既有适合小白学习的零基础资料,也有适合3年以上经验的小伙伴深入学习提升的进阶课程,涵盖了95%以上大数据知识点,真正体系化!
由于文件比较多,这里只是将部分目录截图出来,全套包含大厂面经、学习笔记、源码讲义、实战项目、大纲路线、讲解视频,并且后续会持续更新
m_models = {tmp:nn.Sequential(m_models[tmp], nn.Linear(m_models[tmp].out_features, 3))
for tmp in dataset_train}
m_models2 = {tmp:nn.Sequential(m_models[tmp], nn.Linear(m_models[tmp].out_features, 5))
for tmp in dataset_train if tmp == “target”}
‘’’
m_models = {‘source’:nn.Sequential(m_models1[‘source’],
nn.Linear(m_models1[‘source’].out_features, 2))}
print(m_models)
m_models2 = {‘target’:nn.Sequential(m_models1[‘target’],
nn.Linear(m_models1[‘target’].out_features, 5))}
print(m_models2)
m_models.update(m_models2)
print(“dataset_train:”, dataset_train)
print( m_models)
‘’’
for tmp in dataset_train:
print(tmp)
print(m_models[tmp])
print({tmp:m_models[tmp]})
m_models = {tmp:nn.Sequential(m_models[tmp[‘souce’]],
nn.Linear(m_models[tmp[‘souce’]].out_features, 5))}
m_models2 = {tmp:nn.Sequential(m_models[tmp], nn.Linear(m_models[tmp].out_features, 5))}
print(“tmp”, tmp)
m_models = {tmp:nn.Sequential(m_models[tmp], nn.Linear(m_models[tmp].out_features, 5))}
if tmp == “source”:
m_models = {tmp:nn.Sequential(m_models[tmp], nn.Linear(m_models[tmp].out_features, 5))}
print(‘tmp’, nn.Sequential(m_models[tmp], nn.Linear(m_models[tmp].out_features, 2)))
print(“m_models:”, m_models)
else:
print(dataset_train[tmp])
m_models2 = {tmp:nn.Sequential(m_models[tmp], nn.Linear(m_models[tmp].out_features, 5))}
print(t.get(‘d’))
m_models.update(m_models2)
‘’’
load pretrained parameters
for tmp in m_models:
m_models[tmp].load_state_dict(torch.load(os.path.join(module_path[tmp], ‘best_model.pth’)))
print(‘m_models:’,m_models)
print(‘m_models[source]:’,m_models[‘source’])
print(‘m_models[source][0]’,m_models[‘source’][0])
base_cross_stitch_net = cross_stitch_network.CrossStitchNetwork(m_models[‘source’][0], m_models[‘target’][0])
print(‘base_cross_stitch_net:’,base_cross_stitch_net)
既有适合小白学习的零基础资料,也有适合3年以上经验的小伙伴深入学习提升的进阶课程,涵盖了95%以上大数据知识点,真正体系化!
由于文件比较多,这里只是将部分目录截图出来,全套包含大厂面经、学习笔记、源码讲义、实战项目、大纲路线、讲解视频,并且后续会持续更新
更新**