Datasets:

omnineura

/

Dataset11

like 0

What is the name argument in a Sequential model?

Sequential constructor accepts a name argument, just like any layer or model in Ker

Keras: Sequential Model Keras: Sequential Model The Sequential model The Sequential model A Sequential model is appropriate for a plain stack of layers where each layer has exactly one input tensor and one output tensor. Equivalent to: A Sequential model is not appropriate when: Your model has multiple inputs or multiple outputs Any of your layers has multiple inputs or multiple outputs You need to do layer sharing You want non-linear topology Creating a Sequential model Creating a Sequential model You can create a Sequential model by passing a list of layers to the Sequential constructor: Its layers are accessible via the layers attribute: model.layers You can also create a Sequential model incrementally via the add() method: There's also a corresponding pop() method to remove layers: a Sequential model behaves very much like a list of layers. Sequential constructor accepts a name argument, just like any layer or model in Keras

What is useful to annotate TensorBoard graphs with semantically meaningful

This is useful to annotate TensorBoard graphs with semantically meaningful

This is useful to annotate TensorBoard graphs with semantically meaningful names. Specifying the input shape in advance Specifying the input shape in advance Generally, all layers in Keras need to know the shape of their inputs in order to be able to create their weights

What does this apply to?

Sequential models

So when you create a layer like this, initially, it has no weights: It creates its weights the first time it is called on an input, since the shape of the weights depends on the shape of the inputs: Naturally, this also applies to Sequential models

What happens when a Sequential model is not built?

instantiate a Sequential model without an input shape

When you instantiate a Sequential model without an input shape, it isn't "built": it has no weights (and calling model.weights results in an error stating just this)

What is the current output shape of a Sequential model?

the summary of the model so far, including the current output shape

The weights are created when the model first sees some input data: Once a model is "built", you can call its summary() method to display its contents: model.summary() However, it can be very useful when building a Sequential model incrementally to be able to display the summary of the model so far, including the current output shape

What is a common debugging workflow?

add() + summary()

In this case, you should start your model by passing an Input object to your model, so that it knows its input shape from the start: Note that the Input object is not displayed as part of model.layers, since it isn't a layer: model.layers A simple alternative is to just pass an input_shape argument to your first layer: Models built with a predefined input shape like this always have weights (even before seeing any data) and always have a defined output shape. In general, it's a recommended best practice to always specify the input shape of a Sequential model in advance if you know what it is. A common debugging workflow: add() + summary() A common debugging workflow: add() + summary() When building a new Sequential architecture, it's useful to incrementally stack layers with add() and frequently print model summaries

What does Feature extraction do once a Sequential model behave like?

a Functional API model

For instance, this enables you to monitor how a stack of Conv2D and MaxPooling2D layers is downsampling image feature maps: What to do once you have a model What to do once you have a model Once your model architecture is ready, you will want to: Train your model, evaluate it, and run inference. Save your model to disk and restore it. Speed up model training by leveraging multiple GPUs. Feature extraction with a Sequential model Feature extraction with a Sequential model Once a Sequential model has been built, it behaves like a Functional API model

What does this mean?

every layer has an input and output attribute

This means that every layer has an input and output attribute

What can be done to create a model that extracts the outputs of all intermediate layers

quickly creating a model that extracts the outputs of all intermediate layers in a Sequ

These attributes can be used to do neat things, like quickly creating a model that extracts the outputs of all intermediate layers in a Sequential model: Here's a similar example that only extract features from one layer: Transfer learning & fine-tuning Transfer learning & fine-tuning Transfer learning consists of taking features learned on one problem, and leveraging them on a new, similar problem. For instance, features from a model that has learned to identify racoons may be useful to kick-start a model meant to identify tanukis. Transfer learning is usually done for tasks where your dataset has too little data to train a full-scale model from scratch. The most common incarnation of transfer learning in the context of deep learning is the following worfklow: Take layers from a previously trained model. Freeze them, so as to avoid destroying any of the information they contain during future training rounds. Add some new, trainable layers on top of the frozen layers

What is the list of those that aren't meant to be trained?

non_trainable_weights

They will learn to turn the old features into predictions on a new dataset. Train the new layers on your dataset. A last, optional step, is fine-tuning, which consists of unfreezing the entire model you obtained above (or part of it), and re-training it on the new data with a very low learning rate. This can potentially achieve meaningful improvements, by incrementally adapting the pretrained features to the new data. Freezing layers: understanding the trainable attribute Freezing layers: understanding the trainable attribute Layers & models have three weight attributes: weights is the list of all weights variables of the layer. trainable_weights is the list of those that are meant to be updated (via gradient descent) to minimize the loss during training. non_trainable_weights is the list of those that aren't meant to be trained

What is the typical transfer learning workflow?

Instantiate a base model and load pre-trained weights into it

Typically they are updated by the model during the forward pass. The typical transfer-learning workflow The typical transfer-learning workflow This leads us to how a typical transfer learning workflow can be implemented in Keras: Instantiate a base model and load pre-trained weights into it. Freeze all layers in the base model by setting trainable = False. Create a new model on top of the output of one (or several) layers from the base model. Train your new model on your new dataset. Note that an alternative, more lightweight workflow could also be: Instantiate a base model and load pre-trained weights into it. Run your new dataset through it and record the output of one (or several) layers from the base model

What is the feature extraction?

every layer has an input and output attribute

This is called feature extraction. Use that output as input data for a new, smaller model. A key advantage of that second workflow is that you only run the base model once one your data, rather than once per epoch of training

What is the problem with Transfer learning with a Sequential model?

taking features learned on one problem, and leveraging them on a new, similar problem

So it's a lot faster & cheaper. An issue with that second workflow, though, is that it doesn't allow you to dynamically modify the input data of your new model during training, which is required when doing data augmentation, for instance. Transfer learning with a Sequential model Transfer learning with a Sequential model First, let's say that you have a Sequential model, and you want to freeze all layers except the last one

What is the most common blueprint for a Sequential model?

a Sequential model to stack a pre-trained model and some freshly initialized

In this case, you would simply iterate over model.layers and set layer.trainable = False on each layer, except the last one. Another common blueprint is to use a Sequential model to stack a pre-trained model and some freshly initialized classification layers. Thanks