daviddoji/ISLP_labs
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

standard_lasso

Browse Source 
Jonathan corrected a spurious use of standard_lasso
This commit is contained in:
Trevor Hastie
2024-01-25 17:53:08 -08:00
parent a55798e4d3
commit e5bbb1a5bc
2 changed files with 527 additions and 176744 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
Ch10-deeplearning-lab.Rmd
View File

@@ -1,15 +1,3 @@
---
jupyter:
jupytext:
cell_metadata_filter: -all
formats: Rmd,ipynb
main_language: python
text_representation:
extension: .Rmd
format_name: rmarkdown
format_version: '1.2'
jupytext_version: 1.14.7
---
# Chapter 10 # Chapter 10
@@ -236,7 +224,7 @@ $\lambda$ with an all-zero solution. This value equals the largest absolute inn
X_s = scaler.fit_transform(X_train) X_s = scaler.fit_transform(X_train)
n = X_s.shape[0] n = X_s.shape[0]
lam_max = np.fabs(X_s.T.dot(Y_train - Y_train.mean())).max() / n lam_max = np.fabs(X_s.T.dot(Y_train - Y_train.mean())).max() / n
param_grid = {'alpha': np.exp(np.linspace(0, np.log(0.01), 100)) param_grid = {'lasso__alpha': np.exp(np.linspace(0, np.log(0.01), 100))
* lam_max} * lam_max}
``` ```
Note that we had to transform the data first, since the scale of the variables impacts the choice of $\lambda$. Note that we had to transform the data first, since the scale of the variables impacts the choice of $\lambda$.
@@ -246,7 +234,7 @@ We now perform cross-validation using this sequence of $\lambda$ values.
cv = KFold(10, cv = KFold(10,
shuffle=True, shuffle=True,
random_state=1) random_state=1)
grid = GridSearchCV(lasso, grid = GridSearchCV(standard_lasso,
param_grid, param_grid,
cv=cv, cv=cv,
scoring='neg_mean_absolute_error') scoring='neg_mean_absolute_error')
@@ -868,7 +856,7 @@ for idx, (X_ ,Y_) in enumerate(cifar_dm.train_dataloader()):
Before we start, we look at some of the training images; similar code produced Before we start, we look at some of the training images; similar code produced
Figure 10.5 on page 447. The example below also illustrates Figure 10.5 on page 406. The example below also illustrates
that `TensorDataset` objects can be indexed with integers --- we are choosing that `TensorDataset` objects can be indexed with integers --- we are choosing
random images from the training data by indexing `cifar_train`. In order to display correctly, random images from the training data by indexing `cifar_train`. In order to display correctly,
we must reorder the dimensions by a call to `np.transpose()`. we must reorder the dimensions by a call to `np.transpose()`.
@@ -1388,7 +1376,7 @@ We’ll now make a plot to compare our neural network results with the
lasso. lasso.
```{python} ```{python}
# %%capture %%capture
fig, axes = subplots(1, 2, figsize=(16, 8), sharey=True) fig, axes = subplots(1, 2, figsize=(16, 8), sharey=True)
for ((X_, Y_), for ((X_, Y_),
data_, data_,

177109
Ch10-deeplearning-lab.ipynb
View File

File diff suppressed because one or more lines are too long