diff --git a/01-introduction.html b/01-introduction.html index 0cacfe5..a9bee1c 100644 --- a/01-introduction.html +++ b/01-introduction.html @@ -571,7 +571,7 @@
Two of the most commonly used libraries for image representation and manipulation are NumPy and Pillow (PIL). Additionally, when working with @@ -672,7 +672,7 @@
ChatGPT
Normalizing the RGB values to be between 0 and 1 is a common @@ -890,7 +890,7 @@
A. Training Set
ChatGPT
Data is typically split into the training, validation, and test data diff --git a/03-build-cnn.html b/03-build-cnn.html index 7a5fcc8..0b10e80 100644 --- a/03-build-cnn.html +++ b/03-build-cnn.html @@ -491,7 +491,7 @@
A flatten layer function is typically used to transform the two-dimensional arrays (matrices) generated by the convolutional and @@ -892,7 +892,7 @@
The Model.summary() output has three columns:
ChatGPT
Learning rate is a hyperparameter that determines @@ -583,7 +583,7 @@
ChatGPT
The choice of batch size can have various implications, and there are @@ -822,7 +822,7 @@
Underfitting occurs when the model is too simple or lacks the capacity to capture the underlying patterns and relationships present in @@ -980,7 +980,7 @@
ChatGPT
Regularization methods introduce constraints or penalties to the diff --git a/05-evaluate-predict-cnn.html b/05-evaluate-predict-cnn.html index 2a03550..8a02983 100644 --- a/05-evaluate-predict-cnn.html +++ b/05-evaluate-predict-cnn.html @@ -423,7 +423,7 @@
It depends! Recall in an Episode 02 Introduction to Image Data Callout we talked about the different @@ -480,7 +480,7 @@
A1. The confusion matrix illustrates that the predictions are not bad but can be improved.
@@ -729,7 +729,7 @@A1. Varying the dropout rate
The code below instantiates and trains a model with varying dropout @@ -863,7 +863,7 @@
The code below serves as a practical means for exploring activation performance on an image dataset.
@@ -1128,7 +1128,7 @@This is an open question.
Regarding the model architecture:
diff --git a/aio.html b/aio.html index c389e21..db4a020 100644 --- a/aio.html +++ b/aio.html @@ -621,7 +621,7 @@Two of the most commonly used libraries for image representation and manipulation are NumPy and Pillow (PIL). Additionally, when working with @@ -1342,7 +1342,7 @@
ChatGPT
Normalizing the RGB values to be between 0 and 1 is a common @@ -1601,7 +1601,7 @@
A. Training Set
ChatGPT
Data is typically split into the training, validation, and test data @@ -2008,7 +2008,7 @@