import ANet from ANet import ANet A_model = ANet() try: A_model.load_state_dict(torch.load(os.path.join(ROOT_DIR, 'multisource_cocktail/ANet/ANet_raw_2.pkl'))) except Exception as e: print(e, "A-model not available") # print(A_model) import conv_fc from conv_fc import ResDAE Res_model = ResDAE() try: Res_model.load_state_dict(torch.load(os.path.join(ROOT_DIR, 'multisource_cocktail/DAE/DAE_multi_2.pkl'))) except Exception as e: print(e, "Res-model not available") # print(Res_model) # ============================================ # optimizer # ============================================ criterion = nn.MSELoss() # ============================================
torch.load( os.path.join( ROOT_DIR, 'multisource_cocktail/ANet/ANet_multi_2_trained.pkl'))) else: A_model.load_state_dict( torch.load( os.path.join(ROOT_DIR, 'multisource_cocktail/ANet/ANet_raw_2.pkl'))) except Exception as e: print(e, "A-model not available") # print(A_model) from conv_fc import ResDAE Res_model = ResDAE() try: if ATTEND: Res_model.load_state_dict( torch.load( os.path.join(ROOT_DIR, 'multisource_cocktail/DAE/DAE_multi_2.pkl'))) else: Res_model.load_state_dict( torch.load( os.path.join(ROOT_DIR, 'multisource_cocktail/DAE/DAE_raw_2.pkl'))) except Exception as e: print(e, "Res-model not available") # print(Res_model)
ROOT_DIR, 'multisource_cocktail/ANet/ANet_multi_2_trained.pkl'))) except Exception as e: print(e, "A-model not available") else: try: A_model.load_state_dict( torch.load( os.path.join(ROOT_DIR, 'multisource_cocktail/ANet/ANet_raw_2.pkl'))) except Exception as e: print(e, "A-model not available") # print(A_model) from conv_fc import ResDAE Res_model = ResDAE() if reuse: try: Res_model.load_state_dict( torch.load( os.path.join(ROOT_DIR, 'multisource_cocktail/DAE/DAE_multi_2.pkl'))) except Exception as e: print(e, "Res-model not available") else: try: Res_model.load_state_dict( torch.load( os.path.join(ROOT_DIR, 'multisource_cocktail/DAE/DAE_multi_2.pkl'))) except Exception as e: