graphpipe-go
graphpipe-go copied to clipboard
oracle
GraphPipe for go
On master branch, while the tensorflow go has adopted version 1.11.0, in the Makefile the libtensorflow is still using version 1.8.0: https://github.com/oracle/graphpipe-go/blob/master/cmd/graphpipe-tf/Makefile#L151
If Configure File(https://www.tensorflow.org/tfx/serving/serving_config)? Or how to serve two or more models?
save graph def ``` import tensorflow as tf with tf.Graph().as_default() as g: a = tf.placeholder("double", name="a", shape=[2,2]) b = tf.placeholder("double", name="b", shape=[2,2]) y = tf.multiply(a, b) tf.train.write_graph(g, "./", "ma_plus_mb.pb", as_text=False)...
When I try to load mobilenet to serve on graphpipe--it says BNOperator not found
Step 9/12 : COPY lib/libcaffe2.so /usr/local/lib/libcaffe2.so fails with the message: COPY failed: stat /var/lib/docker/tmp/docker-builder534911080/lib/libcaffe2.so: no such file or directory make: *** [runtime-container] Error 1
my run code: `docker run -it --rm -e https_proxy=${https_proxy} -p 9000:9000 -v "$PWD:/models/" sleepsonthefloor/graphpipe-onnx:cpu --model=/models/simple_pose.onnx --value-inputs=/models/input.json --listen=0.0.0.0:9000` error:  Hi, when I deploy onnx model, the error occured: Don't konw...
Hi, I have a onnx model that I am trying to serve with sleepsonthefloor/graphpipe-onnx:cpu docker image. I am running it with this command: `docker run -it -v "$PWD/models:/models/" -p 9000:9000...
