uMAP code error in example exercise

[Dyn-Walker]

I copied and pasted the code for the examples so as to follow along, confirm the code worked and to use it as a base for my own data. The Penguin example works fine, however, when I move on to the "digits" example, I get the following errors. As I am a newbie to Python, I don't known how to go about resolving the error. Furthermore, as it is the code for the sample exercise, it may require correcting and/or a note in Frequently Asked Questions.

Thank you.

Regards,

Rick

[jlmelville]

I think you are referring to the docs at https://umap-learn.readthedocs.io/en/latest/basic_usage.html#digits-data.

This bit:

    UMAP(a=None, angular_rp_forest=False, b=None,
         force_approximation_algorithm=False, init='spectral', learning_rate=1.0,
         local_connectivity=1.0, low_memory=False, metric='euclidean',
         metric_kwds=None, min_dist=0.1, n_components=2, n_epochs=None,
         n_neighbors=15, negative_sample_rate=5, output_metric='euclidean',
         output_metric_kwds=None, random_state=42, repulsion_strength=1.0,
         set_op_mix_ratio=1.0, spread=1.0, target_metric='categorical',
         target_metric_kwds=None, target_n_neighbors=-1, target_weight=0.5,
         transform_queue_size=4.0, transform_seed=42, unique=False, verbose=False)

is supposed to represent the output you would get from the previous box's input, specifically the output from running reducer.fit(digits.data). Unfortunately, the current formatting of code doesn't make the difference between output and what the user is supposed to enter very clear. Also, the output has changed a little bit with more recent versions of UMAP. As I was the last person to touch this document, I will see what I can do to improve this. Thanks for pointing this out.

For the rest of that document, all but one of the green boxes indicate code you are intended to enter directly and should not give errors. The one exception is the box that consists entirely of (1797, 2). That is unnecessary and pointless to type in, but it won't break anything.

[Dyn-Walker]

James, I feel like an idiot! Thank you very much for clearing that up for me. Your explanation makes total sense. Regards, Rick Walker

…

On Tue, Dec 20, 2022 at 12:14 AM James Melville ***@***.***> wrote: I think you are referring to the docs at https://umap-learn.readthedocs.io/en/latest/basic_usage.html#digits-data. This bit: UMAP(a=None, angular_rp_forest=False, b=None, force_approximation_algorithm=False, init='spectral', learning_rate=1.0, local_connectivity=1.0, low_memory=False, metric='euclidean', metric_kwds=None, min_dist=0.1, n_components=2, n_epochs=None, n_neighbors=15, negative_sample_rate=5, output_metric='euclidean', output_metric_kwds=None, random_state=42, repulsion_strength=1.0, set_op_mix_ratio=1.0, spread=1.0, target_metric='categorical', target_metric_kwds=None, target_n_neighbors=-1, target_weight=0.5, transform_queue_size=4.0, transform_seed=42, unique=False, verbose=False) is supposed to represent the output you would get from the previous box's input, specifically the output from running reducer.fit(digits.data). Unfortunately, the current formatting of code doesn't make the difference between output and what the user is supposed to enter very clear. Also, the output has changed a little bit with more recent versions of UMAP. As I was the last person to touch this document, I will see what I can do to improve this. Thanks for pointing this out. For the rest of that document, all but one of the green boxes indicate code you are intended to enter directly and should not give errors. The one exception is the box that consists entirely of (1797, 2). That is unnecessary and pointless to type in, but it won't break anything. — Reply to this email directly, view it on GitHub <#943 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A4JFDXDKVAJI65OJLQAV7BTWOFMGHANCNFSM6AAAAAAS4INVUA> . You are receiving this because you authored the thread.Message ID: ***@***.***>