search

dusty-nv / jetson-inference

Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.
https://developer.nvidia.com/embedded/twodaystoademo
MIT License
7.82k stars 2.98k forks source link

Retraining SSD Mobilenet Issue -Detectnet failed to initialize #781

Closed irfwas closed 1 year ago

irfwas commented 4 years ago

Hello @dusty-nv,

I have two issues:

  1. Trying to train for 100 epochs, the system shows an error due to low memorry and force quits at ~50 epochs. I tried to resume but there is no clear information on the code on how to resume from this issue. Could you kindly provide me the code or steps to resume back at the 50 epoch mark?

  2. I converted my file to an onxx file and tried to run the file but i got the error shown below. Now, this error occurs for any detection dataset whether its SSD Mobilenet V2, SSD Inception V2 or others, although it worked before I tried to retrain the network.

:~/jetson-inference/python/training/detection/ssd$ detectnet.py --model=models/adas/sss-mobilenet.onnx --labels=models/adas/labels.txt --input=blob=input_0 --output-cvg=scores --output-bbox=boxes csi://0

jetson.inference -- detectNet loading network using argv command line params

detectNet -- loading detection network model from: -- prototxt NULL -- model models/adas/sss-mobilenet.onnx -- input_blob 'data' -- output_cvg 'scores' -- output_bbox 'boxes' -- mean_pixel 0.000000 -- mean_binary NULL -- class_labels models/adas/labels.txt -- threshold 0.500000 -- batch_size 1

[TRT] TensorRT version 7.1.3 [TRT] loading NVIDIA plugins... [TRT] Registered plugin creator - ::GridAnchor_TRT version 1 [TRT] Registered plugin creator - ::NMS_TRT version 1 [TRT] Registered plugin creator - ::Reorg_TRT version 1 [TRT] Registered plugin creator - ::Region_TRT version 1 [TRT] Registered plugin creator - ::Clip_TRT version 1 [TRT] Registered plugin creator - ::LReLU_TRT version 1 [TRT] Registered plugin creator - ::PriorBox_TRT version 1 [TRT] Registered plugin creator - ::Normalize_TRT version 1 [TRT] Registered plugin creator - ::RPROI_TRT version 1 [TRT] Registered plugin creator - ::BatchedNMS_TRT version 1 [TRT] Could not register plugin creator - ::FlattenConcat_TRT version 1 [TRT] Registered plugin creator - ::CropAndResize version 1 [TRT] Registered plugin creator - ::DetectionLayer_TRT version 1 [TRT] Registered plugin creator - ::Proposal version 1 [TRT] Registered plugin creator - ::ProposalLayer_TRT version 1 [TRT] Registered plugin creator - ::PyramidROIAlign_TRT version 1 [TRT] Registered plugin creator - ::ResizeNearest_TRT version 1 [TRT] Registered plugin creator - ::Split version 1 [TRT] Registered plugin creator - ::SpecialSlice_TRT version 1 [TRT] Registered plugin creator - ::InstanceNormalization_TRT version 1 [TRT] detected model format - ONNX (extension '.onnx') [TRT] desired precision specified for GPU: FASTEST [TRT] requested fasted precision for device GPU without providing valid calibrator, disabling INT8 [TRT] native precisions detected for GPU: FP32, FP16 [TRT] selecting fastest native precision for GPU: FP16 [TRT] attempting to open engine cache file .1.1.7103.GPU.FP16.engine [TRT] cache file not found, profiling network model on device GPU

error: model file 'models/adas/sss-mobilenet.onnx' was not found. if loading a built-in model, maybe it wasn't downloaded before.

    Run the Model Downloader tool again and select it for download:

       $ cd <jetson-inference>/tools
       $ ./download-models.sh

[TRT] detectNet -- failed to initialize. jetson.inference -- detectNet failed to load network Traceback (most recent call last): File "/usr/local/bin/detectnet.py", line 51, in net = jetson.inference.detectNet(opt.network, sys.argv, opt.threshold) Exception: jetson.inference -- detectNet failed to load network

dusty-nv commented 3 years ago

Here is a link to the forum post discussing this issue: https://forums.developer.nvidia.com/t/how-to-utilize-class-id-in-object-detection-to-control-gpio-pins/158433/4?u=dusty_nv

ihebchh commented 2 years ago

Hello @dusty-nv, I have a problem with detectnet : [TRT] detectNet -- failed to initialize. jetson.inference -- detectNet failed to load network thanks.

dusty-nv commented 2 years ago

error: model file 'models/adas/sss-mobilenet.onnx' was not found.

It cannot find the path to your model. Are you sure you don't mean ssd-mobilenet.onnx instead of sss-mobilenet.onnx?

ihebchh commented 2 years ago

Thank you for help it works correctly now. I will use an ip camera for this application but the streaming is not in real time there is a late of more than 2 seconds. I'm using this command: detecnet --model=models/person/ssd-mobilenet.onnx --labels=models/person/labels.txt --input_blob=input_0 --output-cvg=scores --output-bbox=boxes @.***:554 Please help Thanks

Le mar. 15 mars 2022 à 14:55, Dustin Franklin @.***> a écrit :

error: model file 'models/adas/sss-mobilenet.onnx' was not found.

It cannot find the path to your model. Are you sure you don't mean ssd-mobilenet.onnx instead of sss-mobilenet.onnx?

— Reply to this email directly, view it on GitHub https://github.com/dusty-nv/jetson-inference/issues/781#issuecomment-1068016183, or unsubscribe https://github.com/notifications/unsubscribe-auth/ARLKILS37EQP26SUJTMZXQDVACJDLANCNFSM4TEWRKYQ . You are receiving this because you commented.Message ID: @.***>