iOS: Can't merge user_target_xcconfig for pod targets: ["MediaPipeTasksCommon", "MediaPipeTasksGenAIC"]. #5302
Comments
|
Hi @DarrylBayliss, Could you please provide additional details about the issue, including the complete steps you are following from our documentation? This information will greatly assist us in better understanding the issue and potentially reproducing it if necessary. Thank you!! |
|
Hi @kuaashish, I followed the setup instructions from these pages:
Some additional things I checked:
I can provide a sample project if neccessary. |
google-ml-butler
bot
removed
the
stat:awaiting response
|
Could you please have look into this issue? Thank you!! |
|
Hi @DarrylBayliss, We acknowledge that utilizing any two or more of our features within an app across both Android and iOS platforms is currently not feasible. This issue may become a priority for us in the near future, as you are encountering errors in this scenario. Thank you!! |
kuaashish
added
stat:awaiting response
|
Hey @kuaashish, Thanks for the update. This might be the case for iOS, I'm not sure it is correct in the case of Android. I have an Android app where I am doing something similar and using the genai / vision task dependencies together in the same project. I have an example incase it is helpful with this issue: https://github.com/DarrylBayliss/Simon-Says-Android |
google-ml-butler
bot
removed
the
stat:awaiting response
|
Hi @DarrylBayliss, following internal discussions, we have identified a conflict between LLM inference and the Vision Task, as the latter relies on OpenCV, which GenAI does not relies. This has resulted in an issue when attempting to integrate both solutions into an application. Rest assured, we have brought this matter to the attention of our team and will keep you informed as soon as any updates are available. For now, we are categorizing this as a feature request and sharing it with our team. Thank you!! |
kuaashish
added
task::all
|
Encoutered the same issue after added LLM and Vision Tasks in the pod file (with the latest LLM 0.10.15 and Vision 0.10.14. It has been 5 months since the opening date of this issue, any progress? |
Some small but important changes: 1. Due to the conflict of LLM and Object Detection framework, we disabled the LLM temporarily. google-ai-edge/mediapipe#5302 (comment) 2. Set CADisableMinimumFrameDurationOnPhone to true to run the KMP app on iOS real device. https://youtrack.jetbrains.com/issue/CMP-5643/Make-CADisableMinimumFrameDurationOnPhone-a-requirement-to-run-Compose-Multiplatform-on-iOS 3. Compose Multiplatform 1.7.0-beta02 had an issue that caused the some extra white space on iPhones like 13mini. Upgrade to 1.7.0-rc01 to avoid that. https://youtrack.jetbrains.com/issue/CMP-6690
|
We have an internal tracking bug for the issue b/378116539 but no concrete update at the moment. Will keep people posted if there is any updates. |
yuhuichen1015
removed
the
stat:awaiting googler
OS Platform and Distribution
iOS 17.2
Compiler version
No response
Programming Language and version
Swift 5
Installed using virtualenv? pip? Conda?(if python)
No response
MediaPipe version
0.10.12
Bazel version
No response
XCode and Tulsi versions (if iOS)
Xcode 15.2
Android SDK and NDK versions (if android)
No response
Android AAR (if android)
None
OpenCV version (if running on desktop)
No response
Describe the problem
Create podfile, run pod install, run build in xcode against simulator
Complete Logs
When I then proceed to build or run the project using the simulator I recieve the following build error.
My Podfile looks like the following:
I assume there's a conflict somewhere between the Common framework and the AIC framework. I'm completely clueless about this though.
The text was updated successfully, but these errors were encountered: