deeplabv3 bug fix (#754)

Co-authored-by: Fzilan <[email protected]>

Author: Fzilan

examples/seg/deeplabv3/README.md

@@ -51,7 +51,7 @@ This example provides implementations of DeepLabV3 and DeepLabV3+ using backbone
 
    * Download Pascal VOC 2012 dataset,  [VOC2012](http://host.robots.ox.ac.uk/pascal/VOC/voc2012/) and Semantic Boundaries Dataset, [SBD](https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/semantic_contours/benchmark.tgz).
 
-   * Prepare training and test data list files with the path to image and annotation pairs. You could simply run `python examples/seg/deeplabv3/preprocess/get_dataset_list.py --data_root=/path/to/data` to generate the list files. This command results in 5 data list files. The lines in a list file should be like as follows:
+   * Prepare training and test data list files with the path to image and annotation pairs. You could simply run `python examples/seg/deeplabv3/preprocess/get_data_list.py --data_root=/path/to/data` to generate the list files. This command results in 5 data list files. The lines in a list file should be like as follows:
 
      ```
      /path/to/data/JPEGImages/2007_000032.jpg /path/to/data/SegmentationClassGray/2007_000032.png
@@ -116,20 +116,20 @@ python examples/seg/deeplabv3/train.py --config [the path to the config file]
 
   ```shell
   # for deeplabv3
-  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/deeplabv3_s16_dilated_resnet101.yaml
+  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/config/deeplabv3_s16_dilated_resnet101.yaml
 
   # for deeplabv3+
-  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/deeplabv3plus_s16_dilated_resnet101.yaml
+  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/config/deeplabv3plus_s16_dilated_resnet101.yaml
   ```
 
 - Step 2: Employ output_stride=8, fine-tune model from step 1 on  *trainaug* dataset with smaller base learning rate. In config file, please specify the path of checkpoint from previous step in `ckpt_path`, set  `ckpt_pre_trained` to `True` and set `output_stride` to `8` .
 
   ```shell
   # for deeplabv3
-  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/deeplabv3_s8_dilated_resnet101.yaml
+  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/config/deeplabv3_s8_dilated_resnet101.yaml
 
   # for deeplabv3+
-  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/deeplabv3plus_s8_dilated_resnet101.yaml
+  mpirun -n 8 python examples/seg/deeplabv3/train.py --config examples/seg/deeplabv3/config/deeplabv3plus_s8_dilated_resnet101.yaml
   ```
 
 ### Test
@@ -138,7 +138,7 @@ For testing the trained model, first specify the path to the model checkpoint at
 
 For example, after replacing  `ckpt_path` in config file with [checkpoint](https://download.mindspore.cn/toolkits/mindcv/deeplabv3/deeplabv3_s8_resnet101-a297e7af.ckpt) from 2-step training of deeplabv3, commands below employ os=8 without left-right filpped or muticale inputs.
 ```shell
-python examples/seg/deeplabv3/eval.py --config examples/seg/deeplabv3/deeplabv3_s8_dilated_resnet101.yaml
+python examples/seg/deeplabv3/eval.py --config examples/seg/deeplabv3/config/deeplabv3_s8_dilated_resnet101.yaml
 ```
 
 ## Results

examples/seg/deeplabv3/config/deeplabv3plus_s8_dilated_resnet101.yaml

@@ -18,7 +18,7 @@ image_std: [57.375, 57.120, 58.395]
 max_scale: 2.0
 min_scale: 0.5
 ignore_label: 255
-num_classes: 21x
+num_classes: 21
 shuffle: True
 
 # backbone

examples/seg/deeplabv3/eval.py

@@ -4,7 +4,7 @@
 import yaml
 from addict import Dict
 from data import create_segment_dataset
-from deeplabv3 import DeeplabInferNetwork, DeepLabV3, DeeplabV3Plus
+from deeplabv3 import DeepLabInferNetwork, DeepLabV3, DeepLabV3Plus
 from dilated_resnet import *  # noqa: F403, F401
 from postprocess import apply_eval
 
@@ -34,11 +34,11 @@ def eval(args):
     if args.model == "deeplabv3":
         deeplab = DeepLabV3(backbone, args, is_training=False)
     elif args.model == "deeplabv3plus":
-        deeplab = DeeplabV3Plus(backbone, args, is_training=False)
+        deeplab = DeepLabV3Plus(backbone, args, is_training=False)
     else:
         NotImplementedError("support deeplabv3 and deeplabv3plus only")
 
-    eval_model = DeeplabInferNetwork(deeplab, input_format=args.input_format)
+    eval_model = DeepLabInferNetwork(deeplab, input_format=args.input_format)
 
     param_dict = load_checkpoint(args.ckpt_path)
     net_param_not_load, _ = load_param_into_net(eval_model, param_dict)