QueMLib is applicable for a wide range of computing platforms, and it is robust, scalable, and portable. The high performing QueMLib prediction models allow designers to focus on actual product development.
A rich set of predictive model functions
The QueMLib includes a comprehensive set of functions for machine learning, data mining, prediction, and classification, including:
- Classification algorithms such as linear and logistic regression, ID3 decision trees, C4.5 decision trees, and KNN, DTW classification.
- Cluster algorithms such as K-means, k-median, hierarchical clustering.
- Statistical functions such as mean, variance and distance calculations.
The library code readily embeds into developers' application environment, without the need of any additional infrastructure such as application management consoles, servers, or coding platforms. The QueMLib allows programmers to write, build, compile and debug code within today's popular IDEs.
The developers can rely on libraries, as the functions are full tested for accuracy. It provides maximum robustness to the developer, because it is implemented using C language, thus, making debugging faster and easier. The library accommodates remarkable error handling functionality that conveys the error state and refers corrective action as applicable.
High performance BLAS (Basic Linear Algebra Subroutines) library
Over the years, BLAS has become a standard around ML application development. QueSSence platform provides hardware accelerated BLAS libraries Level 1 and Level 2 functions. Using BLAS, developers can offload linear algebra related workload to the BLAS library for high performance.
The QueMLib supports scalability through an enhanced ability to analyze time-sequenced data, or streaming, real-time data that is not stored on-board.
QueMLib helps with:
- Improved algorithms that can analyze data sets too large to fit into system memory or exists on separate nodes.
- Memory management that ensures applications will not crash when they encounter a low memory condition.
User friendly nomenclature
The QueSSence C Machine Learning Library has user intractable code and written applications that are extremely easy to use.
Programming interface flexibility
The QueMLib takes full advantage of the features of the C language. The functions support variable-length argument lists, where the concise set of required arguments contains only information necessary for usage. Optional arguments provide added functionality and power to each function. Memory allocation can be handled by the library or by the developer. Finally, user-defined functions are implemented with interfaces that C developers will find natural.
Documentation in the QueMLib is comprehensive, clearly written and standardized. It serves several advantages, as it:
- Is available in multiple formats;
- Provides organized, easy-to-find information;
- Documents, explains, and provides references for algorithms;
- Gives examples of algorithms' usage, with sample input and result.
Rich predefined applications
QueMLib provides predefined applications such as Gesture Recognition, Anomaly detection etc., which has access to the C Machine learning library Functions.
Supervised Machine Learning - Classification Algorithms
- K-nearest neighbors -KNN
- Logistic regression
- Support vector machines - SVM
- Decision trees
Supervised Machine Learning - Regression Algorithms
- Linear regression
- Multiple linear regression
- K-nearest neighbors regression
- Support vector machines regression
Unsupervised Machine Learning - Clustering algorithms
- K-means clustering
- Hierarchical clustering
- Principal component analysis
For further information, please visit documentation section