Jarvis73 commited on
Commit
55a0542
1 Parent(s): f27445d

Upload ./QA.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. QA.md +4 -4
QA.md CHANGED
@@ -1,6 +1,6 @@
1
  # ❓ Q&A for Building TensorRT Engines
2
 
3
- ### 1. Error: `trtexec: not found`
4
 
5
  **Solution:** Try the following command
6
 
@@ -10,13 +10,13 @@ source trt/activate.sh
10
 
11
  This command is used to configure the environment variables required to build the TensorRT Engine.
12
 
13
- ### 2. Error: `Unable to open library: libnvinfer_plugin.so.9 due to libcublas.so.11: cannot open shared object file`
14
 
15
- This error occurs because the `libnvinfer_plugin.so.9` library depends on the `libcublas.so.11` library.
16
  However, you may use a different version of cuda library (see `/usr/local/cuda`), such as cuda-12. It is recommended
17
  to use cuda-11 to build the TensorRT Engine.
18
 
19
- ### 3. The size of the engine file generated by each build is different
20
 
21
  **Explanation:** During the process of building the engine, the building program will automatically select the
22
  optimal kernel and other parameters. Therefore, the size of the engine file generated by each build may be different.
 
1
  # ❓ Q&A for Building TensorRT Engines
2
 
3
+ ## 1. Error: `trtexec: not found`
4
 
5
  **Solution:** Try the following command
6
 
 
10
 
11
  This command is used to configure the environment variables required to build the TensorRT Engine.
12
 
13
+ ## 2. Error: `Unable to open library: libnvinfer_plugin.so.9 due to libcublas.so.11: cannot open shared object file`
14
 
15
+ **Solution:** This error occurs because the `libnvinfer_plugin.so.9` library depends on the `libcublas.so.11` library.
16
  However, you may use a different version of cuda library (see `/usr/local/cuda`), such as cuda-12. It is recommended
17
  to use cuda-11 to build the TensorRT Engine.
18
 
19
+ ## 3. The size of the engine file generated by each build is different
20
 
21
  **Explanation:** During the process of building the engine, the building program will automatically select the
22
  optimal kernel and other parameters. Therefore, the size of the engine file generated by each build may be different.