I am trying to experiment with Keras, and thought I would create my own layer. However, I am using an AMD GPU (PlaidML) without Tensorflow. I have seen some tutorials online, but they all require Tensorflow. How would I create a custom Keras layer without using Tensorflow?
Related
There are manual tools like Makesense.ai to create the labelled data for custom training YOLO, but is there any method for automatic labelling of multiple objects in an image so that the image labelling process would be faster?
Makesense.ai has the very feature you are asking for. You can use a network trained on coco or provide your own yolov5 tensorflow js model.
Is there a way to test out the pretrained image classification models released by Google called 'MobileNets' using only the Keras API?
Models like ResNet50 and InceptionV3 are already available as Keras Applications, but I couldn't find documentation on using custom tensorflow models with Keras. Thanks in advance.
AFAIK, there is no direct access via Keras API to it. However, you can find a good interaction of Keras and MobileNets
here.
I'm working on a project that requires object detection and recognition on images fed by a live camera.
For it to work well it should perform the evaluation for a single frame as quick as possible.
Going straight for TensorFlow instead of using Keras with TF backend will improve the performance of the whole model during the evaluation?
I'm using a bunch of images to train my Tensorflow Image recognition project using this tutorial https://codelabs.developers.google.com/codelabs/tensorflow-for-poets/index.html#4
Actually I need a lot of Cpu to train my model and it takes a lot of time on my laptop.
I have registered a Google ML account and started this tutorial:
https://cloud.google.com/ml/docs/quickstarts/training
Everything is set up and running but this is for mnist sample code. There is no image_retraining sample code like the retrain.py from tensorflow.
Looking for some examples on how to to run the Tensorflow Image Recognition script retrain in Google ML.
After training lenet model in caffe framework using 10k images,i got the model lenet_iter_4000.caffemodel which contains weights and baises. I did in caffe for predicting test image classification, Now i wanted to do classification in OpenCV by loading this caffemodel for test image, Can anybody please help me how to combine caffe and OpenCV for predicting new image..
OpenCV contrib contains a module called dnn that can be used for this, it can load Caffe and Torch models, and here is a tutorial for GoogleNet, you can easily adapt it to use another network, the code is basically the same.
An alternative is the classification.cpp example in Cafee's source, which uses OpenCV to read an image and process it with Caffe.