-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
1 parent
0513e98
commit 5bd5372
Showing
1 changed file
with
22 additions
and
22 deletions.
There are no files selected for viewing
This file contains 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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -4,7 +4,7 @@ | |
| <img src="logo/ntfk-logo.jpg" alt="ntfk" width=25% max-width=125px;/> | ||
| </div> | ||
|
|
||
| **NTFk** is one of the tools in the **SmartTensors** ML framework ([smarttensors.com](https://smarttensors.com)). | ||
| **NTFk** is a module of the **SmartTensors** ML framework ([smarttensors.com](https://smarttensors.com)). | ||
|
|
||
| <div style="text-align: left"> | ||
| <img src="logo/SmartTensorsNewSmall.png" alt="SmartTensors" width=25% max-width=125px;/> | ||
|
|
@@ -15,25 +15,25 @@ | |
| **NTFk** methodology allows for automatic identification of the optimal number of features (signals) present in multi-dimensional data arrays (tensors). | ||
| The number of features (tensor rank / multi-rank) along different dimensions can be estimated jointly and independently. | ||
|
|
||
| **NMFk** can be applied to perform various types of analyses of multi-dimensional data: | ||
| **SmartTensors** can be applied to perform various types of analyses of multi-dimensional data: | ||
| - Feature extraction (**FE**) | ||
| - Blind source separation (**BSS**) | ||
| - Detection of disruptions / anomalies | ||
| - Detection of disruptions/anomalies | ||
| - Image recognition | ||
| - Text mining | ||
| - Data classification | ||
| - Separation (deconstruction) of co-occurring (physics) processes | ||
| - Discovery of unknown dependencies and phenomena | ||
| - Development of reduced-order/surrogate models | ||
| - Identification of dependencies between model inputs and outputs | ||
| - Guiding the development of physics models representing the ML analyzed data | ||
| - Guiding the development of physics models representing the ML-analyzed data | ||
| - Blind predictions | ||
| - Optimization of data acquisition (optimal experimental design) | ||
| - Labeling of datasets for supervised ML analyses | ||
|
|
||
|
|
||
| **NTFk** provides high-performance computing capabilities to solve problems with Shared and Distributed Arrays in parallel. | ||
| The parallelization allows for utilization of multi-core / multi-processor environments. | ||
| The parallelization allows for the utilization of multi-core / multi-processor environments. | ||
| GPU and TPU accelerations are available through existing Julia packages. | ||
|
|
||
| **NTFk** can be employed to perform tensor decomposition using CP (Candecomp/Parafac) and Tucker methods. | ||
|
|
@@ -44,15 +44,15 @@ GPU and TPU accelerations are available through existing Julia packages. | |
| * Wotao Yin's [BCU](https://www.math.ucla.edu/~wotaoyin/papers/bcu/matlab.html) | ||
| * [TensorLab](https://www.tensorlab.net) | ||
|
|
||
| **NTFk** provides also interface to Jean Kossaifi's [Python TensorLy](http://tensorly.org/stable/index.html). | ||
| **NTFk** provides also an interface to Jean Kossaifi's [Python TensorLy](http://tensorly.org/stable/index.html). | ||
|
|
||
| **NTFk** can perform high-performance computing tensor decomposition analyses using TensorFlow, PyTorch and MXNET. | ||
| **NTFk** can perform high-performance computing tensor decomposition analyses using TensorFlow, PyTorch, and MXNET. | ||
|
|
||
| **NTFk** methodology and applications are discussed in the the papers and presentations listed below. | ||
| **NTFk** methodology and applications are discussed in the papers and presentations listed below. | ||
|
|
||
| Tensor network decompositions can be be performed using SmartTensors' [**NTNk**](https://github.com/SmartTensors/NTNk.jl) package. | ||
| Tensor network decompositions can be performed using SmartTensors' [**NTNk**](https://github.com/SmartTensors/NTNk.jl) package. | ||
|
|
||
| Matrix factorization can be be performed using SmartTensors' [**NMFk**](https://github.com/SmartTensors/NMFk.jl) package. | ||
| Matrix factorization can be performed using SmartTensors' [**NMFk**](https://github.com/SmartTensors/NMFk.jl) package. | ||
|
|
||
| ## Awards | ||
|
|
||
|
|
@@ -74,7 +74,7 @@ Pkg.add("NTFk") | |
| ``` | ||
|
|
||
| to access the latest released version. | ||
| To utilize the latest updates (commits) use: | ||
| To utilize the latest updates (commits), use: | ||
|
|
||
| ```julia | ||
| import Pkg | ||
|
|
@@ -138,7 +138,7 @@ Some of the decomposition models can theoretically lead to unique solutions unde | |
| When these conditions are not satisfied, additional minimization constraints can assist the factorization. | ||
| A popular approach is to add sparsity and nonnegative constraints. | ||
| Sparsity constraints on the elements of G reduce the number of features and their mixing (by having as many zero entries as possible). | ||
| Nonnegativity enforces parts-based representation of the original data which also allows the Tensor Decomposition results for <img src="https://latex.codecogs.com/svg.latex?\Large&space;G" /> and <img src="https://latex.codecogs.com/svg.latex?\Large&space;A_1,A_2,\ldots,A_n" /> to be easily interrelated [Cichocki et al, 2009](https://books.google.com/books?hl=en&lr=&id=KaxssMiWgswC&oi=fnd&pg=PR5&ots=Lta2adM6LV&sig=jNPDxjKlON1U3l46tZAYH92mvAE#v=onepage&q&f=false). | ||
| Nonnegativity enforces parts-based representation of the original data, which also allows the Tensor Decomposition results for <img src="https://latex.codecogs.com/svg.latex?\Large&space;G" /> and <img src="https://latex.codecogs.com/svg.latex?\Large&space;A_1,A_2,\ldots,A_n" /> to be easily interrelated [Cichocki et al, 2009](https://books.google.com/books?hl=en&lr=&id=KaxssMiWgswC&oi=fnd&pg=PR5&ots=Lta2adM6LV&sig=jNPDxjKlON1U3l46tZAYH92mvAE#v=onepage&q&f=false). | ||
|
|
||
| ## Examples | ||
|
|
||
|
|
@@ -153,7 +153,7 @@ tsize = (5, 10, 15) | |
| tucker_orig = NTFk.rand_tucker(csize, tsize; factors_nonneg=true, core_nonneg=true) | ||
| ``` | ||
|
|
||
| After that, we can compose a tensor based on this Tucker decomposition: | ||
| After that, we can compose a tensor-based on this Tucker decomposition: | ||
|
|
||
| ```julia | ||
| import TensorDecompositions | ||
|
|
@@ -163,7 +163,7 @@ T_orig .*= 1000 | |
| ``` | ||
|
|
||
| Applying **NTFk**, we can find the unknown core size of the tensor using the tensor by itself as an input only. | ||
| To do this, we explore a series of core sizes and we identify the optimal one: | ||
| To do this, we explore a series of core sizes and identify the optimal one: | ||
|
|
||
| ```julia | ||
| sizes = [csize, (1,3,4), (3,3,4), (2,2,4), (2,4,4), (2,3,3), (2,3,5)] | ||
|
|
@@ -184,9 +184,9 @@ tucker_estimated, csize_estimated = NTFk.analysis(T_orig, sizes, 3; eigmethod=[f | |
| [ Info: Estimated true core size based on the reconstruction: (2, 3, 4) | ||
| ``` | ||
|
|
||
| The final **NTFk** result is the estimated core size `(2,3,4)` which as expected matches the original unknown core size. | ||
| The final **NTFk** result is the estimated core size `(2,3,4)`, which, as expected, matches the original unknown core size. | ||
|
|
||
| **NTFk** also produces a Tucker deconstruction of this tensor with core size `(2,3,4)` which is stored as `tucker_estimated[ibest]` | ||
| **NTFk** also produces a Tucker deconstruction of this tensor with core size `(2,3,4)`, which is stored as `tucker_estimated[ibest]` | ||
|
|
||
| ## Notebooks: | ||
|
|
||
|
|
@@ -217,8 +217,8 @@ The analyzed datasets include model outputs, laboratory experimental data, and f | |
| - Induced seismicity | ||
| - Phase separation of co-polymers | ||
| - Oil / Gas extraction from unconventional reservoirs | ||
| - Geothermal exploration and produciton | ||
| - Geologic carbon storages | ||
| - Geothermal exploration and production | ||
| - Geologic carbon storage | ||
| - Wildfires | ||
|
|
||
| ## Videos: | ||
|
|
@@ -291,7 +291,7 @@ Julia uses git for package management. | |
|
|
||
| Julia uses git and curl to install the necessary packages. | ||
|
|
||
| It is important to set proxies if needed: | ||
| It is important to set proxies, if needed: | ||
|
|
||
| ```bash | ||
| export ftp_proxy=http://proxyout.<your_site>:8080 | ||
|
|
@@ -312,7 +312,7 @@ export https_proxy=http://proxyout.lanl.gov:8080 | |
| export no_proxy=.lanl.gov | ||
| ``` | ||
|
|
||
| Proxies can be also set up directly in the Julia REPL as well: | ||
| Proxies can also be set up directly in the Julia REPL: | ||
|
|
||
| ```julia | ||
| ENV["ftp_proxy"] = "http://proxyout.lanl.gov:8080" | ||
|
|
@@ -332,7 +332,7 @@ ENV["https_proxy"] = "" | |
| ENV["no_proxy"] = "" | ||
| ``` | ||
|
|
||
| In some situations, you may need to add in the `.gitconfig` file in your home directory: | ||
| In some situations, you may need to add the `.gitconfig` file in your home directory: | ||
|
|
||
| ``` | ||
| [url "[email protected]:"] | ||
|
|
@@ -352,4 +352,4 @@ git config --global url."https://".insteadOf git:// | |
| git config --global url."http://".insteadOf git:// | ||
| git config --global url."[email protected]:".insteadOf https://gitlab.com/ | ||
| git config --global url."[email protected]:".insteadOf https://github.com/ | ||
| ``` | ||
| ``` | ||