Skip to content

Commit f5df8f6

Browse files
authored
docs: update files
1 parent facbee0 commit f5df8f6

File tree

1 file changed

+14
-0
lines changed

1 file changed

+14
-0
lines changed

docs/blog/posts/about_model/support_mnn_engine.md

Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -51,6 +51,20 @@ pip install MNN
5151
MNNConvert -f ONNX --modelFile rapidocr/models/ch_PP-OCRv4_det_infer.onnx --MNNModel mnn/ch_PP-OCRv4_det_infer.mnn --bizCode MNN
5252
```
5353

54+
输出下面类似日志,即为转换成功:
55+
56+
```bash linenums="1"
57+
The device supports: i8sdot:1, fp16:1, i8mm: 1, sve2: 0, sme2: 0
58+
Start to Convert Other Model Format To MNN Model..., target version: 3.2
59+
[09:38:49] :46: ONNX Model ir version: 10
60+
[09:38:49] :47: ONNX Model opset version: 14
61+
Start to Optimize the MNN Net...
62+
caution: some weight absolute values are not zero and smaller than float min:1.175494e-38, please check your training process. op name:Add.119
63+
inputTensors : [ x, ]
64+
outputTensors: [ fetch_name_0, ]
65+
Converted Success!
66+
```
67+
5468
### 比较转化前后推理精度差异
5569

5670
#### 运行环境

0 commit comments

Comments
 (0)