askmuhsin
Follow
-
supa-stem-zero-shot-eval Public
Two-stage zero-shot evaluation pipeline for STEM-en-ms dataset
Jupyter Notebook UpdatedFeb 23, 2025 -
-
-
-
Transformer-Neural-Network Public
Forked from ajhalthor/Transformer-Neural-NetworkCode Transformer neural network components piece by piece
Jupyter Notebook MIT License UpdatedApr 11, 2023 -
-
-
github_stale_repo Public
use github API to find stale branches
Jupyter Notebook Apache License 2.0 UpdatedOct 23, 2022 -
pl_hydra_cog_cookiecutter Public template
Forked from ashleve/lightning-hydra-templatecookiecutter project for -> pytorch lightning + hydra + COG
Python UpdatedSep 13, 2022 -
-
legal_maxims Public
legal maxims dataset
-
learn_vit Public
a series of attempts on vision transformers
Jupyter Notebook Apache License 2.0 UpdatedAug 11, 2022 -
ocr_data_management Public
a mock system to do ocr on document, and index for the search and retrieval
Jupyter Notebook Apache License 2.0 UpdatedJun 10, 2022 -
clothing1M_experiments Public
some experiments on learning with noisy data using the clothing1M datset.
-
DETR_for_defect_detection Public
DETR (like) architecture for detecting defects in concrete surface
MIT License UpdatedMar 15, 2022 -
py38_container_template Public
a template for my py38 container, with only basic docker-compose and util scripts
Python Apache License 2.0 UpdatedFeb 18, 2022 -
-
AI-toolkit-MFS Public
A set of optimization functions and pipelines build from basics, for Neural Nets, Evolutionary Algorithms and Swarm Intelligence.
-
Weights_Heist_Flow Public
Forked from eva7wandb/Weights_Heist_FlowJupyter Notebook UpdatedNov 27, 2021 -
-
eva_training_flow Public
This is an updated version of prev training template for eva
-
-
-
-
-
-
cifar10_cnn_model Public
Forked from eva-6-3/cifar10_cnn_modelno maxpool, below 200k params, min 85%+acc, albumenation, make use of depthwise and dilated conv.
Jupyter Notebook UpdatedJun 21, 2021 -
CIFAR_85 Public
Forked from EVA6-Group-15/CIFAR_85Train model having parameters less than 200k to give validation accuracy of 85% on CIFAR 10 dataset
Jupyter Notebook MIT License UpdatedJun 19, 2021 -
eva_training_template Public
Modular training template with seperate model, training, testing, regularization, data, and misc
-
cnn_regularizations Public
Forked from eva-6-3/cnn_regularizationsComparison / interaction study on Batch/Group/Layer Normalizations and L1/L2 losses
Jupyter Notebook UpdatedJun 14, 2021