GraSPy: Graph Statistics in Python
Authors: Jaewon Chung, Benjamin D. Pedigo, Eric W. Bridgeford, Bijan K. Varjavand, Hayden S. Helm, Joshua T. Vogelstein
JMLR 2019 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Given the connectomes of the Drosophila larva left and right mushroom bodies, one natural question to ask is: how similar are these graphs (Eichler et al., 2017)? We can frame this question as whether these graphs are generated from the same distribution of latent positions (Tang et al., 2014). We can use the latent distribution test to test this hypothesis: ... p-value: 0.002 |
| Researcher Affiliation | Academia | Jaewon Chung1, EMAIL Benjamin D. Pedigo1, EMAIL Eric W. Bridgeford2 EMAIL Bijan K. Varjavand1 EMAIL Hayden S. Helm3 EMAIL Joshua T. Vogelstein1, 3, 4, EMAIL 1Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21218 2Department of Biostatistics, Johns Hopkins University, Baltimore, MD 21218 3Center for Imaging Science, Johns Hopkins University, Baltimore, MD 21218 4Kavli Neuroscience Discovery Institute, Institute for Computational Medicine, Johns Hopkins University, Baltimore, MD 21218 |
| Pseudocode | No | The paper describes the functionality of the GraSPy library and its modules, along with a high-level illustration of procedures in Figure 1. However, it does not include any formal pseudocode blocks or algorithms with structured steps. |
| Open Source Code | Yes | Gra SPy can be downloaded from Python Package Index (Py Pi), and is released under the Apache 2.0 open-source license. The documentation and all releases are available at https://neurodata.io/graspy. |
| Open Datasets | Yes | from graspy.datasets import load_drosophila_left , load_drosophila_right ... Given the connectomes of the Drosophila larva left and right mushroom bodies ... (Eichler et al., 2017) |
| Dataset Splits | No | The paper discusses performing a latent distribution test on two specific graphs (Drosophila left and right mushroom bodies) but does not describe any training, validation, or test splits for machine learning models. The problem is framed as a hypothesis test between two given graphs, not a supervised learning task requiring data splits. |
| Hardware Specification | No | The paper describes a software library and its functionalities, providing a code example for a hypothesis test. However, it does not mention any specific hardware (e.g., GPU/CPU models, memory) used to run the experiments or the code example. |
| Software Dependencies | Yes | All descriptions here correspond to Gra SPy version 0.1.1. ... Table 1: Qualitative comparison of Python graph analysis packages. Gra SPy is largely complementary to existing graph analysis packages in Python. Gra SPy does not implement many of the essential algorithms for operating on graphs (rather, it leverages Network X for these implementations). ... Gra SPy 0.1.1 Network X 2.3 graph-tool 2.29 Snap.py 4.1 |
| Experiment Setup | Yes | We can use the latent distribution test to test this hypothesis: ... ldt = Latent Distribution Test (n_components =3, n_bootstraps =500) |