js, std::string(kJsonImages), &image_list));
RETURN_IF_NOT_OK(ImageColumnLoad(image_list, &image_que));
if (task_type_ == TaskType::Detection || task_type_ == TaskType::Panoptic) {
nlohmann::json node_categories;
RETURN_IF_NOT_OK(SearchNodeInJson(js, std::string(kJsonCategories), &node_categories));
RETURN_IF_NOT_OK(CategoriesColumnLoad(node_categories));
}
nlohmann::json annotations_list;
RETURN_IF_NOT_OK(SearchNodeInJson(js, std::string(kJsonAnnotations), &annotations_list));
for (auto annotation : annotations_list) {
@ -440,11 +445,6 @@ Status CocoOp::ParseAnnotationIds() {
for (auto img : image_que) {
if (coordinate_map_.find(img) != coordinate_map_.end()) image_ids_.push_back(img);
}
if (task_type_ == TaskType::Detection || task_type_ == TaskType::Panoptic) {
nlohmann::json node_categories;
RETURN_IF_NOT_OK(SearchNodeInJson(js, std::string(kJsonCategories), &node_categories));
RETURN_IF_NOT_OK(CategoriesColumnLoad(node_categories));
}
num_rows_ = image_ids_.size();
return Status::OK();
}
@ -472,6 +472,9 @@ Status CocoOp::DetectionColumnLoad(nlohmann::json annotation_tree, const std::st
uint32_t category_id = 0, iscrowd = 0;
RETURN_IF_NOT_OK(SearchNodeInJson(annotation_tree, std::string(kJsonAnnoBbox), &node_bbox));
RETURN_IF_NOT_OK(SearchNodeInJson(annotation_tree, std::string(kJsonAnnoCategoryId), &category_id));
auto search_category = category_set_.find(category_id);
if (search_category == category_set_.end())
RETURN_STATUS_UNEXPECTED("category_id can't find in categories where category_id: " + std::to_string(category_id));
auto node_iscrowd = annotation_tree.find(kJsonAnnoIscrowd);
if (node_iscrowd != annotation_tree.end()) iscrowd = *node_iscrowd;
bbox.insert(bbox.end(), node_bbox.begin(), node_bbox.end());
@ -524,22 +527,25 @@ Status CocoOp::PanopticColumnLoad(nlohmann::json annotation_tree, const std::str
RETURN_STATUS_UNEXPECTED("No segments_info found in annotations where image_id: " + std::to_string(image_id));
for (auto info : *itr_segments) {
std::vector<float> bbox;
uint32_t category_id = 0;
auto itr_bbox = info.find(kJsonAnnoBbox);
if (itr_bbox == info.end())
RETURN_STATUS_UNEXPECTED("No bbox found in segments_info where image_id: " + std::to_string(image_id));
bbox.insert(bbox.end(), itr_bbox->begin(), itr_bbox->end());
coordinate_map_[image_file].push_back(bbox);
auto itr_category_id = info.find(kJsonAnnoCategoryId);
if (itr_category_id == info.end())
RETURN_STATUS_UNEXPECTED("No category_id found in segments_info where image_id: " + std::to_string(image_id));
RETURN_IF_NOT_OK(SearchNodeInJson(info, std::string(kJsonAnnoCategoryId), &category_id));
auto search_category = category_set_.find(category_id);
if (search_category == category_set_.end())
RETURN_STATUS_UNEXPECTED("category_id can't find in categories where category_id: " +
std::to_string(category_id));
auto itr_iscrowd = info.find(kJsonAnnoIscrowd);
if (itr_iscrowd == info.end())
RETURN_STATUS_UNEXPECTED("No iscrowd found in segments_info where image_id: " + std::to_string(image_id));
auto itr_area = info.find(kJsonAnnoArea);
if (itr_area == info.end())
RETURN_STATUS_UNEXPECTED("No area found in segments_info where image_id: " + std::to_string(image_id));
simple_item_map_[image_file].push_back(*itr_category_id);
simple_item_map_[image_file].push_back(category_id);
simple_item_map_[image_file].push_back(*itr_iscrowd);
simple_item_map_[image_file].push_back(*itr_area);
}
@ -556,6 +562,7 @@ Status CocoOp::CategoriesColumnLoad(nlohmann::json categories_tree) {
if (itr_id == category.end()) RETURN_STATUS_UNEXPECTED("No id found in categories of " + annotation_path_);
id = *itr_id;
label_info.push_back(id);
category_set_.insert(id);
auto itr_name = category.find(kJsonCategoriesName);
if (itr_name == category.end())

2
mindspore/ccsrc/dataset/engine/datasetops/source/coco_op.h
View File

@ -18,6 +18,7 @@
#include <map>
#include <memory>
#include <set>
#include <string>
#include <utility>
#include <vector>
@ -322,6 +323,7 @@ class CocoOp : public ParallelOp, public RandomAccessOp {
std::vector<std::pair<std::string, std::vector<int32_t>>> label_index_;
std::map<std::string, CoordinateRow> coordinate_map_;
std::map<std::string, std::vector<uint32_t>> simple_item_map_;
std::set<uint32_t> category_set_;
};
} // namespace dataset
} // namespace mindspore

1
tests/ut/data/dataset/testCOCO/annotations/invalid_category_id.json Normal file
View File

@ -0,0 +1 @@
{"info": {"description": "COCO 2017 Dataset", "url": "http://cocodataset.org","version": "1.0","year": 2017,"contributor": "COCO Consortium", "data_created": "2017/09/01"}, "licenses": [{"url": "http://creativecommons.org/license/by-nc-sa/2.0/","id": 3,"name": "Attribution-Noncommercial License"}], "images": [{"license": 3, "file_name": "000000391895.jpg", "id": 391895}, {"license": 3, "file_name": "000000318219.jpg", "id": 318219}, {"license": 3, "file_name": "000000554625.jpg", "id": 554625}, {"license": 3, "file_name": "000000574769.jpg", "id": 574769}, {"license": 3, "file_name": "000000060623.jpg", "id": 60623}, {"license": 3, "file_name": "000000309022.jpg", "id": 309022}], "annotations": [{"segmentation": [[10.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0,20.0]], "category_id": 1, "iscrowd": 0, "image_id": 391895, "bbox": [10,10,10,10], "area": 100, "id": 10000}, {"segmentation": [[20.0,22.0,23.0,24.0,25.0,26.0,27.0,28.0,29.0,30.0,31.0]], "category_id": 2, "iscrowd": 0, "image_id": 318219, "bbox": [20,20,20,20], "area": 400, "id": 10001}, {"segmentation": [[40.0,42.0,43.0,44.0,45.0,46.0,47.0,48.0,49.0,40.0,41.0,42.0]], "category_id": 3, "iscrowd": 0, "image_id": 554625, "bbox": [30,30,30,30], "area": 900, "id": 10002}, {"segmentation": [[50.0,52.0,53.0,54.0,55.0,56.0,57.0,58.0,59.0,60.0,61.0,62.0,63.0]], "category_id": 4, "iscrowd": 0, "image_id": 574769, "bbox": [40,40,40,40], "area": 1600, "id": 10003}, {"segmentation": [[60.0,62.0,63.0,64.0,65.0,66.0,67.0,68.0,69.0,70.0,71.0,72.0,73.0,74.0]], "category_id": 5, "iscrowd": 0, "image_id": 60623, "bbox": [50,50,50,50], "area": 2500, "id": 10004}, {"segmentation": [[60.0,62.0,63.0,64.0,65.0,66.0,67.0],[68.0,69.0,70.0,71.0,72.0,73.0,74.0]], "category_id": 6, "iscrowd": 0, "image_id": 309022, "bbox": [60,60,60,60], "area": 3600, "id": 10005}, {"segmentation": [[70.0,72.0,73.0,74.0,75.0]], "category_id": 7, "iscrowd": 0, "image_id": 391895, "bbox": [70,70,70,70], "area": 4900, "id": 10006}, {"segmentation": {"counts": [10.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0,20.0],"size": [200,300]}, "category_id": 8, "iscrowd": 1, "image_id": 318219, "bbox": [80,80,80,80], "area": 6400, "id": 10007}], "categories": [{"supercategory": "person", "id": 1, "name": "person"},{"supercategory": "vehicle", "id": 2, "name": "bicycle"},{"supercategory": "vehicle", "id": 3, "name": "car"},{"supercategory": "vehicle", "id": 4, "name": "cat"},{"supercategory": "vehicle", "id": 5, "name": "dog"},{"supercategory": "vehicle", "id": 6, "name": "monkey"}]}

2
tests/ut/data/dataset/testCOCO/annotations/train.json
View File

@ -1 +1 @@
{"info": {"description": "COCO 2017 Dataset", "url": "http://cocodataset.org","version": "1.0","year": 2017,"contributor": "COCO Consortium", "data_created": "2017/09/01"}, "licenses": [{"url": "http://creativecommons.org/license/by-nc-sa/2.0/","id": 3,"name": "Attribution-Noncommercial License"}], "images": [{"license": 3, "file_name": "000000391895.jpg", "id": 391895}, {"license": 3, "file_name": "000000318219.jpg", "id": 318219}, {"license": 3, "file_name": "000000554625.jpg", "id": 554625}, {"license": 3, "file_name": "000000574769.jpg", "id": 574769}, {"license": 3, "file_name": "000000060623.jpg", "id": 60623}, {"license": 3, "file_name": "000000309022.jpg", "id": 309022}], "annotations": [{"segmentation": [[10.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0,20.0]], "category_id": 1, "iscrowd": 0, "image_id": 391895, "bbox": [10,10,10,10], "area": 100, "id": 10000}, {"segmentation": [[20.0,22.0,23.0,24.0,25.0,26.0,27.0,28.0,29.0,30.0,31.0]], "category_id": 2, "iscrowd": 0, "image_id": 318219, "bbox": [20,20,20,20], "area": 400, "id": 10001}, {"segmentation": [[40.0,42.0,43.0,44.0,45.0,46.0,47.0,48.0,49.0,40.0,41.0,42.0]], "category_id": 3, "iscrowd": 0, "image_id": 554625, "bbox": [30,30,30,30], "area": 900, "id": 10002}, {"segmentation": [[50.0,52.0,53.0,54.0,55.0,56.0,57.0,58.0,59.0,60.0,61.0,62.0,63.0]], "category_id": 4, "iscrowd": 0, "image_id": 574769, "bbox": [40,40,40,40], "area": 1600, "id": 10003}, {"segmentation": [[60.0,62.0,63.0,64.0,65.0,66.0,67.0,68.0,69.0,70.0,71.0,72.0,73.0,74.0]], "category_id": 5, "iscrowd": 0, "image_id": 60623, "bbox": [50,50,50,50], "area": 2500, "id": 10004}, {"segmentation": [[60.0,62.0,63.0,64.0,65.0,66.0,67.0],[68.0,69.0,70.0,71.0,72.0,73.0,74.0]], "category_id": 6, "iscrowd": 0, "image_id": 309022, "bbox": [60,60,60,60], "area": 3600, "id": 10005}, {"segmentation": [[70.0,72.0,73.0,74.0,75.0]], "category_id": 7, "iscrowd": 0, "image_id": 391895, "bbox": [70,70,70,70], "area": 4900, "id": 10006}, {"segmentation": {"counts": [10.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0,20.0],"size": [200,300]}, "category_id": 8, "iscrowd": 1, "image_id": 318219, "bbox": [80,80,80,80], "area": 6400, "id": 10007}], "categories": [{"supercategory": "person", "id": 1, "name": "person"},{"supercategory": "vehicle", "id": 2, "name": "bicycle"},{"supercategory": "vehicle", "id": 3, "name": "car"},{"supercategory": "vehicle", "id": 4, "name": "cat"},{"supercategory": "vehicle", "id": 5, "name": "dog"},{"supercategory": "vehicle", "id": 6, "name": "monkey"},{"supercategory": "vehicle", "id": 6, "name": "monkey"},{"supercategory": "vehicle", "id": 7, "name": "monkey"}]}
{"info": {"description": "COCO 2017 Dataset", "url": "http://cocodataset.org","version": "1.0","year": 2017,"contributor": "COCO Consortium", "data_created": "2017/09/01"}, "licenses": [{"url": "http://creativecommons.org/license/by-nc-sa/2.0/","id": 3,"name": "Attribution-Noncommercial License"}], "images": [{"license": 3, "file_name": "000000391895.jpg", "id": 391895}, {"license": 3, "file_name": "000000318219.jpg", "id": 318219}, {"license": 3, "file_name": "000000554625.jpg", "id": 554625}, {"license": 3, "file_name": "000000574769.jpg", "id": 574769}, {"license": 3, "file_name": "000000060623.jpg", "id": 60623}, {"license": 3, "file_name": "000000309022.jpg", "id": 309022}], "annotations": [{"segmentation": [[10.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0,20.0]], "category_id": 1, "iscrowd": 0, "image_id": 391895, "bbox": [10,10,10,10], "area": 100, "id": 10000}, {"segmentation": [[20.0,22.0,23.0,24.0,25.0,26.0,27.0,28.0,29.0,30.0,31.0]], "category_id": 2, "iscrowd": 0, "image_id": 318219, "bbox": [20,20,20,20], "area": 400, "id": 10001}, {"segmentation": [[40.0,42.0,43.0,44.0,45.0,46.0,47.0,48.0,49.0,40.0,41.0,42.0]], "category_id": 3, "iscrowd": 0, "image_id": 554625, "bbox": [30,30,30,30], "area": 900, "id": 10002}, {"segmentation": [[50.0,52.0,53.0,54.0,55.0,56.0,57.0,58.0,59.0,60.0,61.0,62.0,63.0]], "category_id": 4, "iscrowd": 0, "image_id": 574769, "bbox": [40,40,40,40], "area": 1600, "id": 10003}, {"segmentation": [[60.0,62.0,63.0,64.0,65.0,66.0,67.0,68.0,69.0,70.0,71.0,72.0,73.0,74.0]], "category_id": 5, "iscrowd": 0, "image_id": 60623, "bbox": [50,50,50,50], "area": 2500, "id": 10004}, {"segmentation": [[60.0,62.0,63.0,64.0,65.0,66.0,67.0],[68.0,69.0,70.0,71.0,72.0,73.0,74.0]], "category_id": 6, "iscrowd": 0, "image_id": 309022, "bbox": [60,60,60,60], "area": 3600, "id": 10005}, {"segmentation": [[70.0,72.0,73.0,74.0,75.0]], "category_id": 7, "iscrowd": 0, "image_id": 391895, "bbox": [70,70,70,70], "area": 4900, "id": 10006}, {"segmentation": {"counts": [10.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0,20.0],"size": [200,300]}, "category_id": 8, "iscrowd": 1, "image_id": 318219, "bbox": [80,80,80,80], "area": 6400, "id": 10007}], "categories": [{"supercategory": "person", "id": 1, "name": "person"},{"supercategory": "vehicle", "id": 2, "name": "bicycle"},{"supercategory": "vehicle", "id": 3, "name": "car"},{"supercategory": "vehicle", "id": 4, "name": "cat"},{"supercategory": "vehicle", "id": 5, "name": "dog"},{"supercategory": "vehicle", "id": 6, "name": "monkey"},{"supercategory": "vehicle", "id": 7, "name": "bag"},{"supercategory": "vehicle", "id": 8, "name": "orange"}]}

12
tests/ut/python/dataset/test_datasets_coco.py
View File

@ -22,6 +22,7 @@ KEYPOINT_FILE = "../data/dataset/testCOCO/annotations/key_point.json"
PANOPTIC_FILE = "../data/dataset/testCOCO/annotations/panoptic.json"
INVALID_FILE = "../data/dataset/testCOCO/annotations/invalid.json"
LACKOFIMAGE_FILE = "../data/dataset/testCOCO/annotations/lack_of_images.json"
INVALID_CATEGORY_ID_FILE = "../data/dataset/testCOCO/annotations/invalid_category_id.json"
def test_coco_detection():
data1 = ds.CocoDataset(DATA_DIR, annotation_file=ANNOTATION_FILE, task="Detection",
@ -150,7 +151,8 @@ def test_coco_panoptic():
def test_coco_detection_classindex():
data1 = ds.CocoDataset(DATA_DIR, annotation_file=ANNOTATION_FILE, task="Detection", decode=True)
class_index = data1.get_class_indexing()
assert class_index == {'person': [1], 'bicycle': [2], 'car': [3], 'cat': [4], 'dog': [5], 'monkey': [7]}
assert class_index == {'person': [1], 'bicycle': [2], 'car': [3], 'cat': [4], 'dog': [5], 'monkey': [6],
'bag': [7], 'orange': [8]}
num_iter = 0
for _ in data1.__iter__():
num_iter += 1
@ -233,6 +235,14 @@ def test_coco_case_exception():
except RuntimeError as e:
assert "Invalid node found in json" in str(e)
try:
data1 = ds.CocoDataset(DATA_DIR, annotation_file=INVALID_CATEGORY_ID_FILE, task="Detection")
for _ in data1.__iter__():
pass
assert False
except RuntimeError as e:
assert "category_id can't find in categories" in str(e)
try:
data1 = ds.CocoDataset(DATA_DIR, annotation_file=INVALID_FILE, task="Detection")
for _ in data1.__iter__():