24 add example gallery to sphinx docs #27
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
bjdebus
reviewed
Oct 24, 2025
| # Testing data | ||
| n_tst = 10000 | ||
| x_tst = rng.random(n=n_tst) # random numbers in [0,1]^d | ||
| y_tst = func(x_tst) + np.random.normal(0, noise_std, size = (n_tst,1)) |
There was a problem hiding this comment.
Why are we adding noise to the test data? I think we should only add noise to the training data, not the testing data.
bjdebus
reviewed
Oct 24, 2025
|
|
||
| # Training data | ||
| np.random.seed(42) | ||
| n_trn = 70 |
There was a problem hiding this comment.
It would be helpful to explain the choice of 70 training points. As BCS is a sparse regression approach, it is designed for the situation where we do not have enough data to fit all basis terms without using regularization.
bjdebus
reviewed
Oct 24, 2025
| print("Number of Basis Terms:", len(pce_surr.pcrv.mindices[0])) | ||
|
|
||
| # (1.5) Set training data | ||
| pce_surr.set_training_data(x_trn, y_trn[:,0]) |
There was a problem hiding this comment.
This data is still defined on [0,1]^d rather than [-1,1]^d. I thought you had reworked this to match the domain of the basis terms. Do I have that wrong?
bjdebus
reviewed
Oct 24, 2025
| np.random.seed(42) | ||
| n_trn = 70 | ||
| x_trn = rng.random(n=n_trn) # random numbers in [0,1]^d | ||
| y_trn = func(x_trn) + np.random.normal(0, noise_std, size = (n_trn,1)) |
bjdebus
reviewed
Oct 24, 2025
| y_trn_approx = pce_surr.evaluate(x_trn) | ||
| y_tst_approx = pce_surr.evaluate(x_tst) | ||
|
|
||
| # Evaluate goodness of fit with RMSE |
There was a problem hiding this comment.
I would move the computation of the RMSE errors below the parity plots. The parity plots already indicate that there is a lot of error in the testing data predictions, and the RMSE then confirms that.
|
Hi Bert,
I should have clarified that the example in that branch is the non-updated one. The branch with the updated example is not yet on GitHub (it’s for the issue regarding integrating the eta optimization function); I actually was hoping we could briefly look over the changes in the example notebook again today.
Best,
Emilie
From: Bert Debusschere ***@***.***>
Date: Friday, October 24, 2025 at 2:27 PM
To: sandialabs/pytuq ***@***.***>
Cc: Baillo, Emilie Grace Eduria ***@***.***>, Author ***@***.***>
Subject: [EXTERNAL] Re: [sandialabs/pytuq] 24 add example gallery to sphinx docs (PR #27)
@bjdebus commented on this pull request.
________________________________
In docs/auto_examples/ex_genz_bcs.py<#27 (comment)>:
+# After importing GenzOscillatory from ``pytuq.func.genz``, we generate the Genz function below, along with training data and testing data with output noise.
+# This data and the corresponding Genz function will be used to create the same PC surrogate fitted in both examples:
+# (1) The first surrogate will be fitted using BCS with a given eta, and (2) the second surrogate will be fitted using BCS with the most optimal eta.
+
+# Use Genz Oscillatory function in multiple dimensions for the true model
+func_dim = 4
+func_weights = [1.0/(i+1)**2 for i in range(func_dim)]
+func = GenzOscillatory(shift=0.25, weights=func_weights)
+noise_std = 0.1
+rng = qmc.LatinHypercube(d=func_dim, seed=rng_seed)
+
+# Training data
+np.random.seed(42)
+n_trn = 70
+x_trn = rng.random(n=n_trn) # random numbers in [0,1]^d
+y_trn = func(x_trn) + np.random.normal(0, noise_std, size = (n_trn,1))
Does func return a 2-D array?
—
Reply to this email directly, view it on GitHub<#27 (review)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BOTIAVYAGUCPTGJ3HHBMKVD3ZKKTJAVCNFSM6AAAAACKEFVOKWVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZTGNZYHA4TGOBVGI>.
You are receiving this because you authored the thread.
|
bjdebus
approved these changes
Oct 24, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Added sphinx-gallery to the docs