graphpipe
graphpipe copied to clipboard
oracle
Machine Learning Model Deployment Made Simple
I have combinatorial optimization code in go designed to consume flatbuffers models. I'm researching how I can integrate it into a larger platform, hosted in kubernetes, so as to help...
1 - GarpePipe using around 7GB of RAM memory after the first request (model size is 500mb) 2- I didn't saw any function for multiple models server. For multi-models, running...
The documentation states that there are "Efficient client implementations in Go, Python, and Java" I don't see where the Java implementation is. Has the project not been made public yet?...
Under normal circumstances, a mtcnn face detection model takes about 0.x seconds, I deploy with graphpipe, then send a picture, it takes 30s to receive a reply, why is it...
Hi, when I deploy onnx model, the error occured: Don't konw how to map unexcepted argument dilations( form operator ConvTranspose) could you please tell me how to fix the problem...
Hello! I want to use `graphpipe` with sklearn model, converted by [this](https://github.com/onnx/onnxmltools/blob/master/onnxmltools/convert/sklearn/convert.py) I can save `model.onnx`, but what can I use for `model-inputs`? I'm trying to save my model as...
requests.exceptions.ConnectionError: HTTPConnectionPool(host='localhost', port=5000): Max retries exceeded with url: / (Caused by NewConnectionError(': Failed to establish a new connection: [WinError 10061] No connection could be made because the target machine actively...
ssd yolo
Can deployments use the yolo SSD faster- RCNN model ? @termie @sleepsonthefloor @Djelibeybi @vishvananda @tjfontaine
Just like the common object detection, three phases are included for a whole cycle. pre-processing: read image, resize or crop (run on cpu) prediction: load inputs, run on GPU, and...
