save_fig is not defined

[ageron]

save_fig is not defined

[UlrikeBoehm]

I used
plt.savefig("attribute_histogram_plots")
instead of
save_fig("attribute_histogram_plots")
then it works!

[ageron]

Hi @loewenm,

Thanks for your kind words, I'm glad you are enjoying my book. :)

The save_fig() function is defined in the first cell of each notebook: just run that cell and the error should go away. You would typically get the NameError: name 'save_fig' is not defined error if you restart Jupyter (or just the Python kernel) and then go back to the cell you were at without running the previous cells again. When you want to continue running a notebook after an interruption, you need to run the previous cells again, because the Python kernel has lost all the context.

I hope this helps. Can you please point me to the discussion threads about this, so I can help them out? Thanks!

Cheers,
Aurélien

[loewenm]

Hi Aurelien, Thank you for the above. I'm still having this issue though... I fully realize that I need to restart the kernel and re-run the initial portion of the section (i.e. all the library imports, etc.) to ensure that everything is running smoothly as I progress through a chapter... However, I get this error in every single chapter. Every one! I am using Spyder instead of Jupyter as I like the interface a little more and the Variable Explorer window is really helpful. If you would indulge me for a minute and wouldn't mind running the following code in Spyder please? import numpy as np import matplotlib.pyplot as plt from sklearn import datasets iris = datasets.load_iris() list(iris.keys()) print(iris.DESCR) X = iris["data"][:, 3:] # petal width y = (iris["target"] == 2).astype(np.int) # 1 if Iris-Virginica, else 0 from sklearn.linear_model import LogisticRegression log_reg = LogisticRegression(random_state=42) log_reg.fit(X, y) # Creates a series that has 1000 datapoints, which sigmoid scale from 0 to 3 X_new = np.linspace(0, 3, 1000).reshape(-1, 1) # Uses log_reg model to predict odds the series is Virginica or not... y_proba = log_reg.predict_proba(X_new) plt.plot(X_new, y_proba[:, 1], "g-", linewidth=2, label="Iris-Virginica") plt.plot(X_new, y_proba[:, 0], "b--", linewidth=2, label="Not Iris-Virginica") # Prettier chart decision_boundary = X_new[y_proba[:, 1] >= 0.5][0] plt.figure(figsize=(8, 3)) plt.plot(X[y==0], y[y==0], "bs") plt.plot(X[y==1], y[y==1], "g^") plt.plot([decision_boundary, decision_boundary], [-1, 2], "k:", linewidth=2) plt.plot(X_new, y_proba[:, 1], "g-", linewidth=2, label="Iris-Virginica") plt.plot(X_new, y_proba[:, 0], "b--", linewidth=2, label="Not Iris-Virginica") plt.text(decision_boundary+0.02, 0.15, "Decision boundary", fontsize=14, color="k", ha="center") plt.arrow(decision_boundary, 0.08, -0.3, 0, head_width=0.05, head_length=0.1, fc='b', ec='b') plt.arrow(decision_boundary, 0.92, 0.3, 0, head_width=0.05, head_length=0.1, fc='g', ec='g') plt.xlabel("Petal width (cm)", fontsize=14) plt.ylabel("Probability", fontsize=14) plt.legend(loc="center left", fontsize=14) plt.axis([0, 3, -0.02, 1.02]) save_fig("logistic_regression_plot") plt.show() Same error as before: Traceback (most recent call last): File "<ipython-input-8-9190cda20295>", line 16, in <module> save_fig("logistic_regression_plot") NameError: name 'save_fig' is not defined Kind regards, Michael

…

On Thu, Jul 6, 2017 at 12:55 PM, Aurélien Geron ***@***.***> wrote: Hi @loewenm <https://github.com/loewenm>, Thanks for your kind words, I'm glad you are enjoying my book. :) The save_fig() function is defined in the first cell of each notebook: just run that cell and the error should go away. You would typically get the NameError: name 'save_fig' is not defined error if you restart Jupyter (or just the Python kernel) and then go back to the cell you were at without running the previous cells again. When you want to continue running a notebook after an interruption, you need to run the previous cells again, because the Python kernel has lost all the context. I hope this helps. Can you please point me to the discussion threads about this, so I can help them out? Thanks! Cheers, Aurélien — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#50 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ASl5NoptZrjsgjhxYlVE-i4msCcdVGE_ks5sLRGZgaJpZM4OOhod> .

-- Signed, Michael Loewen [email protected]

[ageron]

Hi @loewenm ,
Sorry for the delay, I missed your message as the issue was closed. Did you manage to get the code to run in Spyder? The piece of code you asked me to run does not include the definition of the save_fig() function, so it's normal to get an error saying that the function is undefined. You need to make sure you have the following function definition at the beginning of your code:

# Imports
import os
import matplotlib.pyplot as plt

# Where to save the figures
PROJECT_ROOT_DIR = "."
CHAPTER_ID = "fundamentals"

def save_fig(fig_id, tight_layout=True):
    path = os.path.join(PROJECT_ROOT_DIR, "images", CHAPTER_ID, fig_id + ".png")
    print("Saving figure", fig_id)
    if tight_layout:
        plt.tight_layout()
    plt.savefig(path, format='png', dpi=300)

[ageron]

I'm glad this helped! Regarding QML, it's definitely fascinating, but I'm afraid I really don't know much about this topic (yet?). I read an introductory book about Quantum Physics about 10 years ago, and it blew my mind. I was so motivated to learn that I bought an actual university textbook, but I was lost after reading only about 10 pages. I should try again, perhaps it would be easier now that I've practiced a lot of Linear Algebra.
I'd love to read your thoughts about the topic: what do you think will be the main applications? Is it mostly about faster training time? Exploring huge hyperparameter spaces? Training on quantum data?

[olusegun-omisakin]

Thanks, UlrikeBoehm. It does!

[tiagobrc]

Thanks! save_fig my ...

[Neeraj02a]

I used
plt.savefig("attribute_histogram_plots")
instead of
save_fig("attribute_histogram_plots")
then it works!

That was really very help full, it's working rather then what's described above

[ageron]

Hi @Neeraj02a ,

Thanks for your feedback. My save_fig() function is just a thin wrapper for Matplotlib's plt.savefig() function. It just handles the file path (so that each image is saved into the appropriate folder, for example in images/fundamentals if you're running the chapter 1 notebook), and by default it also calls plt.tight_layout() to save some space. Lastly, it ensures the output format is PNG and the resolution is 300 dpi. This is the format expected by O'Reilly: they need print-quality images.

If you get an error when calling save_fig(), it might be because you're missing some of the imports? I added the ones you need in the code example above.

Hope this helps.

[mkayanda]

Thank you @ageron for such a well explained resolution. I had the same issue but coming here gave me the best insights.