{"payload":{"pageCount":2,"repositories":[{"type":"Public","name":"marlin","owner":"IST-DASLab","isFork":false,"description":"FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.","allTopics":["kernel","quantization","4bit","llm"],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":3,"issueCount":20,"starsCount":457,"forksCount":34,"license":"Apache License 2.0","participation":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,3,6,1,0,0,2,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-07-10T03:28:13.078Z"}},{"type":"Public","name":"PanzaMail","owner":"IST-DASLab","isFork":false,"description":"","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":2,"issueCount":3,"starsCount":251,"forksCount":12,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-07-11T10:03:04.466Z"}},{"type":"Public","name":"MicroAdam","owner":"IST-DASLab","isFork":false,"description":"This repository contains code for the MicroAdam paper.","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":5,"forksCount":1,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-06-28T08:46:07.923Z"}},{"type":"Public","name":"ISTA-DASLab-Optimizers","owner":"IST-DASLab","isFork":false,"description":"","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":4,"forksCount":0,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-06-27T17:47:22.377Z"}},{"type":"Public","name":"AutoGPTQRoSA","owner":"IST-DASLab","isFork":false,"description":"","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":0,"forksCount":0,"license":"MIT License","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-06-27T08:21:51.193Z"}},{"type":"Public","name":"GridSearcher","owner":"IST-DASLab","isFork":false,"description":"GridSearcher simplifies running grid searches for machine learning projects in Python, emphasizing parallel execution and GPU scheduling without dependencies on SLURM or other workload managers.","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":0,"forksCount":0,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-06-21T13:13:19.636Z"}},{"type":"Public","name":"Mathador-LM","owner":"IST-DASLab","isFork":false,"description":"Code for the paper \"Mathador-LM: A Dynamic Benchmark for Mathematical Reasoning on LLMs\".","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":0,"forksCount":0,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-06-18T10:29:43.023Z"}},{"type":"Public","name":"sparsegpt","owner":"IST-DASLab","isFork":false,"description":"Code for the ICML 2023 paper \"SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot\".","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":1,"issueCount":13,"starsCount":662,"forksCount":84,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-05-30T14:12:02.860Z"}},{"type":"Public","name":"peft-rosa","owner":"IST-DASLab","isFork":false,"description":"A fork of the PEFT library, supporting Robust Adaptation (RoSA)","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":1,"starsCount":10,"forksCount":3,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-04-26T21:00:58.809Z"}},{"type":"Public","name":"llm-foundry","owner":"IST-DASLab","isFork":true,"description":"LLM training code for Databricks foundation models","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":0,"forksCount":504,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-04-24T11:49:40.438Z"}},{"type":"Public","name":"gptq","owner":"IST-DASLab","isFork":false,"description":"Code for the ICLR 2023 paper \"GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers\".","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":1,"issueCount":20,"starsCount":1790,"forksCount":144,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-03-27T01:46:02.429Z"}},{"type":"Public","name":"RoSA","owner":"IST-DASLab","isFork":false,"description":"","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":26,"forksCount":2,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-02-13T14:40:54.891Z"}},{"type":"Public","name":"SparseFinetuning","owner":"IST-DASLab","isFork":false,"description":"Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":2,"starsCount":35,"forksCount":5,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2024-01-15T12:56:00.527Z"}},{"type":"Public","name":"CAP","owner":"IST-DASLab","isFork":false,"description":"Repository for Correlation Aware Prune (NeurIPS23) source and experimental code","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":4,"forksCount":1,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-11-29T14:52:18.336Z"}},{"type":"Public","name":"qmoe","owner":"IST-DASLab","isFork":false,"description":"Code for the paper \"QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models\".","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":3,"starsCount":255,"forksCount":22,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-11-03T22:37:27.653Z"}},{"type":"Public","name":"TACO4NLP","owner":"IST-DASLab","isFork":false,"description":"Task aware compression for various NLP tasks","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":2,"forksCount":0,"license":null,"participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-10-09T16:57:34.386Z"}},{"type":"Public","name":"KDVR","owner":"IST-DASLab","isFork":false,"description":"Code for the experiments in Knowledge Distillation Performs Partial Variance Reduction, NeurIPS 2023","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":1,"forksCount":0,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-10-06T14:24:33.789Z"}},{"type":"Public","name":"EFCP","owner":"IST-DASLab","isFork":false,"description":"The repository contains code to reproduce the experiments from our paper Error Feedback Can Accurately Compress Preconditioners available below:","allTopics":["computer-vision","deep-learning","top-k","language-model","low-rank-approximation","second-order-optimization","ggt","adaptive-gradient-method","full-matrix-preconditioning","m-fac"],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":3,"forksCount":0,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-09-12T14:07:15.605Z"}},{"type":"Public","name":"QIGen","owner":"IST-DASLab","isFork":false,"description":"Repository for CPU Kernel Generation for LLM Inference","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":25,"forksCount":2,"license":null,"participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-07-13T16:35:04.696Z"}},{"type":"Public","name":"OBC","owner":"IST-DASLab","isFork":false,"description":"Code for the NeurIPS 2022 paper \"Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruning\".","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":3,"starsCount":90,"forksCount":11,"license":null,"participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-07-11T15:48:56.610Z"}},{"type":"Public","name":"spdy","owner":"IST-DASLab","isFork":false,"description":"Code for ICML 2022 paper \"SPDY: Accurate Pruning with Speedup Guarantees\"","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":1,"starsCount":17,"forksCount":4,"license":null,"participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-05-03T08:16:23.704Z"}},{"type":"Public","name":"CrAM","owner":"IST-DASLab","isFork":false,"description":"Code for reproducing the results from \"CrAM: A Compression-Aware Minimizer\" accepted at ICLR 2023","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":7,"forksCount":1,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-03-01T17:08:31.042Z"}},{"type":"Public","name":"gcomp_sim_strip","owner":"IST-DASLab","isFork":false,"description":"Stripped version of gcomp_sim for ML course","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":0,"forksCount":0,"license":null,"participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2023-01-18T09:46:04.510Z"}},{"type":"Public","name":"DeepLearningExamples","owner":"IST-DASLab","isFork":true,"description":"Deep Learning Examples","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":1,"forksCount":3129,"license":null,"participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2022-10-18T14:15:11.686Z"}},{"type":"Public","name":"sparse-imagenet-transfer","owner":"IST-DASLab","isFork":false,"description":"Code for reproducing the results in \"How Well do Sparse Imagenet Models Transfer?\", presented at CVPR 2022","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":8,"forksCount":0,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2022-06-03T14:54:32.105Z"}},{"type":"Public","name":"M-FAC","owner":"IST-DASLab","isFork":false,"description":"Efficient reference implementations of the static &amp; dynamic M-FAC algorithms (for pruning and optimization)","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":16,"forksCount":1,"license":"MIT License","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2022-02-23T14:52:32.010Z"}},{"type":"Public","name":"ACDC","owner":"IST-DASLab","isFork":false,"description":"Code for reproducing \"AC/DC: Alternating Compressed/DeCompressed Training of Deep Neural Networks\" (NeurIPS 2021)","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":20,"forksCount":3,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2021-11-09T12:29:54.937Z"}},{"type":"Public","name":"horovod","owner":"IST-DASLab","isFork":true,"description":"Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":0,"forksCount":2221,"license":"Other","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2021-08-02T11:18:54.023Z"}},{"type":"Public","name":"WoodFisher","owner":"IST-DASLab","isFork":false,"description":"Code accompanying the NeurIPS 2020 paper: WoodFisher (Singh &amp; Alistarh, 2020)","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":45,"forksCount":6,"license":null,"participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2021-03-08T16:24:55.909Z"}},{"type":"Public","name":"distiller","owner":"IST-DASLab","isFork":true,"description":"Neural Network Distiller by Intel AI Lab: a Python package for neural network compression research. <a href=\"https://nervanasystems.github.io/distiller\" rel=\"nofollow\">https://nervanasystems.github.io/distiller</a>","allTopics":[],"primaryLanguage":{"name":"Python","color":"#3572A5"},"pullRequestCount":0,"issueCount":0,"starsCount":0,"forksCount":799,"license":"Apache License 2.0","participation":null,"lastUpdated":{"hasBeenPushedTo":true,"timestamp":"2019-10-07T19:42:44.566Z"}}],"repositoryCount":33,"userInfo":null,"searchable":true,"definitions":[],"typeFilters":[{"id":"all","text":"All"},{"id":"public","text":"Public"},{"id":"source","text":"Sources"},{"id":"fork","text":"Forks"},{"id":"archived","text":"Archived"},{"id":"template","text":"Templates"}],"compactMode":false},"title":"IST-DASLab repositories"}