As　the　core　algorithm　of　artificial　intelligence,　deep　learning　has　brought　new　breakthroughs　and　opportunities　to　all　walks　of　life.　This　paper　summarizes　the　principles　of　deep　learning　algorithms　such　as　Autoencoder　(AE),　Boltzmann　Machine　(BM),　Deep　Belief　Network　(DBM),　Convolutional　Neural　Network　(CNN),　Recurrent　Neural　Network　(RNN)　and　Recursive　Neural　Network　(RNN).　The　characteristics　and　differences　of　deep　learning　frameworks　such　as　Tensorflow,　Caffe,　Theano　and　PyTorch　are　compared　and　analyzed.　Finally,　the　application　and　performance　of　hardware　platforms　such　as　CPU　and　GPU　in　deep　learning　acceleration　are　introduced.　In　this　paper,　the　development　and　application　of　deep　learning　algorithm,　framework　and　hardware　technology　can　provide　reference　and　basis　for　the　selection　of　deep　learning　technology.　©　2022,　The　Author(s).