文件名称:CompRess:压缩表示形式以进行自我监督学习
文件大小:14.27MB
文件格式:ZIP
更新时间:2024-05-30 03:10:05
computer-vision clustering nearest-neighbors representation-learning knn
CompRess:通过压缩制图表达进行自我监督的学习 该存储库是的官方实现 项目网页。 @Article{abbasi2020compress, author = {Koohpayegani, Soroush Abbasi and Tejankar, Ajinkya and Pirsiavash, Hamed}, title = {CompRess: Self-Supervised Learning by Compressing Representations}, journal = {Advances in neural information processing systems}, year = {2020}, } 要求 按照安装PyTorch和ImageNet数据集。 我们使用Python 3.7进行实验。 安装PyTorch( ) 要运行N
【文件预览】:
CompRess-master
----models()
--------resnet.py(6KB)
--------resnet50x4.py(13KB)
--------alexnet.py(1KB)
--------mobilenet.py(6KB)
----train_student.py(10KB)
----eval_cluster_alignment.py(21KB)
----tools.py(4KB)
----docs()
--------_layouts()
--------_includes()
--------assets()
--------_config.yml(130B)
--------index.html(13KB)
--------bib.txt(242B)
--------README.md(29B)
--------libs()
----cache_feats.py(6KB)
----train_student_one_queue.py(9KB)
----train_kmeans.py(12KB)
----eval_linear.py(14KB)
----nn()
--------compress_loss.py(6KB)
----util.py(2KB)
----Readme.md(6KB)
----LICENSE(1KB)
----kmeans.py(5KB)
----train_student_without_momentum.py(8KB)
----eval_knn.py(12KB)