Is it possible to possible to use a pretrained weight of a large model (for example ResNet151) and cut a part out of it to fit a custom, rather small model?
If it is the case, how can it be managed?
Thanks for any ideas!
Related
Am a complete noob and I just want to ask, do we need to preprocess images (like manually) before feeding the image into CNN for training? I've read that CNN already has some filtering techniques to extract features and such. I'm thinking what if all the train images are binary images or even just edges (teaching the model for shapes), is it advisable or I'll just feed grayscale images? Additionally, if the answer is yes, may I know what kind of preparation is done normally or what you would use?
I am aware of the other preprocessing techniques by Keras, such as the VGG16, but I would like something simple and manual.
The preprocessing functions of Keras only preprocess the input according to the state-of-the-art models requirements, e.g., changing the data format etc. May be, what you are talking about is hand-engineered features, which is not simple. And I think, it's not advisable because CNN does a better job. But it may also depend on what you actually want to do.
I want to build a simple image detector for custom Binary shapes on images.
I may train and use the models on object detection zoo such as ssd_inception_v2 and so on. But it's would be extremely un efficient as it has sizes in hundreds of Megabytes.
and I can't even imagine to use that in my simple app. can anybody suggest me how to solve this?
I have already built excellent small size classifiers for my images. but can't build small scale efficient detector. (their position with detection boxes)
I think what you need is transfer learning. I would take one of the lightweight models such as MobileNetV2 and retrain on my dataset. It should be pretty quick.If you want to even decrease your model size further, feel free to only take the first few layers of the CNN and retrain it. It would be a bit more work since you need to re-write the part of network you want to use and load it with the pre-trained weights.
SCENARIO
What if my intention is to train for a dataset of medical images and I have chosen a coco pre-trained model.
My Doubts
1 Since I have chosen medical images there is no point of train it on COCO dataset, right? if so what is a possible solution to do the same?
2 Adding more layers to a pre-trained model will screw the entire model? with classes of around 10 plus and 10000's of training datasets?
3 Without train from scratch what are the possible solutions , like fine-tuning the model?
PS - let's assume this scenario is based on deploying the model for business purposes.
Thanks-
Yes, it is a good idea to reuse the Pre-Trained Models or Transfer Learning in Real World Projects, as it saves Computation Time and as the Architectures are proven.
If your use case is to classify the Medical Images, that is, Image Classification, then
Since I have chosen medical images there is no point of train it on
COCO dataset, right? if so what is a possible solution to do the same?
Yes, COCO Dataset is not a good idea for Image Classification as it is efficient for Object Detection. You can reuse VGGNet or ResNet or Inception Net or EfficientNet. For more information, refer TF HUB Modules.
Adding more layers to a pre-trained model will screw the entire model?
with classes of around 10 plus and 10000's of training datasets?
No. We can remove the Top Layer of the Pre-Trained Model and can add our Custom Layers, without affecting the performance of the Pre-Trained Model.
Without train from scratch what are the possible solutions , like
fine-tuning the model?
In addition to using the Pre-Trained Models, you can Tune the Hyper-Parameters of the Model (Custom Layers added by you) using HParams of Tensorboard.
I have a dataset consist of around 5000 categories of images, but the number of images of every category varies from 20 to 2000, which is quite unbalanced. Also, the number of images are far from enough to train a model from scratch. I decided to do finetuning on pretrained models, like Inception models.
But I am not sure about how to deal with unbalanced data. There are several possible approaches:
Oversampling: Oversample the minority category. But even with aggressive image augmentation technique, we may not be able to deal with overfit.
Also, how to generate balanced batches from unbalanced dataset over so many categories? Do you have some ideas about this pipeline mechanism with TensorFlow?
SMOTE: I think it is not so effective for high dimensional signals like images.
Put weight on cross entropy loss in every batch. This might be useful for single batch, but cannot deal with the overall unbalance.
Any ideas about this? Any feedback will be appreciated.
Use tf.losses.softmax_cross_entropy and set weights for each class inversely proportional to their training frequency to "balance" the optimization.
Start with the pre-trained ImageNet layers, add your own final layers (with appropriate convolution, drop out and flatten layers as required). Freeze all but last few of the ImageNet layers, then train on your dataset.
For unbalanced data (and in general small datasets), use data augmentation to create more training images. Keras has this functionality built-in: Building powerful image classification models using very little data
Currently, I am training a model for image detection, and I want to know how many images do I need per class, do i need to have the same numbers of each object.
Please i need some advice.
I use Tensorflow, and Yolo v2 model.
Thanks,
You need as many as you can get, but definitely in the order of tens of thousands at least if you're training the network from scratch (there are pre-trained weights for YOLOv2 trained on
- http://host.robots.ox.ac.uk/pascal/VOC/).
It's best to have balanced classes, meaning the number of images for each class must be close, it's easier to train this way.
Why are you training the network yourself? Can't you use some pre-trained models, drop the FC layers and insert your own classes? This way it's much faster, and you don't need that many images.