Merge pull request #33 from mlcommons/poc

Poc docs cleanups

.github/workflows/test-mlperf-inference-abtf-poc.yml

@@ -37,7 +37,7 @@ jobs:
         cm run script --quiet --tags=get,sys-utils-cm
     - name: Test MLPerf Inference ABTF POC using ${{ matrix.backend }} on docker
       run: |
-        cm run script --tags=run-abtf,inference,_poc-demo --adr.compiler.tags=gcc --quiet --docker --docker_it=no  -v --gh_token=${{ secrets.ABTF_ACCESS_TOKEN }}
+        cm run script --tags=run-abtf,inference,_poc-demo --adr.compiler.tags=gcc --quiet --docker --docker_it=no  -v 
 
   build_macos:
     runs-on: ${{ matrix.os }}
@@ -66,7 +66,7 @@ jobs:
         cm pull repo --url=${{ github.event.pull_request.head.repo.html_url }} --checkout=${{ github.event.pull_request.head.ref }}
     - name: Test MLPerf Inference ABTF POC using ${{ matrix.backend }} natively
       run: |
-        cm run script --tags=run-abtf,inference,_poc-demo --adr.compiler.tags=gcc --quiet --env.CM_MLPERF_LOADGEN_BUILD_FROM_SRC=off  -v --gh_token=${{ secrets.ABTF_ACCESS_TOKEN }}
+        cm run script --tags=run-abtf,inference,_poc-demo --quiet --env.CM_MLPERF_LOADGEN_BUILD_FROM_SRC=off  -v
 
   build_windows:
     runs-on: ${{ matrix.os }}
@@ -91,7 +91,7 @@ jobs:
         cm pull repo --url=${{ github.event.pull_request.head.repo.html_url }} --checkout=${{ github.event.pull_request.head.ref }}
     - name: Test MLPerf Inference ABTF POC using ${{ matrix.backend }} natively
       run: |
-        cm run script --tags=run-abtf,inference,_poc-demo --quiet --env.CM_MLPERF_LOADGEN_BUILD_FROM_SRC=off --adr.raw-dataset-mlcommons-cognata.tags=_gdrive -v --gh_token=${{ secrets.ABTF_ACCESS_TOKEN }}
+        cm run script --tags=run-abtf,inference,_poc-demo --quiet --env.CM_MLPERF_LOADGEN_BUILD_FROM_SRC=off --adr.raw-dataset-mlcommons-cognata.tags=_gdrive -v 
 
     
 

docs/index.md

@@ -2,11 +2,11 @@
 
 [MLCommons](https://mlcommons.org/) and [AVCC](https://avcc.org/) have jointly defined and developed the MLPerf: Inference Automotive POC which the first release and subset of the MLPerf: Inference Automotive benchmark suite. The benchmark suite has been made possbile with MLCommons expertise in AI benchmarks and AVCC's expertise in automotive. Note that as this is a POC, not the full benchmark suite, there will be no official MLCommons submissions. Find the POC reference block diagram below.
 
-<img src="../img/ABTF-POC_Intended_Use_1_Graphics.svg" alt="POC" width="500">
+<img src="img/ABTF-POC_Intended_Use_1_Graphics.svg" alt="POC" width="500">
 
 For ease of use, many of the benchmark components are packaged in a container which, e.g., reduces the software requirements on the computer running the benchmark. Please note that the POC reference is **not for performance** purposes; it is a reference to enable anyone to familiarize themselves with the benchmark and to give an understanding of how it works. For performant versions of the POC each HW/SW vendor will have their own optimized version of the model for their target hardware.
 
-To try out the POC reference yourself, please visit the [Run](/run/index.md) section. The POC reference only has 20 frames, to see the object detection model run on an entire sequence, follow this link:
+To try out the POC reference yourself, please visit the [Run](run/index.md) section. The POC reference only has 20 frames, to see the object detection model run on an entire sequence, follow this link:
 
 * TODO: Add MLCommons YouTube clip
 

docs/run/index.md

@@ -39,21 +39,35 @@ Note: if you're running Ubuntu, CM automatically installs Docker when running th
 
 By running the script below, you are downloading the POC container and dataset, then launching the benchmark. All in one command!
 
-```
-cm run script --tags=run-abtf,_poc-demo \
-  --quiet \
-  --docker \
-  --gh_token=<GH_TOKEN> \
-  --docker_cache=no
-```
-!!! tip
-    * Use `--rerun` to force overwrite the previously generated results
-    * Use `--env.CM_MLPERF_LOADGEN_BUILD_FROM_SRC=off` to use the prebuilt MLPerf Loadgen binary and not do a source compilation
-    * Use `--docker_os=[rhel|arch|ubuntu]` to change the docker OS
-    * Use `--docker_os_version=[8|9]` for `RHEL`, `[24.04,22.04,20.04]` for `ubuntu` and `[latest]` for `arch`  
-    * Use `--docker_base_image=[IMAGE_NAME]` to override the default base image for docker
-    * Skip `--docker` to do the run on the host machine without using a docker
-    * Github actions for this run can be seen [here](https://github.com/mlcommons/cm4abtf/actions/workflows/test-mlperf-inference-abtf-poc.yml)
+=== "Docker"
+    ### Docker
+
+    ```bash
+    cm run script --tags=run-abtf,_poc-demo --quiet \
+     --docker --docker_cache=no
+    ```
+
+    !!! tip
+        * Use `--rerun` to force overwrite the previously generated results
+        * Use `--env.CM_MLPERF_LOADGEN_BUILD_FROM_SRC=off` to use the prebuilt MLPerf Loadgen binary and not do a source compilation
+        * Use `--docker_os=[rhel|arch|ubuntu]` to change the docker OS
+        * Use `--docker_os_version=[8|9]` for `RHEL`, `[24.04|22.04|20.04]` for `ubuntu` and `[latest]` for `arch`  
+        * Use `--docker_base_image=[IMAGE_NAME]` to override the default base image for docker
+        * Github actions for this run can be seen [here](https://github.com/mlcommons/cm4abtf/actions/workflows/test-mlperf-inference-abtf-poc.yml)
+    We have tested the workflow via docker on Ubuntu 24.04, Ubuntu 22.04, Ubuntu 20.04 and macOS 14
+
+=== "Native"
+    ### Native
+
+    ```
+    cm run script --tags=run-abtf,_poc-demo --quiet
+    ```
+    !!! tip
+        * Use `--rerun` to force overwrite the previously generated results
+        * Use `--env.CM_MLPERF_LOADGEN_BUILD_FROM_SRC=off` to use the prebuilt MLPerf Loadgen binary and not do a source compilation
+        * Use `--adr.raw-dataset-mlcommons-cognata.tags=_gdrive` to download the dataset from gdrive if there is any issue with rclone
+        * Github actions for this run can be seen [here](https://github.com/mlcommons/cm4abtf/actions/workflows/test-mlperf-inference-abtf-poc.yml)
+    We have tested the workflow natively on Ubuntu 24.04, Ubuntu 22.04, Ubuntu 20.04, RHEL 9, macOS 13, macOS 14 and Windows Server 2022
 
 Depending on the computer used and internet connection, this can take a few minutes.