nuluh
b2bf1b0e31
feat(notebooks): Add type hints for data lists, create preview function for STFT visualization, and save plot for Sensor B. Introduce test section with AU data processing.
2025-08-28 10:39:56 +07:00
nuluh
7a7b2a41af
fix(notebooks): update variable names for clarity and add timing evaluation for model predictions on Dataset B
2025-08-28 10:39:55 +07:00
nuluh
3ef656dd28
refactor(notebooks): remove redundant confusion matrix code for Sensor A and update reporting for Sensor B
2025-08-28 10:32:32 +07:00
nuluh
f45614b6d9
fix(notebooks): update import statement for hann function and add window variable display
2025-08-20 08:03:22 +07:00
nuluh
e2a4c80d49
Merge branch 'feat/103-feat-inference-function' into dev
2025-08-19 06:05:32 +07:00
nuluh
855114d633
refactor(notebooks): clean up imports, adjust damage case processing, and improve model training structure
...
- Removed unnecessary imports (os, pandas, numpy) from the STFT notebook.
- Adjusted the number of damage cases in the multiprocessing pool to correctly reflect the range.
- Updated model training code for Sensor B to ensure consistent naming and structure.
- Cleaned up commented-out code for clarity and maintainability.
2025-08-17 23:39:57 +07:00
nuluh
4a1c0ed83e
feat(src): implement inference function with damage probability calculations and visualization
...
Closes #103
2025-08-17 22:21:17 +07:00
nuluh
274cd60d27
refactor(src): update generate_df_tuples function signature to include type hints for better clarity
2025-08-11 18:49:41 +07:00
nuluh
9f23d82fab
fix(src): correct file writing method in process.stft.process_damage_case function to fix incorrect first column name
...
Closes #104
2025-08-11 13:17:46 +07:00
nuluh
a8288b1426
refactor(src): enhance compute_stft function with type hints, improved documentation by moving column renaming process from process_damage_case to compute_stft
2025-08-11 13:15:48 +07:00
nuluh
860542f3f9
refactor(src): restructure compute_stft function to be pure function and include return parameters and improve clarity
2025-08-10 20:02:45 +07:00
nuluh
0e28ed6dd0
feat(notebooks): add cross-dataset validation for Sensor A and Sensor B models
...
Closes #74
2025-07-28 16:41:54 +07:00
nuluh
3e2b153d11
refactor(stft): comment out unused imports and update SVM model loading for consistency
2025-07-28 05:22:24 +07:00
nuluh
3cbef17b0c
feat(model_selection): add timing for model training and validation processes
2025-07-28 05:20:10 +07:00
nuluh
9b018efc15
refactor(notebooks): update STFT notebook to improve clarity and structure of sensor evaluation sections
2025-07-24 17:00:31 +07:00
nuluh
086032c250
refactor(notebooks): clean up to be more readable notebooks
2025-07-18 19:28:43 +07:00
nuluh
f6c71739df
refactor(ml): clean up model_selection.py by removing unused code and improving function structure
2025-07-18 19:27:46 +07:00
nuluh
18824e05c0
refactor(ml): update inference calls to use new model structure and improve clarity
2025-07-17 00:18:01 +07:00
nuluh
2504157b29
feat(src): replace convert.py to src/data_preprocessing.py and fix some functions prefix parameter
2025-07-02 03:25:18 +07:00
nuluh
5ba628b678
refactor(src): make compute_stft and process_damage_case to be pure function that explicitly need STFT arguments to be passed
2025-07-01 14:32:52 +07:00
nuluh
a93adc8af3
feat(notebooks): minimize stft.ipynb notebooks and add STFT data preview plot.
...
- Consolidated import statements for pandas and matplotlib.
- Updated STFT plotting for Sensor 1 and Sensor 2 datasets with improved visualization using pcolormesh.
- Enhanced subplot organization for better clarity in visual representation.
- Added titles and adjusted layout for all plots.
2025-06-30 01:36:44 +07:00
nuluh
c2df42cc2b
feat(ml): add XGBoost model to inference options and update commented inference calls
2025-06-27 10:35:27 +07:00
nuluh
465ed121f9
feat(notebooks): training model with new alternative undamaged (label 0) data
2025-06-27 10:34:23 +07:00
nuluh
d6975b4817
feat(src): update damage base path and adjust test run logic for damage case processing for undamage case new method
2025-06-27 10:33:54 +07:00
nuluh
9921d7663b
feat(src): add inference script for model evaluation
2025-06-24 14:08:38 +07:00
nuluh
459fbcc17a
refactor(notebooks): visualization for sensor analysis and streamline data processing
2025-06-24 14:08:02 +07:00
nuluh
5041ee3feb
feat(src): add confusion matrix plotting and label percentage calculation
2025-06-24 14:06:56 +07:00
nuluh
114ab849b9
feat(src): Add confusion matrix plotting function for model evaluation
2025-06-24 00:27:15 +07:00
nuluh
6196523ea0
feat(notebooks): Add confusion matrix plotting loop for Sensor 1 models
2025-06-21 01:10:03 +07:00
nuluh
18892c1188
WIP(notebooks): Add SVM with StandardScaler and PCA to sensor model definitions
2025-06-18 08:31:55 +07:00
nuluh
a7d8f1ef56
fix(data): Fix pool mapping to include undamaged case and add csv header separator line for Excel compatibility
2025-06-18 08:25:01 +07:00
nuluh
4b0819f94e
feat(notebooks): Enhance STFT notebook and model selection functionality
...
- Updated paths in the STFT notebook to reflect new data files.
- Improved plotting aesthetics for combined plots and added grid lines.
- Introduced a 3D spectrogram visualization for better data representation.
- Refactored model training function to include error handling and model export functionality.
- Adjusted model training calls to include export paths for saved models. Closes #90
- Added additional markdown cells for better documentation and clarity in the notebook.
2025-06-12 03:35:21 +07:00
nuluh
7da3179d08
refactor(nb): Create and implement helper function train_and_evaluate_model
2025-05-29 22:57:28 +07:00
nuluh
254b24cb21
feat(viz): Update plotting for STFT data visualization with color map 'jet' and added color bar
2025-05-29 20:35:35 +07:00
Rifqi D. Panuluh
d151062115
Add Working Milestone with Initial Results and Model Inference ( #82 )
...
* wip: add function to create stratified train-test split from STFT data
* feat(src): implement working function for dataset B to create ready data from STFT files stft_files and add setup.py for package configuration
* feat(notebook): Update variable names for clarity, remove unused imports, and streamline data processing. Implement data concatenation using pandas concat for efficiency. Add validation steps for Dataset B and improve model training consistency across sensors.
* fix(.gitignore): add rule to ignore egg-info directories and ensure proper formatting
* docs(README): add instructions for running stft.ipynb notebook
* feat(notebook): Add evaluation metrics and confusion matrix visualizations for model predictions on Dataset B. Remove commented-out code and integrate data preparation using create_ready_data function.
---------
Co-authored-by: nuluh <dam.ar@outlook.com >
2025-05-24 01:30:10 +07:00
nuluh
c8509aa728
fix(notebooks): fix out of index stft plotting iteration
2025-04-22 10:55:34 +07:00
nuluh
db2947abdf
fix(data): fix the incorrect output of scipy.stft() data to be pandas.DataFrame shaped (513,513) along with its frequencies as the index and times as the columns (transposed) instead of just the magnitude that being flattened out; add checks for empty data and correct file paths for sensor data loading.
...
Closes #43
2025-04-20 14:45:38 +07:00
nuluh
8ed1437d6d
Merge branch 'main' of https://github.com/nuluh/thesis
2025-03-16 14:12:11 +07:00
nuluh
96556a1186
```
...
No code changes detected.
```
2025-03-16 14:07:56 +07:00
nuluh
b890e556cf
fix(notebook): correct execution counts and update file naming conventions for STFT processing
...
Closes #27
2025-03-11 19:08:56 +07:00
nuluh
fa6e1ff72b
refactor(notebook): seperate process_stft function to individual code cell.
2025-03-08 11:04:37 +07:00
nuluh
a2e339a0a0
feat: Implement STFT verification for individual test runs against aggregated data
2024-12-13 16:30:06 +07:00
nuluh
2decff0cfb
Closes #24
...
feat(stft): Implement STFT processing for vibration data with multiprocessing support to include all the data for training process instead of just using `TEST1` only
2024-12-13 16:29:08 +07:00
nuluh
8b4eedab8a
Closes #26
...
feat: Specify `fs` when calling `scipy.signal.stft`
2024-12-09 00:49:25 +07:00
nuluh
832b6c49db
feat(notebook): Implement STFT with Hann windowing. Closes #22
2024-10-21 19:08:46 +07:00
nuluh
9618714d3c
feat: Prepare all damage cases vibration record data to be merged inside two variables "signal_sensor1" and "signal_sensor2". Closes #23
2024-10-19 15:32:05 +07:00
nuluh
2f54e91197
feat: Add absolute value option to time feature extraction
2024-09-03 15:39:44 +07:00
nuluh
758255a24e
feat(notebooks): Implement Time-domain feature extraction with real data from QUGS
2024-09-03 12:52:40 +07:00
nuluh
0306f28a68
docs(notebooks): add extract_numbers docstring
2024-09-03 11:09:47 +07:00
nuluh
adde35ed7e
feat(notebook): Normalize the data by calculating the relative value between two sensors. Along with it, MinMaxScaler and StandardScaler are applied and visualize with Seaborn's Pair Plot.
...
Closes #15
2024-09-01 14:50:04 +07:00
