I am trying to use the Jetson Nano for running an inference using a Mask-Rcnn model with Mobile-net Backbone.
The packages I have installed are: Tensorflow-Gpu 1.14.0, Cuda 10.0.32, Keras 2.1.6, Numpy - 1.16.1, scipy 0.19.1.
Faced lot of difficulties in installing the packages that are compatible with eachother. If possible please give me a proper set of every packages with their versions compatible, to run a Deep-learning model in Jetson Nano.
For now, This is the error I am stuck with.
Traceback (most recent call last):
File “/home/safran/Desktop/fusion/process_video.py”, line 17, in
results = model.detect([frame], verbose=0)
File “/home/safran/Desktop/fusion/model.py”, line 2862, in detect
self.keras_model.predict([molded_images, image_metas, anchors], verbose=0)
File “/usr/local/lib/python3.6/dist-packages/keras/engine/training.py”, line 1462, in predict
File “/usr/local/lib/python3.6/dist-packages/keras/engine/training_arrays.py”, line 324, in predict_loop
batch_outs = f(ins_batch)
File “/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/backend.py”, line 3292, in call
File “/usr/local/lib/python3.6/dist-packages/tensorflow/python/client/session.py”, line 1458, in call
tensorflow.python.framework.errors_impl.UnknownError: 2 root error(s) found.
(0) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
(1) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
0 successful operations.
0 derived errors ignored.
Please provide me with a solution to move forward.