記事

Accelerate Innovation With Teradata AI Unlimited

Move breakthrough ideas quickly from testing to production and deliver AI innovations faster.

Vidhan Bhonsle
Vidhan Bhonsle
2024年6月7日 4 分で読める
One person, smiling, while holding an open laptop in their hands.

Data engineers, data scientists, and software engineers are constantly exploring and experimenting with data to uncover hidden insights and drive innovation. Often, though, we find ourselves entangled in and delayed by the nuances of infrastructure, data location, and integration challenges. As a result, we lose valuable time that could be better spent on exploration, experimentation, and innovation. 

With Teradata AI Unlimited, an on-demand AI/ML engine in the cloud, developers can now innovate with complete freedom, moving breakthrough ideas quickly from testing to production and delivering AI innovations faster. 

Leveraging massively parallel processing (MPP), the ability to build and run models, and operationalization at scale, AI Unlimited offers the freedom to explore and experiment without boundaries, providing a versatile suite of powerful analytical functions covering the entire data science lifecycle. You can quickly and cost-effectively analyze data, develop AI models, and transition into production seamlessly. 

AI Unlimited is available through leading cloud service providers and spin-up or spin-down compute resources with unlimited scale as needed. With consumption-based pricing, you only pay for the time you spend exploring and making discoveries, minimizing the risk of costly overruns. You can find AI Unlimited in public preview on Amazon Web Services (AWS) and Microsoft Azure marketplaces, and in private preview, via Microsoft Fabric. To learn more about AI Unlimited via Microsoft Fabric, watch a demo video, by Janeth Graziani, Developer Advocate at Teradata. 

To explore just how simple and quick it is to deploy an AI Unlimited instance and run queries, watch the video below:


Maximize the impact of your work by seamlessly moving ideas from exploration to testing to production. 

AI Unlimited offers use cases for: 

  • Data engineers
    • Simplify and scale data transformation pipelines
    • Provide relational abstraction on top of raw or disparate data
    • Manipulate data at scale without impacting the production system
       
  • Data scientists
    • Perform real-time and predictive analytics
    • Develop and test hypotheses
    • Train models with unlimited scale
       
  • Software engineers
    • Use data and models to create applications for business analysts, such as reporting or dashboards
    • Build domain-specific applications, such as anomaly detection, customer churn prediction, and next best offer
       

Powerful ClearScape Analytics™ functions expedite AI innovation

What sets AI Unlimited apart is its structured query language (SQL) proficiency and powerful ClearScape Analytics functions. SQL handles structured queries with proficiency and precision. With a comprehensive suite of ClearScape Analytics tools, AI Unlimited enables users to seamlessly leverage advanced analytics.
 

Data Exploration

Descriptive Statistics
td_columnsummary
td_categoricalsummary
td_univariatestatistics
td_getrowswithoutmissingvalues
td_whichmin
td_whichmax
td_histogram
td_qqnorm

Statistical Tests
td_anova
td_ztest
td_ftest
td_chisq

Path and Pattern
npath
attribution
sessionize

Text Analytics
td_textparser
ngramsplitter
td_wordembeddings
td_naivebayestextclassifiertrainer
naivebayestextclassifierpredict
td_sentimentextractor
td_ifidf

Data Preparation

Handling Outlier
td_getfutilecolumns
td_outlierfilterfit
td_outlierfiltertransform
td_oneclasssvm
td_oneclasssvmpredict

Handling Missing Values
td_getrowswithmissingvalues
td_simpleimputefit
td_simpleimputetransform

Parsing data
td_traintestsplit
td_convertto
td_numapply
td_strapply
td_fillrowid
td_roundcolumns
antiselect
pack
unpack
stringsimilarity
td_unpivoting
td_pivoting

Feature Engineering

Categorical Variable Transform
td_ordinalencodingfit 
td_ordinalencodingtransform 
td_onehotencodingfit 
td_onehotencodingtransform 

Continuous Variable Transform
td_nonlinearcombinefit 
td_nonlinearcombinetransform 
td_scalefit 
td_scaletransform 
td_functionfit 
td_functiontransform 
td_polynomialfeaturefit 
td_polynomialfeaturetransform 
td_rownormalizefit 
td_rownormalizetransform 
td_bincodefit 
td_bincodetransform 
movingaverage

Dimensionality Reduction
td_randomprojectionmincomponents 
td_randomprojectionfit 
td_randomprojectiontransform 

Model Training And Forecasting

Model Training

td_glm
td_decisionforest 
td_kmeans 
td_svm
td_knn
td_xgboost 
td_oneclasssvm 
td_vectordistance

Model Evaluation And Selection

Model Scoring

td_glmpredict 
td_kmeans predict
td_decisionforestpredict 
td_svmpredict 
td_xgboostpredict 
td_oneclasssvmpredict 

Model Evaluation

td_regressionevaluator 
td_classificationevaluator 
td_silhouette 
td_roc

1. Machine learning (ML) functions 


ClearScape Analytics functions include a broad range of native in-database functions covering machine learning (ML), geospatial, time series, and digital signal processing functions as well as various data profiling, cleansing, and feature engineering functions (including statistical tests and model evaluation metrics).
 

Data Exploration

Stationarity Exploration
td_acf 
td_pacf
td_line_spectrum
td_power_spectrum

Eliminate Non-Stationary Data
td_dickey_fuller
td_difference
td_seasonalnormalize
td_smoothma
td_trackingop
td_plot
td_extract_results
td_time aware aggregates
td_temporal

Data Preparation

td_resample
td_inputvalidator
td_sinfo
td_minfo

Time Series Aggregates
primary time index
td_gettimebucket
td_time_bucket_number
td_timeseries_range
group by time
using timecode
$td_timecode_range
$td_group_by_time
average
bottom
count
delta_t
describe
first
kurtosis
last
maximum
median
minimum
mode
percentile
Median Absolute Deviation (MAD)
rank (ANSI)
skew
stddev_pop
stddev_samp
sum
top
var_pop
var_samp

Feature Engineering

td_genseries4sinusoids
td_genseries4formula
td_power_transform
td_dfft
td_idfft
td_dfft2
td_idfft2
td_dfftconv
td_dfft2conv
td_convolve
td_convolve2
td_matrixmultiply
td_undiff
td_unnormalize
td_dtw
td_dwr
td_idwt
td_iqr
td_copyart
td_filterfactorid
td_image2matrix
td_matrix2image
td_sax
td_windowdfft

Model Training and Forecasting

Arima Forecasters
td_arima_forecast
td_movavg-forecast
td_autoarima

Smoothing Forecasters
td_mamean_forecaster
td_single_exp_forecaster
td_adaptive_responsive
td_holt_winters_forecastor
td_holt_linear_forecastor

Regressor Forecasters
td_linear_regr
td_simpleexp
td_multivar_regr

Model Evaluation and Selection

Goodness of Fit
td_arima_estimate
td_durbin_watson
td_breusch_godfrey
td_goldfield_quandt
td_breusch_pagan_godfrey
td_whites_general
td_fit_metrics

Model Validation
td_arima_validate
td_signif_resid_mean
td_signif_periodicities
td_cumul_periodogram
td_portman
td_selection_criteria

2. Time series and digital signal processing functions

Seamless integration and rapid deployment

Teradata offers integration between AI Unlimited and leading cloud service providers, enabling setup and deployment of best-in-class analytics from providers in minutes.


Try AI Unlimited today

Break free from constraints, focus on exploration and experimentation, and push boundaries with AI Unlimited. 

Learn more about AI Unlimited, or try it today →

Tags

Vidhan Bhonsle について

Vidhan is a Developer Advocate at Teradata and has over a decade of experience in developer relations, including developer education.

Vidhan strives to innovate and share his experience and knowledge with the future generation of developers.

Outside of his role at Teradata, Vidhan enjoys watching football games to unwind and chatting with people, sharing tech passions, and creating meaningful connections.

Connect with Vidhan on LinkedIn!

Vidhan Bhonsleの投稿一覧はこちら

最新情報をお受け取りください

メールアドレスをご登録ください。ブログの最新情報をお届けします。



テラデータはソリューションやセミナーに関する最新情報をメールにてご案内する場合があります。 なお、お送りするメールにあるリンクからいつでも配信停止できます。 以上をご理解・ご同意いただける場合には「はい」を選択ください。

テラデータはお客様の個人情報を、Teradata Global Privacy Policyに従って適切に管理します。