Spring's Inversion of Control (IoC) Container is like a factory that creates and manages objects (called beans) in a Spring application. Instead of creating objects manually in your code, you let the IoC Container do it for you. This allows your code to be more flexible and easier to manage.
Approach:
Configuration: You define what objects (beans) you need and how they should interact with each other. This is done through configuration files (XML or Java-based annotations).
Dependency Injection: The IoC Container automatically injects the required dependencies into the beans, meaning it provides the necessary objects to other objects. You don't need to manually create or connect them.
Lifecycle Management: The container also manages the lifecycle of these beans, such as their creation, initialization, and destruction.
Example:
Imagine you're building a car, and the car needs an engine. Instead of you manually assembling the engine and connecting it to the car, you just tell the IoC Container, "Hey, I need a car, and it needs an engine." The IoC Container then finds an engine, puts it in the car, and hands you the fully assembled car.
Benefits:
Decoupling: Your code is less tightly connected, making it easier to change or replace parts.
Easier Testing: Since dependencies are injected, it's easier to test components in isolation.
Flexibility: You can easily switch out components without changing much code.
Conclusion:
Spring's IoC Container simplifies object management in your application by taking care of object creation, dependency injection, and lifecycle management, allowing you to focus on writing business logic rather than managing dependencies.
1.(True/False) Recurrent Neural Networks are a class of neural networks that allow previous outputs to be used as inputs while having hidden states.
ANSWER= (A) True
2. (True/False) Recurrent Neural Networks are well suited in applications in which the context is important and needs to be incorporated in the prediction.
ANSWER= (A) True
3. These are the two main outputs of a recurrent neural network:
ANSWER= (A) Prediction and state
Practice: LSTM and GRU
1.(True/False) The main motivation behind LSTM is to make it easier to keep information from distant past in current memory without reinforcement.
ANSWER= (A) True
2. RNNs are augmented with the following Gate Units:
ANSWER= (B) Input gate, forget gate, output gate
3. Select the correct assertion regarding the gate units of RNNs:
ANSWER= (D) A and B
Practice: Regularization
1. Which regularization technique can shrink the coefficients of the less important features to zero?
ANSWER= (D) L1
2. (True/False) Batch Normalization tackles the internal covariate shift issue by always normalizing the input signals, thus accelerating the training of deep neural nets and increasing the generalization power of the networks.
ANSWER= (A) True
3. Regularization is used to mitigate which issue in model training?
ANSWER= (C) Overfitting
Week 6 Final Quiz
1. (True/False) RNN models are mostly used in the fields of natural language processing and speech recognition.
ANSWER= (A) True
2. (True/False) GRUs and LSTM are a way to deal with the vanishing gradient problem encountered by RNNs.
ANSWER= (A) True
3. (True/False) GRUs will generally perform about as well as LSTMs with shorter training time, especially for smaller datasets.
ANSWER= (A) True
4. (True/False) The main idea of Seq2Seq models is to improve accuracy by keeping necessary information in the hidden state from one sequence to the next.
ANSWER= (A) True
5. (True/False) The main parts of a Seq2Seq model are: an encoder, a hidden state, a sequence state, and a decoder.
ANSWER= (B) False
6. Select the correct option, in the context of Seq2Seq models:
ANSWER= (C) The Greedy Search algorithm selects one best candidate as an input sequence for each time step while the Beam Search produces multiple different hypothesis based on conditional probability.
7. Which is the gating mechanism for RNNs that include a reset gate and an update gate?
ANSWER= (A) GRUs
8.LSTM models are among the most common Deep Learning models used in forecasting. These are other common uses of LSTM models, except:
1.The main idea of transfer learning of a neural network is:
ANSWER= (A) To keep the early layers of a pre-trained network and re-train the later layers for a specific application.
2. In the context of transfer learning, which is a guiding principle of fine tuning?
ANSWER= (B) Using data that is similar to the pre-trained network
3. In the context of transfer learning, what do we call the process in which you only train the last or a few layers instead of all layers of a neural network?
1.This concept came as a solution to CNNs in which each layer is turned into branches of convolutions:
ANSWER= (A) Inception
2. Which CNN Architecture is considered the flash point for modern Deep Learning?
ANSWER= (A) AlexNet
3. Which CNN Architecture can be described as a "simplified, deeper LeNet" in which the more layers, the better?
ANSWER= (C) VGG
4.Which CNN Architecture is the precursor of using convolutions to obtain better features and was first used to solve the MNIST data set?
ANSWER= (E) LeNet
5. The motivation behind this CNN Architecture was to solve the inability of deep neural networks to fit or overfit the training data better when adding layers.
ANSWER= (E) ResNet
6. This CNN Architecture keeps passing both the initial unchanged information and the transformed information to the next layer.
ANSWER= (E) ResNet
7. Which activation function was notably used in AlexNet and contributed to its success?
ANSWER= (A) ReLU (Rectified Linear Unit)
Practice: Regularization
1. Which regularization technique can shrink the coefficients of the less important features to zero?
ANSWER= (D) L1
2. (True/False) Batch Normalization tackles the internal covariate shift issue by always normalizing the input signals, thus accelerating the training of deep neural nets and increasing the generalization power of the networks.
ANSWER= (A) True
3. Regularization is used to mitigate which issue in model training?
ANSWER= (C) Overfitting
Week 5 Final Quiz
1. (True/False) In Keras, the Dropout layer has an argument called rate, which is a probability that represents how often we want to invoke the layer in the training.
ANSWER= (B) False
2. What is a benefit of applying transfer learning to neural networks?
ANSWER= (B) Save early layers for generalization before re-training later layers for specific applications.
3. By setting ` layer.trainable = False` for certain layers in a neural network, we____
ANSWER= (D) freeze the layers such that their weights don’t update during training.
4.Which option correctly orders the steps of implementing transfer learning?
1. Freeze the early layers of the pre-trained model.
2. Improve the model by fine-tuning.
3. Train the model with a new output layer in place.
4. Select a pre-trained model as the base of our training.
ANSWER= (B) 4, 1, 3, 2
5. Given a 100x100 pixels RGB image, there are _____ features.
ANSWER= (C) 30000
6. Before a CNN is ready for classifying images, what layer must we add as the last?
ANSWER= (A) Dense layer with the number of units corresponding to the number of classes
7. In a CNN, the depth of a layer corresponds to the number of:
