Analysis

Query Data Commons for public demographic, economic and health statistics

Version and fetch scientific datasets with DataLad and capture run provenance

Run long computations with checkpoints, fixed seeds and capped threads

Profile scientific data files across 200+ formats and summarize their structure

Generate and test hypotheses on tabular data with an LLM-driven loop

Resolve free-text labels to ontology term IDs and validate CURIEs with OLS

Fit nonlinear curves with error propagation, chi-squared checks and AIC or BIC

Build hierarchical Bayesian models with PyMC, NUTS and LOO model comparison

Analyze frequencies with FFT, Welch PSD, spectrograms, wavelets and coherence

Choose paired, rank-based and multiple-comparison tests the way referees expect

Calculate sample sizes, power curves and minimum detectable effects for studies

Reduce high-dimensional data to 2D or 3D with UMAP and HDBSCAN clustering

Analyze billion-row tabular datasets out of core with lazy DataFrame operations

Store and process large chunked N-D arrays on S3 or GCS with Zarr

Analyze CSV, parquet and JSON data with DuckDB or Polars and plot results

Speed up pandas-style DataFrame work on NVIDIA GPUs with cuDF

Compute federated statistics and histograms across NVFLARE sites

Model and publish semantic terms, relationships and measures in Auto Ontology

Pull false positive and false negative cases from VLM yes or no predictions

Read, write, format and filter Google Sheets data and manage tabs through Composio

Call vehicle, demand, market and GitHub data APIs per request with USDC payments

Analyze video comment sections to gauge content quality and find recommendations

Choose statistical tests, run analyses and report results in APA format

Inspect and validate a dataset before drawing conclusions or writing transformations

Catch price scale, ticker and date errors in market articles before you publish

Score the sentiment of financial text with FinBERT or an LLM

Run hypothesis tests, regression and power analysis with APA-style reporting

Fit OLS, GLM, logistic and ARIMA models with statsmodels diagnostics and AIC

Compare GWAS studies and run meta-analyses on effect sizes and replication

Analyze ImageJ and CellProfiler measurements with statistics for cell counts

Fit regression, Cox survival and mixed-effects models on biomedical datasets

Answer DataChain SDK API questions with method signatures and code patterns

Forecast a single time series with ARIMA models and report the results

Profile a dataset's structure, quality and readiness for analysis

Split customers into deciles to find high-value and low-value groups

Diagnose missing values in a dataset and choose how to handle them

Run binary or multiclass logistic regression and interpret the results

Forecast with Prophet, decomposing trend and seasonality in time series

Run linear regression and explain which variables drive the outcome

Model seasonal time series and forecast them with SARIMA

Screen variables one by one to shortlist candidate explanatory features

Trim extreme outlier samples and check how much the result shifts

Model how several time series influence each other with vector autoregression

Winsorize extreme values in a dataset and verify how much it changes results

Create and manage Feishu Bitable records, fields, views and batch imports

Build and maintain business knowledge network definitions for objects, relations and actions

Guide business knowledge network modeling of object types, relations and actions

Create BKN definition files for knowledge networks, object types and relation types

Query knowledge network schemas, run actions, manage metrics and check data sources