|
|
|
@ -41,7 +41,7 @@ Here we used 4 datasets for training, and 1 datasets for Evaluation.
|
|
|
|
|
- Train: 27.7MB, 410 images
|
|
|
|
|
- Dataset3: SCUT-FORU: Flickr OCR Universal Database
|
|
|
|
|
- Train: 388MB, 1715 images
|
|
|
|
|
- Dataset4: CocoText v2(Subset of MSCOCO2014):
|
|
|
|
|
- Dataset4: CocoText v2(Subset of MSCOCO2017):
|
|
|
|
|
- Train: 13GB, 63686 images
|
|
|
|
|
|
|
|
|
|
# [Features](#contents)
|
|
|
|
@ -100,9 +100,9 @@ Here we used 4 datasets for training, and 1 datasets for Evaluation.
|
|
|
|
|
# distribute training example(8p)
|
|
|
|
|
sh run_distribute_train_ascend.sh [IMGS_PATH] [ANNOS_PATH] [RANK_TABLE_FILE] [PRETRAINED_PATH] [COCO_TEXT_PARSER_PATH]
|
|
|
|
|
# standalone training
|
|
|
|
|
sh run_standalone_train_ascend.sh [IMGS_PATH] [ANNOS_PATH] [PRETRAINED_PATH] [COCO_TEXT_PARSER_PATH]
|
|
|
|
|
sh run_standalone_train_ascend.sh [IMGS_PATH] [ANNOS_PATH] [PRETRAINED_PATH] [COCO_TEXT_PARSER_PATH] [DEVICE_ID]
|
|
|
|
|
# evaluation:
|
|
|
|
|
sh run_eval_ascend.sh [IMGS_PATH] [ANNOS_PATH] [CHECKPOINT_PATH] [COCO_TEXT_PARSER_PATH]
|
|
|
|
|
sh run_eval_ascend.sh [IMGS_PATH] [ANNOS_PATH] [CHECKPOINT_PATH] [COCO_TEXT_PARSER_PATH] [DEVICE_ID]
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
> Notes:
|
|
|
|
@ -122,7 +122,7 @@ sh run_eval_ascend.sh [IMGS_PATH] [ANNOS_PATH] [CHECKPOINT_PATH] [COCO_TEXT_PARS
|
|
|
|
|
# distribute training example(8p)
|
|
|
|
|
sh run_distribute_train_ascend.sh [IMGS_PATH] [ANNOS_PATH] [RANK_TABLE_FILE] [PRETRAINED_PATH] [COCO_TEXT_PARSER_PATH]
|
|
|
|
|
# standalone training
|
|
|
|
|
sh run_standalone_train_ascend.sh [IMGS_PATH] [ANNOS_PATH] [PRETRAINED_PATH] [COCO_TEXT_PARSER_PATH]
|
|
|
|
|
sh run_standalone_train_ascend.sh [IMGS_PATH] [ANNOS_PATH] [PRETRAINED_PATH] [COCO_TEXT_PARSER_PATH] [DEVICE_ID]
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
### Result
|
|
|
|
@ -144,7 +144,7 @@ You can start training using python or shell scripts. The usage of shell scripts
|
|
|
|
|
- Ascend:
|
|
|
|
|
|
|
|
|
|
```bash
|
|
|
|
|
sh run_eval_ascend.sh [IMGS_PATH] [ANNOS_PATH] [CHECKPOINT_PATH] [COCO_TEXT_PARSER_PATH]
|
|
|
|
|
sh run_eval_ascend.sh [IMGS_PATH] [ANNOS_PATH] [CHECKPOINT_PATH] [COCO_TEXT_PARSER_PATH] [DEVICE_ID]
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
### Launch
|
|
|
|
@ -153,7 +153,7 @@ You can start training using python or shell scripts. The usage of shell scripts
|
|
|
|
|
# eval example
|
|
|
|
|
shell:
|
|
|
|
|
Ascend:
|
|
|
|
|
sh run_eval_ascend.sh [IMGS_PATH] [ANNOS_PATH] [CHECKPOINT_PATH] [COCO_TEXT_PARSER_PATH]
|
|
|
|
|
sh run_eval_ascend.sh [IMGS_PATH] [ANNOS_PATH] [CHECKPOINT_PATH] [COCO_TEXT_PARSER_PATH] [DEVICE_ID]
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
> checkpoint can be produced in training process.
|
|
|
|
@ -186,7 +186,6 @@ class 1 precision is 88.01%, recall is 82.77%
|
|
|
|
|
| Optimizer | Momentum |
|
|
|
|
|
| Loss Function | SoftmaxCrossEntropyWithLogits for classification, SmoothL2Loss for bbox regression|
|
|
|
|
|
| Loss | ~0.008 |
|
|
|
|
|
| Accuracy (8p) | precision=0.8854, recall=0.8024 |
|
|
|
|
|
| Total time (8p) | 4h |
|
|
|
|
|
| Scripts | [deeptext script](https://gitee.com/mindspore/mindspore/tree/r1.1/mindspore/official/cv/deeptext) |
|
|
|
|
|
|
|
|
|
@ -200,7 +199,7 @@ class 1 precision is 88.01%, recall is 82.77%
|
|
|
|
|
| MindSpore Version | 1.1.0 |
|
|
|
|
|
| Dataset | 229 images |
|
|
|
|
|
| Batch_size | 2 |
|
|
|
|
|
| Accuracy | precision=0.8854, recall=0.8024 |
|
|
|
|
|
| Accuracy | precision=0.8801, recall=0.8277 |
|
|
|
|
|
| Total time | 1 min |
|
|
|
|
|
| Model for inference | 3492M (.ckpt file) |
|
|
|
|
|
|
|
|
|
@ -208,11 +207,11 @@ class 1 precision is 88.01%, recall is 82.77%
|
|
|
|
|
|
|
|
|
|
| **Ascend** | train performance |
|
|
|
|
|
| :--------: | :---------------: |
|
|
|
|
|
| 1p | 42 img/s |
|
|
|
|
|
| 1p | 14 img/s |
|
|
|
|
|
|
|
|
|
|
| **Ascend** | train performance |
|
|
|
|
|
| :--------: | :---------------: |
|
|
|
|
|
| 8p | 330 img/s |
|
|
|
|
|
| 8p | 50 img/s |
|
|
|
|
|
|
|
|
|
|
# [Description of Random Situation](#contents)
|
|
|
|
|
|
|
|
|
|