Amulya Baniya

Issues with 2 principal components

1

 How do I fix this?

Chapter 14, MNIST test set plot

1

**Page 482** The code given is not working as expected. I think it should be `mnist_test_dataset.data`. ```python fig = plt.figure(figsize=(12, 4)) for i in range(12): ax = fig.add_subplot(2, 6, i+1)...

$$ \frac{\partial a_1^{(out)}}{\partial z_1^{(out)}} = \frac{\partial}{\partial z_1^{out}}\cdot \frac{1}{1+e^{z_1^{(out)}}} = ... = \left(\frac{1}{1+e^{z_1^{(out)}}}\right) \cdot \left(1-\frac{1}{1+e^{z_1^{(out)}}}\right)$$ In this equation, the exponents should have negative sign

We were only using Petal Length and Petal width but in the plot of decision trees you have used `feature_names = ["Sepal Length", "Sepal width", "Petal Length", "Petal width"]`. Shouldn't...

BaselineShift fails on 2d array with `dtype=int`

9

BaselineShift class fails on an `int` dtype 2d array. The issue lies on the implementation [here](https://github.com/paucablop/chemotools/blob/067ccd580eb8545b808a88d7ab42ed9615681ea1/chemotools/augmentation/baseline_shift.py#L103). The issue is this: ```python x = np.array([[1,2,3]]) # dtype int new_x = np.add(x,...