[Refactor] Refactor config v1 (open-mmlab#80)

* [Refactor] Refactor configs according to new standard (open-mmlab#67)

* modify cfg and cfg_util

* modify tensorrt config

* fix bug

* lint

* Fix

1. Delete print
2. Modify the return value from "False, None" to "None" and related code
3. Rename 2 get functions

* modify apply_marks

* [Feature] Refactor ocr config (open-mmlab#71)

* add text detection config refactor

* add text recognition refactor

* add static exporting for mmocr

* fix lint

* set max space in child config

* use Sequence[int] instead

* add assert input_shape

* fix static bug and add ppl ort and trt static (open-mmlab#77)

* [Feature] Refine setup.py (open-mmlab#61)

* add setup.py and related files

* lint

* Edit requirements

* modify onnx version

* modify according to comments

* [Refactor] Refactor mmseg configs  (open-mmlab#73)

* refactor mmseg config

* change create_input

* fix lint

* fix lint

* fix lint

* fix yapf

* fix yapf

* update export

* remove Segmentation

* remove tast assert

* add onnx_config

* remove hardcode

* Inherit with static

* Remove blank line

* Add segmentation task enum

* add assert task

* mmocr version 0.3.0 (open-mmlab#79)

* add dump_info

* [Feature]: Refactor config in mmdet (open-mmlab#75)

* support onnxruntime

* add two stage

* test two-stage ort and ppl

* update fcos post_params

* fix calib

* test ok with maskrcnn dynamic

* add empty line

* add static into config filename

* add input_shape to create_input in mmdet

* add static to some configs

* remove todo codes

* remove partition config in base

* refactor create_input

* rename task name in mmdet

* return None if input_shape is None

* add size info into mmdet configs filenames

* reorganize mmdet configs

* add object detection task for mmdet

* rename get_mmdet_params

* keep naming style consistent

* update post_params for fcos

* fix typo in ncnn config

* [Refactor] Refactor mmedit static config (open-mmlab#78)

* add static cfg

* update create_input

* [Refactor]: Refactor mmcls configs (open-mmlab#74)

* refactor mmcls2.0

* fix classify_tensorrt_dynamic.py

* fix classify_tensorrt_dynmic.py

* classify_tensorrt_dynamic_int8.py

* fix file name

* fix ncnn ppl

* updata prepare_input.py

* update utils.py

* updata constant.py

* add

* fix prepare_input.py

* fix prepare_input.py

* add static config file

* add blank lines

* fix prepare_input.py(wait test)

* fix input_shape(wait test)

* Update prepare_input.py

* fix classification_tensorrt_dynamic(wait test)

* fix classification_tensorrt_dynamic_int8(wait test)

* fix classification_tensorrt_static_int8(wait test)

* Rename classification_tensorrt_dynamic.py to classification_tensorrt_dynamic-224x224-224x224.py

* Rename classification_tensorrt_dynamic_int8.py to classification_tensorrt_dynamic_int8-224x224-224x224.py

* Rename classification_tensorrt_dynamic_int8-224x224-224x224.py to classification_tensorrt_int8_dynamic_224x224-224x224.py

* Rename classification_tensorrt_dynamic-224x224-224x224.py to classification_tensorrt_dynamic_224x224-224x224.py

* Rename classification_tensorrt_static.py to classification_tensorrt_static_224x224.py

* Rename classification_tensorrt_static_int8.py to classification_tensorrt_int8_static_224x224.py

* Update prepare_input.py

* Rename classification_tensorrt_dynamic_224x224-224x224.py to classification_tensorrt_dynamic-224x224-224x224.py

* Rename classification_tensorrt_int8_dynamic_224x224-224x224.py to classification_tensorrt_int8-dynamic_224x224-224x224.py

* Rename classification_tensorrt_int8-dynamic_224x224-224x224.py to classification_tensorrt_int8_dynamic-224x224-224x224.py

* Rename classification_tensorrt_int8_static_224x224.py to classification_tensorrt_int8_static-224x224.py

* Rename classification_tensorrt_static_224x224.py to classification_tensorrt_static-224x224.py

* Update prepare_input.py

* Update prepare_input.py

* Update prepare_input.py

* Update prepare_input.py

* Update prepare_input.py

* Update prepare_input.py

* Update prepare_input.py

* change logging msg

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

* fix

* fix else branch

* fix bug for trt in mmseg

* enable dump trt info

* fix trt static for mmdet

* remove two-stage_partition_tensorrt_static-800x1344 config

* fix wrong backend in ppl config

* fix partition calibration

Co-authored-by: Yifan Zhou <[email protected]>
Co-authored-by: AllentDan <[email protected]>
Co-authored-by: hanrui1sensetime <[email protected]>
Co-authored-by: RunningLeon <[email protected]>
Co-authored-by: VVsssssk <[email protected]>
Co-authored-by: maningsheng <[email protected]>
Co-authored-by: AllentDan <[email protected]>

MANIFEST.in

@@ -0,0 +1 @@
+include requirements/*.txt

backend_ops/CMakeLists.txt

@@ -1,4 +1,4 @@
-add_definitions(-std=c++11)
+set(CMAKE_CXX_STANDARD 11)
 set(CMAKE_CXX_FLAGS_RELEASE "-O3")
 
 set(CMAKE_ARCHIVE_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/lib)

configs/_base_/backends/ncnn.py

@@ -1 +1 @@
-backend = 'ncnn'
+backend_config = dict(type='ncnn')

configs/_base_/backends/onnxruntime.py

@@ -1 +1 @@
-backend = 'onnxruntime'
+backend_config = dict(type='onnxruntime')

configs/_base_/backends/ppl.py

@@ -1 +1 @@
-backend = 'ppl'
+backend_config = dict(type='ppl')

configs/_base_/backends/tensorrt.py

@@ -1,5 +1,6 @@
 import tensorrt as trt
 
-backend = 'tensorrt'
-tensorrt_params = dict(
-    shared_params=dict(fp16_mode=False, log_level=trt.Logger.INFO))
+backend_config = dict(
+    type='tensorrt',
+    common_config=dict(
+        fp16_mode=False, log_level=trt.Logger.INFO, max_workspace_size=0))

configs/_base_/backends/tensorrt_int8.py

@@ -1,5 +1,5 @@
 _base_ = ['./tensorrt.py']
 
-create_calib = True
-calib_params = dict(calib_file='calib_data.h5')
-tensorrt_params = dict(shared_params=dict(fp16_mode=True, int8_mode=True))
+backend_config = dict(common_config=dict(fp16_mode=True, int8_mode=True))
+
+calib_config = dict(create_calib=True, calib_file='calib_data.h5')

configs/_base_/onnx_config.py

@@ -0,0 +1,9 @@
+onnx_config = dict(
+    type='onnx',
+    export_params=True,
+    keep_initializers_as_inputs=False,
+    opset_version=11,
+    save_file='end2end.onnx',
+    input_names=['input'],
+    output_names=['output'],
+    input_shape=None)

configs/mmcls/classification_dynamic.py

@@ -0,0 +1,13 @@
+_base_ = ['./classification_static.py']
+
+onnx_config = dict(
+    dynamic_axes={
+        'input': {
+            0: 'batch',
+            2: 'height',
+            3: 'width'
+        },
+        'output': {
+            0: 'batch'
+        }
+    }, )

configs/mmcls/classification_ncnn_dynamic.py

@@ -0,0 +1 @@
+_base_ = ['./classification_dynamic.py', '../_base_/backends/ncnn.py']

configs/mmcls/classification_ncnn_static.py

@@ -0,0 +1,3 @@
+_base_ = ['./classification_static.py', '../_base_/backends/ncnn.py']
+
+onnx_config = dict(input_shape=None)

configs/mmcls/classification_onnxruntime_dynamic.py

@@ -0,0 +1 @@
+_base_ = ['./classification_dynamic.py', '../_base_/backends/onnxruntime.py']

configs/mmcls/classification_onnxruntime_static.py

@@ -0,0 +1,3 @@
+_base_ = ['./classification_static.py', '../_base_/backends/onnxruntime.py']
+
+onnx_config = dict(input_shape=None)

configs/mmcls/classification_ppl_dynamic.py

@@ -0,0 +1 @@
+_base_ = ['./classification_dynamic.py', '../_base_/backends/ppl.py']

configs/mmcls/classification_ppl_static.py

@@ -0,0 +1,3 @@
+_base_ = ['./classification_static.py', '../_base_/backends/ppl.py']
+
+onnx_config = dict(input_shape=None)

configs/mmcls/classification_static.py

@@ -0,0 +1,3 @@
+_base_ = ['../_base_/onnx_config.py']
+
+codebase_config = dict(type='mmcls', task='Classification')

configs/mmcls/classification_tensorrt_dynamic-224x224-224x224.py

@@ -0,0 +1,13 @@
+_base_ = ['./classification_dynamic.py', '../_base_/backends/tensorrt.py']
+
+onnx_config = dict(input_shape=[224, 224])
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 224, 224],
+                    opt_shape=[4, 3, 224, 224],
+                    max_shape=[64, 3, 224, 224])))
+    ])

configs/mmcls/classification_tensorrt_int8_dynamic-224x224-224x224.py

@@ -0,0 +1,13 @@
+_base_ = ['./classification_dynamic.py', '../_base_/backends/tensorrt_int8.py']
+
+onnx_config = dict(input_shape=[224, 224])
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 224, 224],
+                    opt_shape=[4, 3, 224, 224],
+                    max_shape=[64, 3, 224, 224])))
+    ])

configs/mmcls/classification_tensorrt_int8_static-224x224.py

@@ -0,0 +1,13 @@
+_base_ = ['./classification_static.py', '../_base_/backends/tensorrt_int8.py']
+
+onnx_config = dict(input_shape=[224, 224])
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 224, 224],
+                    opt_shape=[1, 3, 224, 224],
+                    max_shape=[1, 3, 224, 224])))
+    ])

configs/mmcls/classification_tensorrt_static-224x224.py

@@ -0,0 +1,13 @@
+_base_ = ['./classification_static.py', '../_base_/backends/tensorrt.py']
+
+onnx_config = dict(input_shape=[224, 224])
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 224, 224],
+                    opt_shape=[1, 3, 224, 224],
+                    max_shape=[1, 3, 224, 224])))
+    ])

configs/mmdet/base_dynamic.py → configs/mmdet/_base_/base_dynamic.py

@@ -1,5 +1,5 @@
 _base_ = ['./base_static.py']
-pytorch2onnx = dict(
+onnx_config = dict(
     dynamic_axes={
         'input': {
             0: 'batch',

configs/mmdet/_base_/base_static.py

@@ -0,0 +1,14 @@
+_base_ = ['../../_base_/onnx_config.py']
+
+onnx_config = dict(output_names=['dets', 'labels'], input_shape=None)
+codebase_config = dict(
+    type='mmdet',
+    task='ObjectDetection',
+    post_processing=dict(
+        score_threshold=0.05,
+        iou_threshold=0.5,
+        max_output_boxes_per_class=200,
+        pre_top_k=-1,
+        keep_top_k=100,
+        background_label_id=-1,
+    ))

configs/mmdet/_base_/base_tensorrt_dynamic-320x320-1344x1344.py

@@ -0,0 +1,12 @@
+_base_ = ['./base_dynamic.py', '../../_base_/backends/tensorrt.py']
+
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 320, 320],
+                    opt_shape=[1, 3, 800, 1344],
+                    max_shape=[1, 3, 1344, 1344])))
+    ])

configs/mmdet/_base_/base_tensorrt_int8_dynamic-320x320-1344x1344.py

@@ -0,0 +1,12 @@
+_base_ = ['./base_dynamic.py', '../../_base_/backends/tensorrt_int8.py']
+
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 320, 320],
+                    opt_shape=[1, 3, 800, 1344],
+                    max_shape=[1, 3, 1344, 1344])))
+    ])

configs/mmdet/_base_/base_tensorrt_int8_static-800x1344.py

@@ -0,0 +1,14 @@
+_base_ = ['./base_static.py', '../../_base_/backends/tensorrt_int8.py']
+
+onnx_config = dict(input_shape=(1344, 800))
+
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 800, 1344],
+                    opt_shape=[1, 3, 800, 1344],
+                    max_shape=[1, 3, 800, 1344])))
+    ])

configs/mmdet/_base_/base_tensorrt_static-800x1344.py

@@ -0,0 +1,14 @@
+_base_ = ['./base_static.py', '../../_base_/backends/tensorrt.py']
+
+onnx_config = dict(input_shape=(1344, 800))
+
+backend_config = dict(
+    common_config=dict(max_workspace_size=1 << 30),
+    model_inputs=[
+        dict(
+            input_shapes=dict(
+                input=dict(
+                    min_shape=[1, 3, 800, 1344],
+                    opt_shape=[1, 3, 800, 1344],
+                    max_shape=[1, 3, 800, 1344])))
+    ])

configs/mmdet/mask_base.py → configs/mmdet/_base_/mask_base_dynamic.py

@@ -1,6 +1,5 @@
-_base_ = ['./base_dynamic.py']
-pytorch2onnx = dict(
-    output_names=['dets', 'labels', 'masks'],
+_base_ = ['./mask_base_static.py']
+onnx_config = dict(
     dynamic_axes={
         'input': {
             0: 'batch',
@@ -21,5 +20,4 @@
             2: 'height',
             3: 'width'
         },
-    },
-)
+    })

configs/mmdet/_base_/mask_base_static.py

@@ -0,0 +1,3 @@
+_base_ = ['./base_static.py']
+
+onnx_config = dict(output_names=['dets', 'labels', 'masks'])

configs/mmdet/single-stage/single-stage_onnxruntime_dynamic.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_dynamic.py', '../../_base_/backends/onnxruntime.py']

configs/mmdet/single-stage/single-stage_onnxruntime_static.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_static.py', '../../_base_/backends/onnxruntime.py']

configs/mmdet/single-stage/single-stage_partition_ncnn_static.py

@@ -0,0 +1,3 @@
+_base_ = ['../_base_/base_static.py', '../../_base_/backends/ncnn.py']
+
+partition_config = dict(type='single_stage', apply_marks=True)

configs/mmdet/single-stage/single-stage_partition_onnxruntime_dynamic.py

@@ -0,0 +1,3 @@
+_base_ = ['./single-stage_onnxruntime_dynamic.py']
+
+partition_config = dict(type='single_stage', apply_marks=True)

configs/mmdet/single-stage/single-stage_partition_onnxruntime_static.py

@@ -0,0 +1,3 @@
+_base_ = ['./single-stage_onnxruntime_static.py']
+
+partition_config = dict(type='single_stage', apply_marks=True)

configs/mmdet/single-stage/single-stage_ppl_dynamic.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_dynamic.py', '../../_base_/backends/ppl.py']

configs/mmdet/single-stage/single-stage_ppl_static.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_static.py', '../../_base_/backends/ppl.py']

configs/mmdet/single-stage/single-stage_tensorrt_dynamic-320x320-1344x1344.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_tensorrt_dynamic-320x320-1344x1344.py']

configs/mmdet/single-stage/single-stage_tensorrt_int8_dynamic-320x320-1344x1344.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_tensorrt_int8_dynamic-320x320-1344x1344.py']

configs/mmdet/single-stage/single-stage_tensorrt_int8_static-800x1344.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_tensorrt_int8_static-800x1344.py']

configs/mmdet/single-stage/single-stage_tensorrt_static-800x1344.py

@@ -0,0 +1 @@
+_base_ = ['../_base_/base_tensorrt_static-800x1344.py']