TensorRT
TensorRT copied to clipboard
one bash is not correct of quickstart/deploy_to_triton/README.md
Description
In Step 1 of quickstart/deploy_to_triton/README.md. , here is a bash: polygraphy inspect model model.plan --mode=basic
and a error come out as following:
polygraphy inspect model: error: argument --model-type: invalid choice: 'basic' (choose from 'frozen', 'keras', 'ckpt', 'onnx', 'engine', 'uff', 'trt-network-script', 'caffe')
Solution
change the --mode=basic into --mode=engine, then we can see the details of the .plan model
[W] 'colored' module is not installed, will not use colors when logging. To enable colors, please install the 'colored' module: python3 -m pip install colored
[I] Loading bytes from /trt_optimize/model.plan
[I] ==== TensorRT Engine ====
Name: Unnamed Network 0 | Explicit Batch Engine
---- 1 Engine Input(s) ----
{input [dtype=float32, shape=(1, 3, 224, 224)]}
---- 1 Engine Output(s) ----
{output [dtype=float32, shape=(1, 1000)]}
---- Memory ----
Device Memory: 7325696 bytes
---- 1 Profile(s) (2 Tensor(s) Each) ----
- Profile: 0
Tensor: input (Input), Index: 0 | Shapes: min=(1, 3, 224, 224), opt=(1, 3, 224, 224), max=(1, 3, 224, 224)
Tensor: output (Output), Index: 1 | Shape: (1, 1000)
---- 87 Layer(s) ----