@GRAPElib

🍇 GRAPE is a Rust/Python library for high-performance Graph Representation learning, Predictions and Evaluations.

 Milano, Lombardia
github.com/AnacletoLAB/grape
Joined June 2022
  • Tweets53
  • Following0
  • Followers44
36Media
#GraphML 🍇 preprint is on @arxiv! ❤️ to @LucaCappellett6, @Zommiommy, Casiraghi, @RavanmehrVida, @Tiff_callahan, @realmarcin, @chrismungall, @pnrobins, @justaddcoffee & Valentini 📍@MonarchInit, @ConsorzioCINI, @LaStatale, @jacksonlab & @BerkeleyLab 📜:arxiv.org/pdf/2110.06196.pdf
5
8
13
Woohoo! #GRAPE just hit 100 stars on GitHub! Thank you to all the amazing developers who have supported our graph representation learning library. We couldn't have done it without you! 🍇💜🍇 #opensource #machinelearning
1
5
I've been asked how to use 🍇 to run and visualize @keenuniverse's implementation of @mnick's HolE model, so here's the one-liner!
2
Pushing the ✉️ of 🍇's @psresnik score implementation by computing 3T, i.e. 3*10^12, scores from @NCBI Taxonomy (2438821 nodes) upper triangular matrix. This is heavily parallelized and takes ≈3h on a 💻 with 8GBs of RAM and 96 cores.
2
2
You can now use 🍇 to compute @hp_ontology pairwise @psresnik scores, ~270M, on your notebook in about 1 second ⚡
2
9
🎥>>🗣️ #8: @MKoutrouli's FAVA functional association networks, embedded using @phanein's DeepWalk + SkipGram with Right Laplacian sampling by @LucaCappellett6 & @zommiommy Done in ~2m on my desktop! ⚡ The edge prediction looks excellent (holdout 70/30)! ❤️
3
3
10
🎥>>🗣️ #7: @justaddcoffee's KGCOVID19 node & edge labels + properties, embedded using First-order LINE ⚡
2
4
#GraphML & #MachineLearning to 🔮 #GameOfThrones 5th book using characters' interactions from all previous 📚!
1
3
💻:github.com/AnacletoLAB/grape…
Positive predictions

ALT Positive predictions

Negative predictions

ALT Negative predictions

1
🎥>>🗣️ #6! Edge prediction of @sanyabt11's NPKG (745.51K N, 7.25M E) embedded using 🍇's #Rust + #Python implementation of @tangjianpku's First-order LINE, ~30s on my PC Test set, 70/30 split 💽 from @zenodo_org ❤️ Edge types & 💻 in the 1st comment! #GraphML #MachineLearning
1
3
Edge types of NPKG. The training is unsupervised. 💻 Tutorial:github.com/AnacletoLAB/grape… 💻 NPKG #github repo:github.com/sanyabt/np-kg
The Twitter handle of 🍇's PI, Prof. Valentini, is @gg_valentini
1
🎥>>🗣️ #5! Edge types of @AstrorEnales's PharMeBINet (2.65M N, 15.88M E) embedded using 🍇's #Rust + #Python implementation of @tangjianpku's First-order LINE, ~1m on my PC 💽 from @zenodo_org❤️ Edge prediction & 💻 in the 1st comment! #GraphML #MachineLearning
1
2
3
Visualization of edge prediction on PharMeBINet, connected holdout with 70/30 split. Good separation between existing and non-existing edges is achieved, suggesting an edge prediction task could achieve good performance. 💻:github.com/AnacletoLAB/grape…
1
Introducing #slurm across 🍇 #GraphML evaluation pipelines to make 🧪 happen faster! We thank @pnrobins & @lnrekerle for testing it on @jacksonlab's #HPC! ❤️ 💻:github.com/AnacletoLAB/grape…
Flux diagram of how the SLURM parallelization of the holdouts is achieved. GRAPE parallelizes the computation of holdouts across different SLURM nodes.

ALT Flux diagram of how the SLURM parallelization of the holdouts is achieved. GRAPE parallelizes the computation of holdouts across different SLURM nodes.

Script showing how the parallelization can actually be executed. Simply set the NUMBER_OF_SLURM_NODES parameter, and you are set to go.

ALT Script showing how the parallelization can actually be executed. Simply set the NUMBER_OF_SLURM_NODES parameter, and you are set to go.

2
4
From 🍇 V 0.1.19, we integrated @bioregistry ❤️ in the graph retrieval, which includes all of the @OBOFoundry #KnowledgeGraph made available by @harry_caufield's KGOBO! 🚄 Just enable the flag, and you are set to go!
1
1
6
🎥>>🗣️ #4! Directed English @Wikipedia (17.37M N, 140.81M E) embedded using 🍇's #Rust + #Python implementation of @tangjianpku's Second-order LINE, ~20m on my PC 💽 from @Wikimedia❤️ 💻:github.com/AnacletoLAB/grape… #GraphML #MachineLearning
1
2
4
🎀 Round! Since this is a direct graph, I believed the Second-order LINE model was best suited. Nonetheless, I have now completed running the First-order version, which rocks!
Tiny overfitting planets 🪐🌎 Even overfitting can be beautiful ❤️ #GraphML #MachineLearning
2
9
🎥>>🗣️ #3! @realmarcin's KGMicrobe (276K N, 525K E) embedded using 🍇's #Rust + #Python implementation of Jeffrey Pennington's GloVe model with @phanein's DeepWalk & Walklets sample, unsupervised training. 💻:github.com/AnacletoLAB/grape… 🧵(1/2) DeepWalk #GraphML #MachineLearning
1
4
3
🧵(2/2) Walklets
2
Load more