RRoundTable
commited on
Commit
•
91b99e6
1
Parent(s):
0232313
Upload model
Browse files
Makefile
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
triton:
|
2 |
+
docker run --rm -d \
|
3 |
+
--name dinov2_vitl14_triton \
|
4 |
+
-p 8000:8000 --gpus "device=0" -p 8001:8001 -p 8002:8002 \
|
5 |
+
-v $(PWD)/model_repository:/models \
|
6 |
+
nvcr.io/nvidia/tritonserver:23.04-py3 \
|
7 |
+
tritonserver --model-repository=/models
|
8 |
+
|
9 |
+
perf:
|
10 |
+
docker run --gpus all --rm -it --net host nvcr.io/nvidia/tritonserver:23.04-py3-sdk perf_analyzer -m dinov2_vitl14 --percentile=95 -i grpc -u 0.0.0.0:8001 --concurrency-range 16:16 --shape input:3,280,280
|
model_repository/dinov2_vitl14/config.pbtxt
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
name: "dinov2_vitl14"
|
2 |
+
max_batch_size: 8
|
3 |
+
dynamic_batching {}
|
4 |
+
platform: "onnxruntime_onnx"
|
5 |
+
instance_group [
|
6 |
+
{
|
7 |
+
count: 1
|
8 |
+
kind: KIND_GPU
|
9 |
+
}
|
10 |
+
]
|
11 |
+
input [
|
12 |
+
{
|
13 |
+
name: "input"
|
14 |
+
data_type: TYPE_FP32
|
15 |
+
dims: [ 3, 280, 280 ]
|
16 |
+
}
|
17 |
+
]
|
18 |
+
output [
|
19 |
+
{
|
20 |
+
name: "output"
|
21 |
+
data_type: TYPE_FP32
|
22 |
+
dims: [-1]
|
23 |
+
}
|
24 |
+
]
|
model_repository/dinov2_vitl14/dinov2_vitl14.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1c997a103722ead6ba8993152e7d1a1670a2662ad552a43a6649c67660bf2efc
|
3 |
+
size 88358024
|