synthcity

synthcity copied to clipboard

Time series with missingness

5

## Feature Description I have developed a GAN framework for generating irregularly sampled time series with missing values, however, I cannot add it to synthcity as it does not support...

hojjatkarami

I had a question about interpreting [synthcity's k anonymity metric](https://github.com/vanderschaarlab/synthcity/blob/943fa280687d236d783e53f40302838f5924f422/src/synthcity/metrics/eval_privacy.py#L70) for a synthetic dataset. Consider the following example train dataset: | | Age | Gender | Zip Code | Medical...

amad-person

Allow use of Pytorch 2x, Pandas 2x, and Numpy 1.26

1

## Description Allow installing Pytorch 2 and pandas 2. This solves four open issues #172, #171, #234 #245 ## Affected Dependencies Only removed restrictions on pytorch, numpy, and pandas versions...

bvanbreugel

PydanticUserError in schema.py

1

## Description In tutorial0 (or wherever Plugins is used) I encounter the following error: A non-annotated attribute was detected: `protected_cols = ['seq_id']`. All model fields require a type annotation; if...

AsafKov

Session crashed and installation issue on google colab

1

## Question **Hi developer, I have an issue when running tutorial examples on google colab.** ## Further Information I think I successfully installed the package but when running the first...

yunwezhang

TabDDPM generating NaNs for large datasets

1

## Question When fitting TabDDPM for more than a single iteration NaNs are generated which leads to a `ValueError` during sampling here: https://github.com/vanderschaarlab/synthcity/blob/41e6e5acfd886dd4ebc0528039e9395a2a93b380/src/synthcity/plugins/core/models/tabular_ddpm/gaussian_multinomial_diffsuion.py#L954-L955 ## Further Information I'm fitting TabDDPM to...

HLasse

Benchmark problem on survivalgan

1

## What? I want to evaluate the synthetic data generated but I have this code: ``` from synthcity.benchmark import Benchmarks df=df.loc[df['OS_time']>0,:] loader = SurvivalAnalysisDataLoader( df, target_column="os_event", time_to_event_column="OS_time", ) score =...

BioinfoLabImmuno

## Question How is splitting the real input dataset into train and test parts beneficial model evaluation? ## Further Information Based on line 130 and lines 232-273 in file `eval.py`,...

ms2017

Upgrade PyTorch requirements from ">=1.10.0,<2.0" to "2.0.1"

3

https://github.com/vanderschaarlab/synthcity/blob/7c1d9b5d3397334ce8d79ffd48ac31449d72b730/setup.cfg#L36 Hello synthcity Team, Are there any plans to update the library file from ">=1.10.0,

kacpersh

## Question Hi, thanks again for an excellent package. Is it possible to upgrade the upper bound of the numpy version to e.g. 1.26.0? https://github.com/vanderschaarlab/synthcity/blob/ab47bcf38cf7edec946fc630185850dc6eceb5fa/setup.cfg#L39 At the moment this is...

HLasse