YakumoAkayi commited on
Commit
dd80042
·
verified ·
1 Parent(s): 968e3aa

Agora board builds (Model Farm, verified on real boards)

Browse files
.gitattributes CHANGED
@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ targets/QCS8550/FP16/model.aidem filter=lfs diff=lfs merge=lfs -text
37
+ targets/QCS8550/W8A16/model.aidem filter=lfs diff=lfs merge=lfs -text
38
+ targets/QCS8550/W8A8/model.aidem filter=lfs diff=lfs merge=lfs -text
LICENSE_MODEL_FARM.pdf ADDED
Binary file (71.1 kB). View file
 
NOTICE.md ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ # Provenance
2
+
3
+ Resources are copied without binary changes from the APLUX Model Farm build `PP-HumanSegV2-Lite` that Agora verified on real boards (decoded sample check, then measured), pinned by sha256 in SHA256SUMS.
4
+
5
+ Agora arranges files under targets/<SoC>/<precision>/ and adds model.json. AidLite, AidGen, AidVoice and the Qualcomm QNN runtime are separate dependencies with their own terms.
README.md ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ license_name: aplux-model-farm
4
+ license_link: https://huggingface.co/APLUX-Agora/PP-HumanSegV2-Lite/blob/main/LICENSE_MODEL_FARM.pdf
5
+ pipeline_tag: image-segmentation
6
+ tags: [Agora, QNN, AIoT, Qualcomm]
7
+ ---
8
+
9
+ # PP-HumanSegV2-Lite model resources
10
+
11
+ PP-HumanSegV2-Lite,使用 Model Farm 官方预处理与样例,逐板卡验证。
12
+
13
+ Board builds that [Agora](https://mega.aidlux.com) runs and measures on real Qualcomm boards. Try it online, without a board: https://mega.aidlux.com/models/pp-humansegv2-lite
14
+
15
+ | SoC · board | Precisions | Measured on Agora (default precision) |
16
+ | --- | --- | --- |
17
+ | QCS6490 · 犀牛派 A1 (Rhino Pi A1) | W8A16, W8A8 | 2.3 ms |
18
+ | QCS8550 · 犀牛派 X1 (Rhino Pi X1) | FP16, W8A16, W8A8 | 0.9 ms |
19
+
20
+ Task: 分割. Inputs: image (image). Outputs: masks (masks).
21
+
22
+ ## Download
23
+
24
+ ```bash
25
+ export AGORA_WORKSPACE="$HOME/my-pp-humansegv2-lite"
26
+ export AGORA_SOC='QCS8550'
27
+ hf download APLUX-Agora/PP-HumanSegV2-Lite \
28
+ --include "targets/$AGORA_SOC/*" --local-dir "$AGORA_WORKSPACE/resources"
29
+ ```
30
+
31
+ Each `targets/<SoC>/` holds `model.json` (default precision and variant paths) and one directory per precision. Context binaries are named `model.bin`; multi-file packages are kept as published by Model Farm.
32
+
33
+ Pre- and post-processing, the runner and an exported project: open the model on Agora and choose “部署指南”, or export any workflow that uses it.
34
+
35
+ ## Licence and provenance
36
+
37
+ Source model licence: see the original model's terms.
38
+ Converted, deployable resources: APLUX Model Farm License (LICENSE_MODEL_FARM.pdf). See NOTICE.md and SHA256SUMS.
SHA256SUMS ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ 3be4bfd3639b9f1b6b54209ec172e9a56f042864dd877a31034d5e023349504c targets/QCS6490/W8A16/model.bin
2
+ f629009fdbc15d675c77af8bb610c3124c8d785b3981e47d690b378ae843061a targets/QCS6490/W8A8/model.bin
3
+ d1ad5e9a397d481ed661f4fa3297ce663594fb2ebe9a2d1ee4b073eca886f163 targets/QCS8550/FP16/model.aidem
4
+ ed885a02c858ae6e3101eb0860cb1f898f7157bb8b6ac41b3c6f4c71cb856e67 targets/QCS8550/W8A16/model.aidem
5
+ dbdadd155db07bd72a2e68cc865a42e1cc7bb848af8779dcc239a3505b2bcdb4 targets/QCS8550/W8A8/model.aidem
targets/QCS6490/W8A16/model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3be4bfd3639b9f1b6b54209ec172e9a56f042864dd877a31034d5e023349504c
3
+ size 6142176
targets/QCS6490/W8A8/model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f629009fdbc15d675c77af8bb610c3124c8d785b3981e47d690b378ae843061a
3
+ size 6117600
targets/QCS6490/model.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "soc": "QCS6490",
3
+ "default_precision": "W8A16",
4
+ "variants": {
5
+ "W8A16": "W8A16/model.bin",
6
+ "W8A8": "W8A8/model.bin"
7
+ }
8
+ }
targets/QCS8550/FP16/model.aidem ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1ad5e9a397d481ed661f4fa3297ce663594fb2ebe9a2d1ee4b073eca886f163
3
+ size 8214680
targets/QCS8550/W8A16/model.aidem ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed885a02c858ae6e3101eb0860cb1f898f7157bb8b6ac41b3c6f4c71cb856e67
3
+ size 4183032
targets/QCS8550/W8A8/model.aidem ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbdadd155db07bd72a2e68cc865a42e1cc7bb848af8779dcc239a3505b2bcdb4
3
+ size 3941368
targets/QCS8550/model.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "soc": "QCS8550",
3
+ "default_precision": "FP16",
4
+ "variants": {
5
+ "FP16": "FP16/model.aidem",
6
+ "W8A16": "W8A16/model.aidem",
7
+ "W8A8": "W8A8/model.aidem"
8
+ }
9
+ }