Skip to content

Commit 31402ff

Browse files
authored
Update supported mmseg models (#181)
* fix ocrnet cascade decoder * update mmseg support models * update mmseg configs * support emanet and icnet * set max K of TopK for tensorrt * update supported models for mmseg in docs * add test for emamodule * add configs and update docs * Update docs * update benchmark
1 parent 61b0146 commit 31402ff

20 files changed

Lines changed: 736 additions & 44 deletions

configs/mmseg/segmentation_ncnn_static.py

Lines changed: 0 additions & 3 deletions
This file was deleted.
Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/onnxruntime.py']
2+
3+
onnx_config = dict(input_shape=[2048, 1024])
Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
_base_ = ['./segmentation_static.py', '../_base_/backends/onnxruntime.py']
22

3-
onnx_config = dict(input_shape=None)
3+
onnx_config = dict(input_shape=[512, 512])
Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,4 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/openvino.py']
2+
3+
backend_config = dict(
4+
model_inputs=[dict(opt_shapes=dict(input=[1, 3, 512, 512]))])
Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/tensorrt-fp16.py']
2+
3+
onnx_config = dict(input_shape=[1024, 1024])
4+
backend_config = dict(
5+
common_config=dict(max_workspace_size=1 << 30),
6+
model_inputs=[
7+
dict(
8+
input_shapes=dict(
9+
input=dict(
10+
min_shape=[1, 3, 1024, 1024],
11+
opt_shape=[1, 3, 1024, 1024],
12+
max_shape=[1, 3, 1024, 1024])))
13+
])
Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/tensorrt-fp16.py']
2+
3+
onnx_config = dict(input_shape=[512, 512])
4+
backend_config = dict(
5+
common_config=dict(max_workspace_size=1 << 30),
6+
model_inputs=[
7+
dict(
8+
input_shapes=dict(
9+
input=dict(
10+
min_shape=[1, 3, 512, 512],
11+
opt_shape=[1, 3, 512, 512],
12+
max_shape=[1, 3, 512, 512])))
13+
])
Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/tensorrt-int8.py']
2+
3+
onnx_config = dict(input_shape=[1024, 1024])
4+
backend_config = dict(
5+
common_config=dict(max_workspace_size=1 << 30),
6+
model_inputs=[
7+
dict(
8+
input_shapes=dict(
9+
input=dict(
10+
min_shape=[1, 3, 1024, 1024],
11+
opt_shape=[1, 3, 1024, 1024],
12+
max_shape=[1, 3, 1024, 1024])))
13+
])
Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/tensorrt-int8.py']
2+
3+
onnx_config = dict(input_shape=[512, 512])
4+
backend_config = dict(
5+
common_config=dict(max_workspace_size=1 << 30),
6+
model_inputs=[
7+
dict(
8+
input_shapes=dict(
9+
input=dict(
10+
min_shape=[1, 3, 512, 512],
11+
opt_shape=[1, 3, 512, 512],
12+
max_shape=[1, 3, 512, 512])))
13+
])
Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/tensorrt.py']
2+
3+
onnx_config = dict(input_shape=[1024, 1024])
4+
backend_config = dict(
5+
common_config=dict(max_workspace_size=1 << 30),
6+
model_inputs=[
7+
dict(
8+
input_shapes=dict(
9+
input=dict(
10+
min_shape=[1, 3, 1024, 1024],
11+
opt_shape=[1, 3, 1024, 1024],
12+
max_shape=[1, 3, 1024, 1024])))
13+
])
Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
_base_ = ['./segmentation_static.py', '../_base_/backends/tensorrt.py']
2+
3+
onnx_config = dict(input_shape=[512, 512])
4+
backend_config = dict(
5+
common_config=dict(max_workspace_size=1 << 30),
6+
model_inputs=[
7+
dict(
8+
input_shapes=dict(
9+
input=dict(
10+
min_shape=[1, 3, 512, 512],
11+
opt_shape=[1, 3, 512, 512],
12+
max_shape=[1, 3, 512, 512])))
13+
])

0 commit comments

Comments
 (0)