UbiquitousLearning
Android app crash while Image Reading
Fail To Load Models! Please Check if models exists at /sdcard/Download/model and restart but i have copied model in that location. still getting same error message what would be the reason.
Hi,
- Currently the android lib is only shipped with LLaMA and Fuyu support, please make sure your model files are supported.
- Could you double check the file is renamed properly according to This notice? it would be better if you attach a screenshot or else to show the file tree structure of your
/sdcard/Download/model.
while loading fuyu app is getting crashed getting below errors:
FATAL EXCEPTION: main Process: org.saltedfish.chatbot, PID: 10488 java.lang.IllegalArgumentException: Software rendering doesn't support hardware bitmaps
Process is going to kill itself! java.lang.Exception at android.os.Process.killProcess(Process.java:1356) at com.android.internal.os.RuntimeInit$KillApplicationHandler.uncaughtException(RuntimeInit.java:174) at java.lang.ThreadGroup.uncaughtException(ThreadGroup.java:1071) at java.lang.ThreadGroup.uncaughtException(ThreadGroup.java:1066) at java.lang.Thread.dispatchUncaughtException(Thread.java:2306)
Uh... It seems something wrong about the UI framework. What kind of device you are testing on?
Uh... It seems something wrong about the UI framework. What kind of device you are testing on?
Snapdragon 8 gen3 (xiaomi14)
Opps... It seems fine on our testing devices (of same model). We may need some time to investigate it.😂
Opps... It seems fine on our testing devices (of same model). We may need some time to investigate it.😂
okay, Especially it is failing ImageReader Mode.
Hi, we tried to reproduce the bug but sadly failed. Could you double check the model file tree?
hi @lx200916 , could able to load the model , but app is getting crashed after passing image.
Fatal signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0x0 in tid 7578 (RenderThread), pid 7538 (tedfish.chatbot)
Understood. Could you attach the image file here? I guess maybe something wrong about the file size or PreProcessors.
The input image needs to be 3-channel (RGB), @Vinaysukhesh98 you need to check if it is 4-channel (RGBA)
Understood. Could you attach the image file here? I guess maybe something wrong about the file size or PreProcessors.
https://github.com/UbiquitousLearning/mllm/blob/main/assets/dog_image.jpg
The input image needs to be 3-channel (RGB), @Vinaysukhesh98 you need to check if it is 4-channel (RGBA)
can you please help me with more info or any steps to resolve.