Exam AWS Certified Machine Learning - Specialty topic 1 question 63 discussion

Ans: A XGBoost multi class classification. https://medium.com/@gabrielziegler3/multiclass-multilabel-classification-with-xgboost-66195e4d9f2d CNN is used for image classificaiton problems

Answer is A. This a classification problem thus XGBoost and the fact that there are six categories SOFTMAX is the right activation function

Deep convolutional neural networks (CNNs) are primarily used for image processing tasks. Given that the dataset provided is structured/tabular in nature (with features like dimensions, weight, and price) and does not mention image data, a CNN is not the most appropriate choice.

A. YES - perfect fit, multi:softmax the highest probability class is assigned B. NO - CNN is for imaging C. NO - regression forest is for continuuous variables, we can discrete classification D. NO - it is classification, not forecasting

Option A XGBoost multi class classification

A is the answer.

The XGBoost algorithm is a popular and effective technique for multi-class classification. The objective parameter can be set to multi:softmax, which uses a softmax objective function for multi-class classification. This will train the model to predict the probability of each product belonging to each category, and the most probable category will be chosen as the final prediction. A deep convolutional neural network (CNN) (B) is a powerful technique commonly used for image recognition tasks. However, it is less appropriate for tabular data like the dataset provided.

A, CNN is used for image classification. It would be suitable if we were classifying products using pictures of them.

https://xgboost.readthedocs.io/en/stable/parameter.html

B - CNN is used for dataset that have "local intermediate features" ex) images, or textCNN, etc C - We need classfication model, not regression model D - RNN is used for dataset that have sequential features A is correct

A is the best option here. Only 1200 items and 6 classes are not enough data to involve a deep neural architecture for classification.

Ans- A … For multiclassification - multi: SoftMax

That is a classification problem so A is the answer

Easy one. A is correct

A is correct

Definitely A.

100% is A; the the others are clearly wrong Convolutional Neural Network (ConvNet or CNN) is a special type of Neural Network used effectively for image recognition and classification Recurrent neural networks (RNN) are a class of neural networks that is powerful for modeling sequence data such as time series or natural language