update examples readme

Signed-off-by: hsj576 <[email protected]>

examples/pedestrian_tracking/README.md → examples/MOT17/multiedge_inference_bench/pedestrian_tracking/README.md

@@ -47,7 +47,7 @@ Next, download pretrained model via [[google]](https://drive.google.com/file/d/1
 We are now ready to run the ianvs for benchmarking pedestrian tracking on the MOT17 dataset.
 
 ```python
-ianvs -f ./examples/pedestrian_tracking/multiedge_inference_bench/tracking_job.yaml
+ianvs -f ./examples/MOT17/multiedge_inference_bench/pedestrian_tracking/tracking_job.yaml
 ```
 
 The benchmarking process takes a few minutes and varies depending on devices.
@@ -78,7 +78,7 @@ Next, download pretrained model via [[google]](https://drive.google.com/drive/fo
 We are now ready to run the ianvs for benchmarking pedestrian re-identification on the MOT17 dataset.
 
 ```python
-ianvs -f ./examples/pedestrian_tracking/multiedge_inference_bench/reid_job.yaml
+ianvs -f ./examples/MOT17/multiedge_inference_bench/pedestrian_tracking/reid_job.yaml
 ```
 
 The benchmarking process takes a few minutes and varies depending on devices.
@@ -93,9 +93,9 @@ The final output might look like this:
 ## Step 3. Generate test report
 
 ```shell
-python ./examples/pedestrian_tracking/multiedge_inference_bench/generate_reports.py \
--t ./examples/pedestrian_tracking/multiedge_inference_bench/tracking_job.yaml \
--r ./examples/pedestrian_tracking/multiedge_inference_bench/reid_job.yaml
+python ./examples/MOT17/multiedge_inference_bench/pedestrian_tracking/generate_reports.py \
+-t ./examples/MOT17/multiedge_inference_bench/pedestrian_tracking/tracking_job.yaml \
+-r ./examples/MOT17/multiedge_inference_bench/pedestrian_tracking/reid_job.yaml
 ```
 
 Finally, the report is generated under <Ianvs_HOME>/examples/pedestrian_tracking/multiedge_inference_bench/reports. You can also check the sample report under the current directory.

examples/bdd/lifelong_learning_bench/README.md → examples/bdd/lifelong_learning_bench/curb-detection/README.md

@@ -64,7 +64,7 @@ we have done that for you and the interested readers can refer to [testenv.yaml]
 
 Related algorithm is also ready in this quick start. 
 ``` shell
-export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/curb-detection/lifelong_learning_bench/testalgorithms/rfnet/RFNet
+export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/bdd/lifelong_learning_bench/curb-detection/testalgorithms/rfnet/RFNet
 ```
 
 The URL address of this algorithm then should be filled in the configuration file ``algorithm.yaml``. In this quick
@@ -115,7 +115,7 @@ We are now ready to run the ianvs for benchmarking.
 
 ``` shell
 cd /ianvs/project
-ianvs -f examples/bdd/lifelong_learning_bench/benchmarkingjob.yaml
+ianvs -f examples/bdd/lifelong_learning_bench/curb-detection/benchmarkingjob.yaml
 ```
 
 Finally, the user can check the result of benchmarking on the console and also in the output path(

examples/bdd/lifelong_learning_bench/testalgorithms/yolo/model_selector/choose_net_b64.py → examples/bdd/lifelong_learning_bench/curb-detection/testalgorithms/yolo/model_selector/choose_net_b64.py

@@ -1,106 +1,106 @@
-model = dict(
-    type='ImageClassifier',
-    backbone=dict(
-        type='ResNet',
-        depth=18,
-        num_stages=4,
-        out_indices=(3, ),
-        style='pytorch'),
-    neck=dict(type='GlobalAveragePooling'),
-    head=dict(
-        type='MultiLabelLinearClsHead',
-        num_classes=20,
-        in_channels=512,
-        loss=dict(type='CrossEntropyLoss', loss_weight=1.0, use_soft=True)))
-dataset_type = 'BDD_Performance'
-img_norm_cfg = dict(
-    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
-train_pipeline = [
-    dict(type='LoadImageFromFile'),
-    dict(type='Resize', size=(256, -1)),
-    dict(
-        type='Normalize',
-        mean=[123.675, 116.28, 103.53],
-        std=[58.395, 57.12, 57.375],
-        to_rgb=True),
-    dict(type='ImageToTensor', keys=['img']),
-    dict(type='ToTensor', keys=['gt_label']),
-    dict(type='Collect', keys=['img', 'gt_label'])
-]
-test_pipeline = [
-    dict(type='LoadImageFromFile'),
-    dict(type='Resize', size=(256, -1)),
-    dict(
-        type='Normalize',
-        mean=[123.675, 116.28, 103.53],
-        std=[58.395, 57.12, 57.375],
-        to_rgb=True),
-    dict(type='ImageToTensor', keys=['img']),
-    dict(type='Collect', keys=['img'])
-]
-data = dict(
-    samples_per_gpu=64,
-    workers_per_gpu=2,
-    train=dict(
-        type='BDD_Performance',
-        data_prefix='',
-        ann_file=
-        '/home/liyunzhe/Mobile-Inference/algorithm/labels/0129_real_world_multi_label_remo_xyxy_bdd_train.txt',
-        pipeline=[
-            dict(type='LoadImageFromFile'),
-            dict(type='Resize', size=(256, -1)),
-            dict(
-                type='Normalize',
-                mean=[123.675, 116.28, 103.53],
-                std=[58.395, 57.12, 57.375],
-                to_rgb=True),
-            dict(type='ImageToTensor', keys=['img']),
-            dict(type='ToTensor', keys=['gt_label']),
-            dict(type='Collect', keys=['img', 'gt_label'])
-        ]),
-    val=dict(
-        type='BDD_Performance',
-        data_prefix='',
-        ann_file=
-        '/home/liyunzhe/Mobile-Inference/algorithm/labels/0129_real_world_multi_label_remo_xyxy_bdd_val.txt',
-        pipeline=[
-            dict(type='LoadImageFromFile'),
-            dict(type='Resize', size=(256, -1)),
-            dict(
-                type='Normalize',
-                mean=[123.675, 116.28, 103.53],
-                std=[58.395, 57.12, 57.375],
-                to_rgb=True),
-            dict(type='ImageToTensor', keys=['img']),
-            dict(type='Collect', keys=['img'])
-        ]),
-    test=dict(
-        type='BDD_Performance',
-        data_prefix='',
-        ann_file=
-        '/home/liyunzhe/Mobile-Inference/algorithm/labels/0129_real_world_multi_label_remo_xyxy_bdd_val.txt',
-        pipeline=[
-            dict(type='LoadImageFromFile'),
-            dict(type='Resize', size=(256, -1)),
-            dict(
-                type='Normalize',
-                mean=[123.675, 116.28, 103.53],
-                std=[58.395, 57.12, 57.375],
-                to_rgb=True),
-            dict(type='ImageToTensor', keys=['img']),
-            dict(type='Collect', keys=['img'])
-        ]))
-evaluation = dict(interval=1, metric='mAP')
-optimizer = dict(type='SGD', lr=0.1, momentum=0.9, weight_decay=0.0001)
-optimizer_config = dict(grad_clip=None)
-lr_config = dict(policy='step', step=[30, 60, 90])
-runner = dict(type='EpochBasedRunner', max_epochs=100)
-checkpoint_config = dict(interval=1)
-log_config = dict(interval=100, hooks=[dict(type='TextLoggerHook')])
-dist_params = dict(backend='nccl')
-log_level = 'INFO'
-load_from = None
-resume_from = None
-workflow = [('train', 1)]
-work_dir = 'work_dirs/220208-bdd-best'
+model = dict(
+    type='ImageClassifier',
+    backbone=dict(
+        type='ResNet',
+        depth=18,
+        num_stages=4,
+        out_indices=(3, ),
+        style='pytorch'),
+    neck=dict(type='GlobalAveragePooling'),
+    head=dict(
+        type='MultiLabelLinearClsHead',
+        num_classes=20,
+        in_channels=512,
+        loss=dict(type='CrossEntropyLoss', loss_weight=1.0, use_soft=True)))
+dataset_type = 'BDD_Performance'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='Resize', size=(256, -1)),
+    dict(
+        type='Normalize',
+        mean=[123.675, 116.28, 103.53],
+        std=[58.395, 57.12, 57.375],
+        to_rgb=True),
+    dict(type='ImageToTensor', keys=['img']),
+    dict(type='ToTensor', keys=['gt_label']),
+    dict(type='Collect', keys=['img', 'gt_label'])
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='Resize', size=(256, -1)),
+    dict(
+        type='Normalize',
+        mean=[123.675, 116.28, 103.53],
+        std=[58.395, 57.12, 57.375],
+        to_rgb=True),
+    dict(type='ImageToTensor', keys=['img']),
+    dict(type='Collect', keys=['img'])
+]
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    train=dict(
+        type='BDD_Performance',
+        data_prefix='',
+        ann_file=
+        '/home/liyunzhe/Mobile-Inference/algorithm/labels/0129_real_world_multi_label_remo_xyxy_bdd_train.txt',
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='Resize', size=(256, -1)),
+            dict(
+                type='Normalize',
+                mean=[123.675, 116.28, 103.53],
+                std=[58.395, 57.12, 57.375],
+                to_rgb=True),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='ToTensor', keys=['gt_label']),
+            dict(type='Collect', keys=['img', 'gt_label'])
+        ]),
+    val=dict(
+        type='BDD_Performance',
+        data_prefix='',
+        ann_file=
+        '/home/liyunzhe/Mobile-Inference/algorithm/labels/0129_real_world_multi_label_remo_xyxy_bdd_val.txt',
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='Resize', size=(256, -1)),
+            dict(
+                type='Normalize',
+                mean=[123.675, 116.28, 103.53],
+                std=[58.395, 57.12, 57.375],
+                to_rgb=True),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img'])
+        ]),
+    test=dict(
+        type='BDD_Performance',
+        data_prefix='',
+        ann_file=
+        '/home/liyunzhe/Mobile-Inference/algorithm/labels/0129_real_world_multi_label_remo_xyxy_bdd_val.txt',
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='Resize', size=(256, -1)),
+            dict(
+                type='Normalize',
+                mean=[123.675, 116.28, 103.53],
+                std=[58.395, 57.12, 57.375],
+                to_rgb=True),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img'])
+        ]))
+evaluation = dict(interval=1, metric='mAP')
+optimizer = dict(type='SGD', lr=0.1, momentum=0.9, weight_decay=0.0001)
+optimizer_config = dict(grad_clip=None)
+lr_config = dict(policy='step', step=[30, 60, 90])
+runner = dict(type='EpochBasedRunner', max_epochs=100)
+checkpoint_config = dict(interval=1)
+log_config = dict(interval=100, hooks=[dict(type='TextLoggerHook')])
+dist_params = dict(backend='nccl')
+log_level = 'INFO'
+load_from = None
+resume_from = None
+workflow = [('train', 1)]
+work_dir = 'work_dirs/220208-bdd-best'
 gpu_ids = range(0, 1)

examples/curb-detection/lifelong_learning_bench/README.md → examples/cityscapes-synthia/lifelong_learning_bench/curb-detection/README.md

@@ -68,7 +68,7 @@ we have done that for you and the interested readers can refer to [testenv.yaml]
 
 Related algorithm is also ready in this quick start. 
 ``` shell
-export PYTHONPATH=$PYTHONPATH:/ianvs/project/ianvs/examples/curb-detection/lifelong_learning_bench/testalgorithms/rfnet/RFNet
+export PYTHONPATH=$PYTHONPATH:/ianvs/project/ianvs/examples/cityscapes-synthia/lifelong_learning_bench/curb-detection/testalgorithms/rfnet/RFNet
 ```
 
 The URL address of this algorithm then should be filled in the configuration file ``algorithm.yaml``. In this quick
@@ -80,7 +80,7 @@ We are now ready to run the ianvs for benchmarking.
 
 ``` shell
 cd /ianvs/project/ianvs
-ianvs -f examples/curb-detection/lifelong_learning_bench/benchmarkingjob.yaml
+ianvs -f examples/cityscapes-synthia/lifelong_learning_bench/curb-detection/benchmarkingjob.yaml
 ```
 
 Finally, the user can check the result of benchmarking on the console and also in the output path(

examples/cItyscapes+synthia/lifelong_learning_bench/README.md → examples/cityscapes-synthia/lifelong_learning_bench/semantic-segmentation/README.md

@@ -13,6 +13,7 @@ Before using Ianvs, you might want to have the device ready:
 - Internet connection for GitHub and pip, etc
 - Python 3.6+ installed
 
+
 In this example, we are using the Linux platform with Python 3.8.5. If you are using Windows, most steps should still apply but a few like commands and package requirements might be different. 
 
 ## Step 1. Ianvs Preparation
@@ -65,15 +66,15 @@ we have done that for you and the interested readers can refer to [testenv.yaml]
 Copy the index files of dataset.
 
 ``` shell
-cp /ianvs/project/ianvs/examples/semantic_segmentation/lifelong_learning_bench/indexes/* /root/data/semantic_segmentation_dataset/
+cp /ianvs/project/ianvs/examples/cityscapes-synthia/lifelong_learning_bench/semantic-segmentation/indexes/* /root/data/semantic_segmentation_dataset/
 ```
 
 <!-- Please put the downloaded dataset on the above dataset path, e.g., `/ianvs/dataset`. One can transfer the dataset to the path, e.g., on a remote Linux system using [XFTP].  -->
 
 
 Related algorithm is also ready in this quick start. 
 ``` shell
-export PYTHONPATH=$PYTHONPATH:/ianvs/project/ianvs/examples/semantic_segmentation/lifelong_learning_bench/testalgorithms/rfnet/RFNet
+export PYTHONPATH=$PYTHONPATH:/ianvs/project/ianvs/examples/cityscapes-synthia/lifelong_learning_bench/semantic-segmentation/testalgorithms/rfnet/RFNet
 ```
 
 The URL address of this algorithm then should be filled in the configuration file ``algorithm.yaml``. In this quick
@@ -85,7 +86,7 @@ We are now ready to run the ianvs for benchmarking.
 
 ``` shell
 cd /ianvs/project/ianvs
-ianvs -f examples/semantic_segmentation/lifelong_learning_bench/benchmarkingjob-smalltest.yaml
+ianvs -f examples/cityscapes-synthia/lifelong_learning_bench/semantic-segmentation/benchmarkingjob-smalltest.yaml
 ```
 
 Finally, the user can check the result of benchmarking on the console and also in the output path(

examples/curb-detection/scene-based-unknown-task-recognition/lifelong_learning_bench/Readme.md → examples/cityscapes-synthia/scene-based-unknown-task-recognition/curb-detection/Readme.md

@@ -59,7 +59,7 @@ unzip dataset.zip
 The URL address of this dataset then should be filled in the configuration file `testenv.yaml`. In this quick start, we have done that for you and interested readers can refer to [testenv.yaml](https://ianvs.readthedocs.io/en/latest/guides/how-to-test-algorithms.html#step-1-test-environment-preparation) for more details.
 
 ```shell
-cd /ianvs/examples/scene-based-unknown-task-recognition/lifelong_learning_bench/testalgorithms/rfnet
+cd /ianvs/examples/cityscapes-synthia/scene-based-unknown-task-recognition/curb-detection/testalgorithms/rfnet
 mkdir results
 ```
 
@@ -68,8 +68,8 @@ Put the model to results.Download [model](https://pan.baidu.com/s/18MA8Gaw7ptpip
 The related algorithm is also ready for this quick start.
 
 ```shell
-export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/scene-based-unknown-task-recognition/lifelong_learning_bench/testalgorithms/rfnet/RFNet
-export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/scene-based-unknown-task-recognition/lifelong_learning_bench/testalgorithms/rfnet/
+export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/cityscapes-synthia/scene-based-unknown-task-recognition/curb-detection/testalgorithms/rfnet/RFNet
+export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/cityscapes-synthia/scene-based-unknown-task-recognition/curb-detection/testalgorithms/rfnet/
 ```
 
 The URL address of this algorithm then should be filled in the configuration file `algorithm.yaml`. In this quick start, we have done that for you and interested readers can refer to [algorithm.yaml](https://ianvs.readthedocs.io/en/latest/guides/how-to-test-algorithms.html#step-1-test-environment-preparation) for more details.
@@ -80,7 +80,7 @@ We are now ready to run the ianvs for benchmarking.
 
 ```shell
 cd /ianvs/project
-ianvs -f examples/scene-based-unknown-task-recognition/lifelong_learning_bench/benchmarkingjob.yaml
+ianvs -f examples/cityscapes-synthia/scene-based-unknown-task-recognition/curb-detection/benchmarkingjob.yaml
 ```
 
 Finally, the user can check the result of benchmarking on the console and also in the output path( e.g. `/ianvs/lifelong_learning_bench/workspace`) defined in the benchmarking config file ( e.g. `benchmarkingjob.yaml`). In this quick start, we have done all configurations for you and the interested readers can refer to [benchmarkingJob.yaml](https://ianvs.readthedocs.io/en/latest/guides/how-to-test-algorithms.html#step-1-test-environment-preparation) for more details.

examples/how-to-contribute-examples.md

@@ -0,0 +1,34 @@
+# How to contribute examples
+
+## Overall contribution workflow
+
+1. Apply for a topic.
+   Once you have a new example, you can apply for a topic to discuss it on [SIG AI weekly meeting](http://github.com/kubeedge/ianvs.git).
+2. Submit proposal.
+   After the idea is fully discussed, the former proposal PR is needed to submit to the [Ianvs repository](http://github.com/kubeedge/ianvs.git).
+3. Fix proposal review comments.  
+   If other Ianvs maintainers leave review comments to the PR, you need to fix them and get at least 2 reviewers' `/lgtm`, and 1 approver's `/approve`.
+4. Submit code.
+   Then you can implement your code, and a good code style is encouraged.
+5. Fix code review comments.  
+   Besides the merge requirements of the proposal, CI passing is needed before reviewing this step.
+
+## Add a new example
+
+The new example should be stored in the following path:
+
+~~~bash
+examples/dataset_name/algorithm_name/task_name/
+~~~
+
+Here is an example:
+
+~~~bash
+examples/robot/lifelong_learning_bench/semantic-segmentation/
+~~~
+
+For contributing a new example, you can follow these steps to determine its storage path:
+
+1. Under the examples directory, choose the dataset you used in this example, such as cityscapes, robot, and so on. Only when you use a new dataset can you create a new folder under the examples directory to store the example related to that dataset.
+2. After determining the dataset, select the algorithm paradigm you used, such as lifelong learning, single-task learning, and so on. If you used a new algorithm paradigm, you can create a new folder under the dataset directory to store examples of that type of algorithm.
+3. After determining the algorithm paradigm, select the task for your example, such as semantic segmentation, curb detection, and so on. If you used a new task, you can create a new folder under the algorithm paradigm directory to store examples of that type of task.

examples/robot+cItyscapes+synthia/lifelong_learning_bench/README.md → examples/robot-cityscapes-synthia/lifelong_learning_bench/semantic-segmentation/README.md

@@ -64,7 +64,7 @@ we have done that for you and the interested readers can refer to [testenv.yaml]
 Related algorithm is also ready in this quick start. 
 
 ``` shell
-export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/class_increment_semantic_segmentation/lifelong_learning_bench/testalgorithms/erfnet/ERFNet
+export PYTHONPATH=$PYTHONPATH:/ianvs/project/examples/robot-cityscapes-synthia/lifelong_learning_bench/semantic-segmentation/testalgorithms/erfnet/ERFNet
 ```
 
 The URL address of this algorithm then should be filled in the configuration file ``algorithm.yaml``. In this quick
@@ -77,7 +77,7 @@ We are now ready to run the ianvs for benchmarking.
 
 ``` shell
 cd /ianvs/project
-ianvs -f examples/class_increment_semantic_segmentation/lifelong_learning_bench/benchmarkingjob.yaml
+ianvs -f examples/robot-cityscapes-synthia/lifelong_learning_bench/semantic-segmentation/benchmarkingjob.yaml
 ```
 
 Finally, the user can check the result of benchmarking on the console and also in the output path(