iku50
/
mindspore2022
forked from huawei/mindspore2022
0
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fasterrcnn add export.py in r1.0

This commit is contained in:
linqingke 2020-11-23 16:39:47 +08:00
parent 9fe624527b
commit 81c6acadc1
2 changed files with 53 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
model_zoo/official/cv/faster_rcnn/README.md
View File

@ -33,7 +33,9 @@ FasterRcnn is a two-stage target detection network,This network uses a region pr
# Dataset
Dataset used: [COCO2017](<http://images.cocodataset.org/>)
Note that you can run the scripts based on the dataset mentioned in original paper or widely used in relevant domain/network architecture. In the following sections, we will introduce how to run the scripts using the related dataset below.
Dataset used: [COCO2017](<https://cocodataset.org/>)
- Dataset size：19G
- Train：18G，118000 images
@ -148,7 +150,7 @@ sh run_standalone_train_ascend.sh [PRETRAINED_MODEL]
sh run_distribute_train_ascend.sh [RANK_TABLE_FILE] [PRETRAINED_MODEL]
```
> Rank_table.json which is specified by RANK_TABLE_FILE is needed when you are running a distribute task. You can generate it by using the [hccl_tools](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/utils/hccl_tools).
> Rank_table.json which is specified by RANK_TABLE_FILE is needed when you are running a distribute task. You can generate it by using the [hccl_tools](https://gitee.com/mindspore/mindspore/tree/r1.0/model_zoo/utils/hccl_tools).
> As for PRETRAINED_MODEL，it should be a ResNet50 checkpoint that trained over ImageNet2012. Ready-made pretrained_models are not available now. Stay tuned.
> The original dataset path needs to be in the config.py,you can select "coco_root" or "image_dir".
@ -207,9 +209,9 @@ Eval result will be stored in the example path, whose folder name is "eval". Und
| Parameters | FasterRcnn |
| -------------------------- | ----------------------------------------------------------- |
| Model Version | V1 |
| Resource | Ascend 910 ；CPU 2.60GHz，56cores；Memory，314G |
| Resource | Ascend 910 ；CPU 2.60GHz，192cores；Memory，755G |
| uploaded Date | 08/31/2020 (month/day/year) |
| MindSpore Version | 0.7.0-beta |
| MindSpore Version | 1.0.0 |
| Dataset | COCO2017 |
| Training Parameters | epoch=12, batch_size=2 |
| Optimizer | SGD |
@ -217,7 +219,7 @@ Eval result will be stored in the example path, whose folder name is "eval". Und
| Speed | 1pc: 190 ms/step; 8pcs: 200 ms/step |
| Total time | 1pc: 37.17 hours; 8pcs: 4.89 hours |
| Parameters (M) | 250 |
| Scripts | https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/cv/faster_rcnn |
| Scripts | [fasterrcnn script](https://gitee.com/mindspore/mindspore/tree/r1.0/model_zoo/official/cv/faster_rcnn) |
### Evaluation Performance
@ -227,7 +229,7 @@ Eval result will be stored in the example path, whose folder name is "eval". Und
| Model Version | V1 |
| Resource | Ascend 910 |
| Uploaded Date | 08/31/2020 (month/day/year) |
| MindSpore Version | 0.7.0-beta |
| MindSpore Version | 1.0.0 |
| Dataset | COCO2017 |
| batch_size | 2 |
| outputs | mAP |
@ -235,4 +237,5 @@ Eval result will be stored in the example path, whose folder name is "eval". Und
| Model for inference | 250M (.ckpt file) |
# [ModelZoo Homepage](#contents)
Please check the official [homepage](https://gitee.com/mindspore/mindspore/tree/master/model_zoo).
Please check the official [homepage](https://gitee.com/mindspore/mindspore/tree/r1.0/model_zoo).

43
model_zoo/official/cv/faster_rcnn/export.py Normal file
View File

@ -0,0 +1,43 @@
# Copyright 2020 Huawei Technologies Co., Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ============================================================================
"""export checkpoint file into air models"""
import argparse
import numpy as np
import mindspore as ms
from mindspore import Tensor, load_checkpoint, load_param_into_net, export
from src.FasterRcnn.faster_rcnn_r50 import Faster_Rcnn_Resnet50
from src.config import config
parser = argparse.ArgumentParser(description='fasterrcnn_export')
parser.add_argument('--ckpt_file', type=str, default='', help='fasterrcnn ckpt file.')
parser.add_argument('--output_file', type=str, default='', help='fasterrcnn output air name.')
parser.add_argument('--file_format', type=str, choices=["AIR", "ONNX", "MINDIR"], default='AIR', help='file format')
args = parser.parse_args()
if __name__ == '__main__':
net = Faster_Rcnn_Resnet50(config=config)
param_dict = load_checkpoint(args.ckpt_file)
load_param_into_net(net, param_dict)
img = Tensor(np.zeros([config.test_batch_size, 3, config.img_height, config.img_width]), ms.float16)
img_metas = Tensor(np.random.uniform(0.0, 1.0, size=[config.test_batch_size, 4]), ms.float16)
gt_bboxes = Tensor(np.random.uniform(0.0, 1.0, size=[config.test_batch_size, config.num_gts]), ms.float16)
gt_label = Tensor(np.random.uniform(0.0, 1.0, size=[config.test_batch_size, config.num_gts]), ms.int32)
gt_num = Tensor(np.random.uniform(0.0, 1.0, size=[config.test_batch_size, config.num_gts]), ms.bool_)
export(net, img, img_metas, gt_bboxes, gt_label, gt_num, file_name=args.output_file, file_format=args.file_format)