I just got an intern position in which I need to help them with serving inference requests on multiple GPUs. However, the available Github resource I could find are all about training.
Is there any example using multiple GPU to do inference in Tensorflow (python).
Related
I have a python code which uses keras and tensorflow backend. My system doesn't support training this model due to low memory space. I want to take use of Amazon sagemaker.
However all the tutorials I find are about deploying your model in docker containers. My model isn't trained and I want to train it on Amazon Sagemaker.
Is there a way to do this?
EDIT : Also can I make a script of my python code and run on it on AWS sagemaker?
SageMaker provides the capability for users to bring in their custom training scripts and train their algorithms using the script it on SageMaker using one of the pre-built containers for frameworks like TensorFlow, MXNet, PyTorch.
Please take a look at https://github.com/aws/amazon-sagemaker-examples/blob/master/frameworks/tensorflow/get_started_mnist_train.ipynb
It walks through how you can bring in your training script using TensorFlow and train it using SageMaker.
There are several other examples in the repository which will help you answer other questions you might have as you progress on with your SageMaker journey.
I want to fine tune ALBERT.
I see one can distribute neural net training over multiple gpus using tensorflow: https://www.tensorflow.org/guide/distributed_training
I was wondering if it's possible to distribute fine-tuning across both my laptop's gpu and a colab gpu?
I don't think that's possible. Because in order to do GPU distributed training, you need NVLinks among your GPUs. You don't have such a link between your laptop's GPU and Colab GPUs. This is a good read https://lambdalabs.com/blog/introduction-multi-gpu-multi-node-distributed-training-nccl-2-0/
I am currently working with neural networks in keras and I know that it works with tensorflow in the back-end, I have it installed on the GPU, but I don't know if keras uses the GPU or if it is something completely different from tensorflow.
TensorFlow is a mid-level framework that performs operations on tensors. Keras is a high-level API that simplifies the creation and training of neural networks. Keras doesn't do any of the tensor ops itself; it delegates those to its backend, which is a mid-level framework of your choosing: TensorFlow, CNTK, or Theano. Each of those frameworks can be configured to do the tensor ops in whatever ways they can (as far as I am aware, each of them can use either CPUs or GPUs). Keras, however, doesn't really care how the ops get done. It just tells the backend to do them, and they get done.
I've created a script to train a keras neural net and have run it successfully on my machine (at the end of training there is roughly 0.8 validation accuracy). However, when I try to run the exact same code (on the same data) on a Google Cloud VM instance I get drastically worse results (~0.2 validation accuracy).
Git status confirms that the repo in the VM is up to date with master (same with my local machine), and I have verified that its versions of tf and keras are up to date (and same as my local machine). I've also set the numpy and tensorflow random seeds before importing Keras.
Has anyone run into a problem like this before? I'm at a loss for what could be causing this... the only difference I can think of is that my machine is running Python 3.6 whereas the VM is running Python 2.7. Could that account for the vast difference is training results?
I found a buggy interaction between Keras and the Estimator API in tensorflow 1.10 (current gcloud version), but not in >=1.11 (what I was using locally).
Not sure if it applies to you (do you use Keras+Estimator and tensorflow >=1.11 for local?)
I filed a bug report here: https://github.com/tensorflow/tensorflow/issues/24299
I am testing the new Tensorflow Object Detection API in Python, and I succeeded in installing it on Windows using docker. However, my trained model (Faster RCNN resnet101 COCO) takes up to 15 seconds to make a prediction (with very good accuracy though), probably because I only use Tensorflow CPU.
My three questions are:
Considering the latency, where is the problem? I heard Faster RCNN was a good model for low latency visual detection, is it because of the CPU-only execution?
With such latency, is it possible to make efficient realtime video processing by using tensorflow GPU, or should I use a more popular model like YOLO?
The popular mean to use tensorflow GPU in docker is nvidia-docker but is not supported on windows. Should I continue to look for a docker (or conda) solution for local prediction, or should I deploy my model directly to a virtual instance with GPU (I am comfortable with Google Cloud Platform)?
Any advice and/or good practice concerning real-time video processing with Tensorflow is very welcome!
Considering the latency, where is the problem ? I heard Faster RCNN
was a good model for low latency visual detection, is it because of
the CPU-only execution ?
Of course, it's because you are using CPU.
With such latency, is it possible to make efficient realtime video
processing by using tensorflow GPU, or should I use a more popular
model like YOLO ?
Yolo is fast, but I once used it for face and accuracy was not that great. But a good alternative.
The popular mean to use tensorflow GPU in docker is nvidia-docker but
is not supported on windows. Should I continue to look for a docker
(or conda) solution for local prediction, or should I deploy my model
directly to a virtual instance with GPU (I am comfortable with Google
Cloud Platform) ?
I think you can still use your local GPU in windows, as Tensorflow supports GPU on python.
And here is an example, simply to do that. It has a client which can read webcam or IP cam stream. The server is using Tensorflow python GPU version and ready to use pre-trained model for predictions.
Unfortunately, Tensoflow does not support tensorflow-serving on windows. Also as you said Nvidia-Docker is not supported on windows. Bash on windows has no support for GPU either. So I think this is the only easy way to go for now.