68 Commits

Author SHA1 Message Date
Palash Tyagi
ed01c4b8f2 Enhance documentation with usage examples for crate::compute::models 2025-08-03 16:48:37 +01:00
Palash Tyagi
e6964795e3 Enhance documentation with usage examples for statistical routines and utilities 2025-08-03 16:48:02 +01:00
Palash Tyagi
7720312354 Improve comments for clarity in logistic regression, stats overview, PCA, correlation, descriptive statistics, and matrix tests 2025-08-02 21:59:22 +01:00
Palash Tyagi
ef322fc6a2 Refactor assertions in tests to simplify error messages for KMeans, CryptoRng, and Prng modules 2025-07-29 22:15:45 +01:00
Palash Tyagi
750adc72e9 Add missing #[cfg(test)] attribute to tests module in activations.rs 2025-07-29 21:42:47 +01:00
Palash Tyagi
5934b163f5 Refactor random number generation to use rustframe's random module 2025-07-28 20:37:08 +01:00
Palash Tyagi
252c8a3d29 Refactor KMeans module to use inbuilt random 2025-07-28 20:23:59 +01:00
Palash Tyagi
a8a532f252 Remove Spearman correlation function and unused rank import from stats module 2025-07-16 01:50:28 +01:00
Palash Tyagi
19c3dde169 Add Pearson and Spearman correlation functions to stats module 2025-07-16 01:32:18 +01:00
Palash Tyagi
a335d29347 Simplify t-test assertion in unit test for clarity 2025-07-15 01:05:32 +01:00
Palash Tyagi
b2f6794e05 Add inferential module to stats module exports 2025-07-15 01:02:20 +01:00
Palash Tyagi
5f1f0970da Implement statistical tests: t-test, chi-square test, and ANOVA with corresponding unit tests 2025-07-15 01:02:14 +01:00
Palash Tyagi
7bbfb5394f Add tests for sample variance and standard deviation calculations 2025-07-15 01:01:40 +01:00
Palash Tyagi
285147d52b Refactor variance functions to distinguish between population and sample variance 2025-07-15 01:00:03 +01:00
Palash Tyagi
64722914bd Add test for KMeans empty cluster reinitialization logic 2025-07-13 02:24:29 +01:00
Palash Tyagi
86ea548b4f Remove test for KMeans empty cluster reinitialization 2025-07-13 01:51:43 +01:00
Palash Tyagi
1bdcf1b113 Refactor test for KMeans empty cluster reinitialization to use distinct data points and remove redundant assertion 2025-07-13 01:49:15 +01:00
Palash Tyagi
7c7c8c2a16 Remove redundant assertion message in empty cluster reinitialization test 2025-07-13 01:41:41 +01:00
Palash Tyagi
4d8ed2e908 Add test for KMeans empty cluster reinitialization logic 2025-07-13 01:41:20 +01:00
Palash Tyagi
62d4803075 Simplify assertion for unique labels in KMeans tests when k equals m 2025-07-13 01:35:02 +01:00
Palash Tyagi
19bc09fd5a Refactor KMeans centroid initialization and improve handling of edge cases 2025-07-13 01:29:19 +01:00
Palash Tyagi
bda9b84987 Refactor KMeans centroid initialization to handle k=1 case by setting centroid to mean of data 2025-07-13 00:16:29 +01:00
Palash Tyagi
c24eb4a08c Relax assertion tolerance in KMeans tests to align with algorithm's convergence criteria 2025-07-12 01:47:40 +01:00
Palash Tyagi
12a72317e4 Refactor KMeans fit and predict methods for improved clarity and performance 2025-07-12 01:45:59 +01:00
Palash Tyagi
049dd02c1a Remove unreachable panic 2025-07-12 01:35:51 +01:00
Palash Tyagi
bc87e40481 Add test for variance smoothing with zero smoothing in GaussianNB 2025-07-12 01:34:08 +01:00
Palash Tyagi
eebe772da6 Add test for invalid activation count in DenseNNConfig to ensure proper configuration 2025-07-12 01:11:41 +01:00
Palash Tyagi
7b0d34384a Refactor test assertions to improve readability by removing error messages from assert macros 2025-07-12 01:06:02 +01:00
Palash Tyagi
9182ab9fca Add test for PCA fit with n_components greater than n_features to verify behavior 2025-07-12 01:00:00 +01:00
Palash Tyagi
de18d8e010 applied formatting 2025-07-12 00:56:09 +01:00
Palash Tyagi
10018f7efe Refactor covariance_matrix to improve mean calculation and add broadcasting for centered data; add tests for vertical and horizontal covariance matrices 2025-07-12 00:50:14 +01:00
Palash Tyagi
b7480b20d4 Add correlation module and update exports in stats module 2025-07-12 00:30:26 +01:00
Palash Tyagi
d5afb4e87a Refactor PCA fit method to use covariance matrix directly and improve mean calculation 2025-07-12 00:30:21 +01:00
Palash Tyagi
493eb96a05 Implement covariance functions for matrices with comprehensive tests 2025-07-12 00:29:50 +01:00
Palash Tyagi
37c0d312e5 Add tests for activation functions, initializers, and loss gradient in DenseNN 2025-07-10 23:47:36 +01:00
Palash Tyagi
e7c181f011 Refactor error handling in GaussianNB fit method to use assert instead of panic for empty class labels 2025-07-10 23:27:23 +01:00
Palash Tyagi
2cd2e24f57 Add test for gamma_cdf_func to validate behavior for negative input 2025-07-08 23:21:59 +01:00
Palash Tyagi
61aeedbf76 Simplify assertion in lower_incomplete_gamma test for clarity 2025-07-08 23:18:16 +01:00
Palash Tyagi
8ffa278db8 Add tests for uniform and binomial distributions; enhance gamma function tests 2025-07-08 23:16:19 +01:00
Palash Tyagi
b2a799fc30 Add test for median_horizontal function to validate horizontal median calculation 2025-07-08 21:03:05 +01:00
Palash Tyagi
5779c6b82d Refactor median and percentile functions to handle vertical and horizontal calculations correctly; add corresponding tests for validation 2025-07-08 21:00:19 +01:00
Palash Tyagi
a2fcaf1d52 Add tests for mean, variance, and standard deviation calculations in vertical and horizontal directions 2025-07-07 23:36:43 +01:00
Palash Tyagi
122a972a33 Add statistical distribution functions for matrices 2025-07-07 21:22:09 +01:00
Palash Tyagi
e48ce7d6d7 Add descriptive statistics functions and module integration 2025-07-07 00:38:09 +01:00
Palash Tyagi
a08fb546a9 fixed typo 2025-07-07 00:02:24 +01:00
Palash Tyagi
87d14bbf5f Moved activations module and update imports in dense_nn and logreg 2025-07-06 23:50:32 +01:00
Palash Tyagi
4648800a09 fixed incorrectly commited file 2025-07-06 21:16:57 +01:00
Palash Tyagi
96f434bf94 Add tests for DenseNN training and MSE loss calculation 2025-07-06 20:48:04 +01:00
Palash Tyagi
46abeb12a7 applied formatting 2025-07-06 20:43:01 +01:00
Palash Tyagi
75d07371b2 Increase training epochs from 5000 to 10000 for improved model performance 2025-07-06 20:16:06 +01:00