Tutorial Notebook Results¶
Executed outputs from the tutorial notebooks in notebooks/tutorials/. These pages are generated from notebook outputs, including text tables and plots.
Environment¶
- Generated: stable
- Git commit:
stable - Python:
3.12.1 - Package version:
0.2.14 - Matplotlib backend:
Agg
Summary¶
Classical vs Quantum Classifiers¶
Notebook: notebooks/tutorials/01-classical-vs-quantum-classifier.ipynb
Result block 1:
VQC train accuracy: 0.84
VQC test accuracy: 0.92
Result block 2:
Quantum kernel train accuracy: 0.8666666666666667
Quantum kernel test accuracy: 0.92
Result block 3:
Classical SVM train accuracy: 0.96
Classical SVM test accuracy: 1.0
Result block 4:
{'VQC': {'train_accuracy': 0.84, 'test_accuracy': 0.92},
'Quantum kernel': {'train_accuracy': 0.8666666666666667,
'test_accuracy': 0.92},
'Classical SVM': {'train_accuracy': 0.96, 'test_accuracy': 1.0}}
Classical vs quantum regression¶
Notebook: notebooks/tutorials/02-classical-vs-quantum-regressor.ipynb
Result block 1:
[{'model': 'VQR',
'train_mse': 0.29700758438371283,
'test_mse': 0.17865945115851434,
'train_mae': 0.23254695445727222,
'test_mae': 0.20425433701803775},
{'model': 'Ridge',
'train_mse': 4.490711703649848e-05,
'test_mse': 4.315219471074882e-05,
'train_mae': 0.00516673344748882,
'test_mae': 0.005244853133929798},
{'model': 'MLP',
'train_mse': 0.0057038784522001765,
'test_mse': 0.004103491527765519,
'train_mae': 0.05753716767756181,
'test_mae': 0.05229663963250704}]
Result block 2:
VQR | train MSE = 0.297008 | test MSE = 0.178659 | train MAE = 0.232547 | test MAE = 0.204254
Ridge | train MSE = 0.000045 | test MSE = 0.000043 | train MAE = 0.005167 | test MAE = 0.005245
MLP | train MSE = 0.005704 | test MSE = 0.004103 | train MAE = 0.057537 | test MAE = 0.052297
Result block 3:
Lowest test MSE: Ridge (0.000043)
Lowest test MAE: Ridge (0.005245)
Quantum Variational Classifier¶
Notebook: notebooks/tutorials/03-variational-quantum-classifier.ipynb
Result block 1:
Train accuracy: 0.8533333333333334
Test accuracy: 0.84
Final loss: 0.3296512422612351
Result block 2:
[1.3637149385725453,
1.3106343893072079,
1.150096754736344,
0.9990183939256337,
0.8762037652494321,
0.775059270097408,
0.6885867746531164,
0.613274820659994,
0.5479665993633804,
0.4922192570203015]
Result block 3:
['ansatz_params',
'dataset',
'early_stopping_min_delta',
'early_stopping_patience',
'embedding',
'embedding_layers',
'embedding_params',
'final_loss',
'loss_history',
'model',
'n_layers',
'n_qubits',
'n_samples',
'noise',
'optimizer',
'optimizer_kwargs',
'params',
'seed',
'shots',
'step_size',
'steps',
'test_accuracy',
'test_probabilities',
'test_size',
'train_accuracy',
'train_probabilities',
'y_test',
'y_test_pred',
'y_train',
'y_train_pred']
Variational Quantum Regressor¶
Notebook: notebooks/tutorials/04-variational-quantum-regressor.ipynb
Result block 1:
Train MSE: 0.5133790198164415
Test MSE: 0.5351584510286133
Train MAE: 0.5291734269340835
Test MAE: 0.49500624808166144
Final loss: 0.5160344040102691
Result block 2:
[1.53959574370404,
1.480014268088658,
1.4249729606232324,
1.3747701545384936,
1.3292252417575974,
1.287867793360639,
1.2495917135548547,
1.2127663351701832,
1.1754339606139674,
1.1359078512368066]
Result block 3:
['dataset',
'early_stopping_min_delta',
'early_stopping_patience',
'final_loss',
'loss_history',
'model',
'n_layers',
'n_qubits',
'n_samples',
'noise',
'optimizer',
'optimizer_kwargs',
'params',
'seed',
'shots',
'step_size',
'steps',
'test_mae',
'test_mse',
'test_size',
'train_mae',
'train_mse',
'x_test',
'x_train',
'y_test',
'y_test_pred',
'y_train',
'y_train_pred']
Quantum Kernel Classifier¶
Notebook: notebooks/tutorials/05-quantum-kernel-classifier.ipynb
Result block 1:
Train accuracy: 0.8666666666666667
Test accuracy: 0.92
Result block 2:
(75, 75)
Result block 3:
array([1, 1, 0, 0, 0, 1, 0, 1, 0, 1])
Result block 4:
['dataset',
'embedding',
'kernel_matrix_test',
'kernel_matrix_train',
'model',
'n_qubits',
'n_samples',
'noise',
'seed',
'shots',
'test_accuracy',
'test_size',
'train_accuracy',
'x_test',
'x_train',
'y_test',
'y_test_pred',
'y_train',
'y_train_pred']
Reusable Quantum Kernel Estimators¶
Notebook: notebooks/tutorials/06-quantum-kernel-estimators.ipynb
Result block 1:
Dataset
+------------------------------+---------------------+
| Metric | Value |
+------------------------------+---------------------+
| Classification samples | 20 |
| Classification feature_shape | [20, 2] |
| Regression samples | 20 |
| Regression feature_shape | [20, 2] |
| Regression target range | [-0.89437, 0.46061] |
+------------------------------+---------------------+
Result block 2:
Results
+-----------------------------+-----------+
| Metric | Value |
+-----------------------------+-----------+
| Kernel-target alignment | 0.634204 |
| Quantum classifier accuracy | 1 |
| SVC classifier accuracy | 1 |
| Quantum regression MAE | 0.0861475 |
| Quantum regression MSE | 0.0186331 |
| Kernel ridge regression MAE | 0.0208563 |
+-----------------------------+-----------+
Result block 3:
Validation
Dataset
+------------------------------+--------------------------------------+
| Metric | Value |
+------------------------------+--------------------------------------+
| Problem | reusable_quantum_kernel_estimators |
| Classification features | [sensor_feature_1, sensor_feature_2] |
| Classification target | operating_regime |
| Regression features | [calibration_axis, cosine_feature] |
| Regression target | calibration_value |
| Classification train samples | 14 |
| Classification test samples | 6 |
| Regression train samples | 14 |
| Regression test samples | 6 |
+------------------------------+--------------------------------------+
Results
+-----------------------------+-----------+
| Metric | Value |
+-----------------------------+-----------+
| Kernel-target alignment | 0.634204 |
| Quantum classifier accuracy | 1 |
| SVC classifier accuracy | 1 |
| Quantum regression MAE | 0.0861475 |
| Quantum regression MSE | 0.0186331 |
| Kernel ridge regression MAE | 0.0208563 |
+-----------------------------+-----------+
Sample classification predictions
1. Actual: 0, Quantum: 0, SVC: 0
2. Actual: 0, Quantum: 0, SVC: 0
3. Actual: 0, Quantum: 0, SVC: 0
4. Actual: 1, Quantum: 1, SVC: 1
5. Actual: 1, Quantum: 1, SVC: 1
6. Actual: 1, Quantum: 1, SVC: 1
Sample regression predictions
1. Actual: -0.172629, Quantum: -0.109190, Kernel ridge: -0.155921
2. Actual: 0.410242, Quantum: 0.425287, Kernel ridge: 0.402479
3. Actual: -0.764573, Quantum: -0.791361, Kernel ridge: -0.777201
4. Actual: 0.457576, Quantum: 0.428382, Kernel ridge: 0.464813
5. Actual: 0.460610, Quantum: 0.397573, Kernel ridge: 0.465578
6. Actual: 0.250000, Quantum: -0.069381, Kernel ridge: 0.174167
Interpretation
The reusable quantum kernel estimators produced finite predictions and useful held-out metrics.
Classical baselines are included as sanity checks; no quantum advantage is claimed.
Passed: True
Dataset-Agnostic Variational Quantum Estimators¶
Notebook: notebooks/tutorials/07-variational-quantum-estimators.ipynb
Result block 1:
Dataset
+------------------------------+-----------+
| Metric | Value |
+------------------------------+-----------+
| Classification samples | 15 |
| Classification feature_shape | [15, 2] |
| Classes | [0, 1, 2] |
| Regression samples | 6 |
| Regression feature_shape | [6, 2] |
| Regression target_shape | [6, 2] |
+------------------------------+-----------+
Result block 2:
Results
+------------------------------+-------------+
| Metric | Value |
+------------------------------+-------------+
| Quantum classifier accuracy | 0.666667 |
| SVC classifier accuracy | 1 |
| Quantum regression MAE | 0.393804 |
| Quantum regression MSE | 0.207483 |
| Ridge regression MAE | 0.000269655 |
| Classifier initial loss mean | 0.776098 |
| Classifier final loss mean | 0.392458 |
| Regressor initial loss mean | 0.34266 |
| Regressor final loss mean | 0.211391 |
+------------------------------+-------------+
Result block 3:
Validation
Dataset
+-------------------------+------------------------------------------+
| Metric | Value |
+-------------------------+------------------------------------------+
| Problem | dataset_agnostic_variational_estimators |
| Classification features | [feature_1, feature_2] |
| Classification target | operating_regime |
| Regression features | [input_1, input_2] |
| Regression targets | [response_channel_1, response_channel_2] |
| Classification samples | 15 |
| Regression samples | 6 |
+-------------------------+------------------------------------------+
Results
+------------------------------+-------------+
| Metric | Value |
+------------------------------+-------------+
| Quantum classifier accuracy | 0.666667 |
| SVC classifier accuracy | 1 |
| Quantum regression MAE | 0.393804 |
| Quantum regression MSE | 0.207483 |
| Ridge regression MAE | 0.000269655 |
| Classifier initial loss mean | 0.776098 |
| Classifier final loss mean | 0.392458 |
| Regressor initial loss mean | 0.34266 |
| Regressor final loss mean | 0.211391 |
+------------------------------+-------------+
Sample class predictions
1. Actual: 0, Quantum: 2, SVC: 0, Probabilities: [0.4941, 0.0114, 0.4945]
2. Actual: 0, Quantum: 2, SVC: 0, Probabilities: [0.4738, 0.0519, 0.4743]
3. Actual: 0, Quantum: 2, SVC: 0, Probabilities: [0.4949, 0.0097, 0.4954]
4. Actual: 0, Quantum: 2, SVC: 0, Probabilities: [0.4712, 0.057, 0.4717]
5. Actual: 0, Quantum: 2, SVC: 0, Probabilities: [0.4952, 0.0091, 0.4956]
6. Actual: 1, Quantum: 1, SVC: 1, Probabilities: [0.0797, 0.8408, 0.0795]
Sample regression predictions
1. Actual: [-0.2, -0.1], Quantum: [-0.523667, -0.517271], Ridge: [-0.199475, -0.10035]
2. Actual: [-0.04, -0.04], Quantum: [0.034384, 0.041868], Ridge: [-0.04019, -0.039824]
3. Actual: [0.12, 0.08], Quantum: [0.580979, 0.587058], Ridge: [0.119784, 0.080175]
4. Actual: [0.28, 0.2], Quantum: [0.934008, 0.936657], Ridge: [0.279758, 0.200175]
Interpretation
The sklearn-like quantum estimators fit, expose learned attributes, and return finite predictions.
The validation threshold is intentionally modest because this notebook is an API usage example, not a benchmark claim.
Passed: True
Sequence Windows for Quantum Forecasting¶
Notebook: notebooks/tutorials/08-sequence-window-quantum-forecasting.ipynb
Result block 1:
Dataset
+------------------------+---------+
| Metric | Value |
+------------------------+---------+
| Raw signal steps | 44 |
| Window size | 3 |
| Horizon | 1 |
| Windowed feature_shape | [41, 3] |
| Target shape | [41] |
| Train windows | 30 |
| Test windows | 11 |
+------------------------+---------+
Result block 2:
Results
+----------------------+----------+
| Metric | Value |
+----------------------+----------+
| Quantum forecast MAE | 0.406761 |
| Quantum forecast MSE | 0.207272 |
| Ridge forecast MAE | 0.182912 |
| Initial loss | 0.715744 |
| Final loss | 0.222213 |
| Loss steps | 20 |
+----------------------+----------+
Result block 3:
Validation
Dataset
+------------------+--------------------------------------------------------+
| Metric | Value |
+------------------+--------------------------------------------------------+
| Problem | sequence_window_quantum_forecasting |
| Features | [signal_t_minus_2, signal_t_minus_1, signal_t_minus_0] |
| Target | next_signal_value |
| Raw signal steps | 44 |
| Train windows | 30 |
| Test windows | 11 |
+------------------+--------------------------------------------------------+
Results
+----------------------+----------+
| Metric | Value |
+----------------------+----------+
| Quantum forecast MAE | 0.406761 |
| Quantum forecast MSE | 0.207272 |
| Ridge forecast MAE | 0.182912 |
| Initial loss | 0.715744 |
| Final loss | 0.222213 |
| Loss steps | 20 |
+----------------------+----------+
Sample forecasts
1. Actual: 0.569579, Quantum: 0.107778, Ridge: 0.342875
2. Actual: 0.864542, Quantum: 0.942297, Ridge: 1.049405
3. Actual: 0.850233, Quantum: 0.522847, Ridge: 0.875183
4. Actual: 0.856505, Quantum: 0.144220, Ridge: 0.549189
5. Actual: 0.490632, Quantum: 0.163618, Ridge: 0.522962
6. Actual: -0.309106, Quantum: 0.140184, Ridge: 0.009358
Interpretation
The preprocessing helper created valid supervised windows and the quantum regressor reduced training loss.
The ridge baseline is included as a sanity check; no quantum advantage is claimed.
Passed: True
Quantum Metric Learning¶
Notebook: notebooks/tutorials/09-quantum-metric-learning.ipynb
Result block 1:
[metric_learning] step=0001 loss=0.427937
[metric_learning] step=0010 loss=0.305071
[metric_learning] step=0020 loss=0.298624
[metric_learning] step=0030 loss=0.255549
[metric_learning] step=0040 loss=0.286283
[metric_learning] step=0050 loss=0.260128
[metric_learning] step=0060 loss=0.143402
[metric_learning] step=0070 loss=0.156889
[metric_learning] step=0075 loss=0.098516
Train accuracy: 0.6111111111111112
Test accuracy: 0.6
Final loss: 0.09851639525405846
Result block 2:
Train embedding shape: (90, 2)
Test embedding shape: (30, 2)
Result block 3:
[metric_learning] step=0001 loss=0.087691
[metric_learning] step=0010 loss=0.225662
[metric_learning] step=0020 loss=0.134055
[metric_learning] step=0030 loss=0.130827
[metric_learning] step=0040 loss=0.144779
[metric_learning] step=0050 loss=0.261926
[metric_learning] step=0060 loss=0.167996
[metric_learning] step=0070 loss=0.194980
[metric_learning] step=0075 loss=0.151230
layers=1 test accuracy: 0.7
Result block 4:
[metric_learning] step=0001 loss=0.427937
[metric_learning] step=0010 loss=0.305071
[metric_learning] step=0020 loss=0.298624
[metric_learning] step=0030 loss=0.255549
[metric_learning] step=0040 loss=0.286283
[metric_learning] step=0050 loss=0.260128
[metric_learning] step=0060 loss=0.143402
[metric_learning] step=0070 loss=0.156889
[metric_learning] step=0075 loss=0.098516
moons
train accuracy: 0.6111111111111112
test accuracy: 0.6
[metric_learning] step=0001 loss=0.356441
[metric_learning] step=0010 loss=0.428865
[metric_learning] step=0020 loss=0.303805
[metric_learning] step=0030 loss=0.204626
[metric_learning] step=0040 loss=0.247740
[metric_learning] step=0050 loss=0.143872
[metric_learning] step=0060 loss=0.121123
[metric_learning] step=0070 loss=0.149328
[metric_learning] step=0075 loss=0.112340
circles
train accuracy: 0.7222222222222222
test accuracy: 0.6333333333333333
[metric_learning] step=0001 loss=0.140471
[metric_learning] step=0010 loss=0.056122
[metric_learning] step=0020 loss=0.094982
[metric_learning] step=0030 loss=0.065302
[metric_learning] step=0040 loss=0.064415
[metric_learning] step=0050 loss=0.110236
[metric_learning] step=0060 loss=0.092765
[metric_learning] step=0070 loss=0.086375
[metric_learning] step=0075 loss=0.093646
blobs
train accuracy: 0.9
test accuracy: 0.9
Quantum Convolutional Neural Network¶
Notebook: notebooks/tutorials/10-quantum-convolutional-neural-network.ipynb
Result block 1:
Train accuracy: 0.9444444444444444
Test accuracy: 1.0
Final loss: 0.42221349893739046
Result block 2:
Embedding params shape: (4, 3)
First convolution block shape: (2, 6)
Second convolution block shape: (1, 6)
Dense readout params shape: (2,)
Result block 3:
moons
train accuracy: 0.6666666666666666
test accuracy: 0.8
circles
train accuracy: 0.4666666666666667
test accuracy: 0.4
blobs
train accuracy: 1.0
test accuracy: 1.0
xor
train accuracy: 0.5333333333333333
test accuracy: 0.4
Result block 4:
Finite-shot train accuracy: 0.4666666666666667
Finite-shot test accuracy: 1.0
Finite-shot final loss: 2.022702148492722
Quantum Autoencoder¶
Notebook: notebooks/tutorials/11-quantum-autoencoder.ipynb
Result block 1:
Train compression fidelity: 0.9729841780176779
Test compression fidelity: 0.971444977321234
Train reconstruction fidelity: 0.9729841780176779
Test reconstruction fidelity: 0.9714449773212341
Final loss: 0.027027868067930227
Result block 2:
Parameter tensor shape: (2, 4, 2)
Latent qubits: 2
Trash qubits: 2
Result block 3:
correlated
train compression fidelity: 0.9725666481383045
test compression fidelity: 0.9710059982389014
test reconstruction fidelity: 0.9710059982389015
entangled
train compression fidelity: 0.4378060715512072
test compression fidelity: 0.4831713546984486
test reconstruction fidelity: 0.48317135469844863
hybrid
train compression fidelity: 0.7725086540592571
test compression fidelity: 0.8327539703789598
test reconstruction fidelity: 0.8327539703789596
Advanced Quantum Kernel and Reservoir Models¶
Notebook: notebooks/tutorials/12-advanced-quantum-kernel-and-reservoir-models.ipynb
Result block 1:
Classification dataset
+-------------+--------------------+
| Metric | Value |
+-------------+--------------------+
| dataset | two Gaussian blobs |
| train_shape | [29, 2] |
| test_shape | [13, 2] |
| classes | [0, 1] |
+-------------+--------------------+
Result block 2:
Unsupervised kernel models
+------------------------------------+----------+
| Metric | Value |
+------------------------------------+----------+
| kpca_component_count | 2 |
| one_class_accuracy_against_class_1 | 0.846154 |
| first_eigenvalue | 11.1637 |
+------------------------------------+----------+
Result block 3:
Reservoir classifier
+----------------------------+-------+
| Metric | Value |
+----------------------------+-------+
| quantum_reservoir_accuracy | 1 |
+----------------------------+-------+
Result block 4:
Regression dataset
+-------------+-----------------------------+
| Metric | Value |
+-------------+-----------------------------+
| dataset | synthetic linear regression |
| train_shape | [25, 2] |
| test_shape | [11, 2] |
+-------------+-----------------------------+
Result block 5:
Regression models
+----------------------------+----------+
| Metric | Value |
+----------------------------+----------+
| trainable_kernel_mse | 0.145834 |
| quantum_gpr_mse | 1.59169 |
| quantum_reservoir_mse | 0.642639 |
| trainable_kernel_alignment | 0.28523 |
| quantum_gpr_mean_std | 0.031456 |
+----------------------------+----------+
Result block 6:
Validation
Classification
+----------------------------+-------+
| Metric | Value |
+----------------------------+-------+
| quantum_reservoir_accuracy | 1 |
+----------------------------+-------+
Unsupervised
+------------------------------------+----------+
| Metric | Value |
+------------------------------------+----------+
| kpca_component_count | 2 |
| one_class_accuracy_against_class_1 | 0.846154 |
| first_eigenvalue | 11.1637 |
+------------------------------------+----------+
Regression
+----------------------------+----------+
| Metric | Value |
+----------------------------+----------+
| trainable_kernel_mse | 0.145834 |
| quantum_gpr_mse | 1.59169 |
| quantum_reservoir_mse | 0.642639 |
| trainable_kernel_alignment | 0.28523 |
| quantum_gpr_mean_std | 0.031456 |
+----------------------------+----------+
Passed
+--------+-------+
| Metric | Value |
+--------+-------+
| passed | True |
+--------+-------+
Model Selection and Cross-Validation¶
Notebook: notebooks/tutorials/13-model-selection-and-cross-validation.ipynb
Result block 1:
Dataset
+----------+--------+
| Metric | Value |
+----------+--------+
| dataset | moons |
| samples | 48 |
| features | 2 |
| classes | [0, 1] |
+----------+--------+
Result block 2:
Cross-validation summary
+-----------+---------------+----------+-----------+-------+
| model | mean_accuracy | ci95_low | ci95_high | folds |
+-----------+---------------+----------+-----------+-------+
| logistic | 0.8125 | 0.659716 | 0.965284 | 3 |
| reservoir | 0.5 | 0.442253 | 0.557747 | 3 |
+-----------+---------------+----------+-----------+-------+
Selection
+------------+----------+
| Metric | Value |
+------------+----------+
| best_model | logistic |
| best_score | 0.8125 |
+------------+----------+
Result block 3:
Reservoir folds
+------+-------------+------------+-------------+
| fold | train_score | test_score | fit_seconds |
+------+-------------+------------+-------------+
| 1 | 0.625 | 0.6875 | 0.0892158 |
| 2 | 0.6875 | 0.5625 | 0.0730951 |
| 3 | 0.625 | 0.625 | 0.0802958 |
+------+-------------+------------+-------------+
Result block 4:
Final split
+----------------+----------+
| Metric | Value |
+----------------+----------+
| model | logistic |
| train_size | 36 |
| test_size | 12 |
| train_accuracy | 0.861111 |
| test_accuracy | 0.833333 |
+----------------+----------+
Reproduce¶
Regenerate notebook result pages from committed notebook outputs:
python docs/pages/generate_results.py --skip-api-results