search

yuanli2333 / Hadamard-Matrix-for-hashing

CVPR2020/TNNLS2023: Central Similarity Quantization/Hashing for Efficient Image and Video Retrieval
MIT License
233 stars 46 forks source link

nus_wide MAP ? #6

Open xy4869 opened 4 years ago

xy4869 commented 4 years ago

May I know your specific parameters in NUS Wide？I use your library's parameters to run out the result is only 0.8188 using resnet model and 64 bit hash code in NUS Wide dataset

swuxyj commented 4 years ago

I get mAP 0.844 when use optimizer = torch.optim.RMSprop(params_list, lr = 1e-5, weight_decay=10 ** -5)

yuanli2333 commented 4 years ago

I get mAP 0.844 when use optimizer = torch.optim.RMSprop(params_list, lr = 1e-5, weight_decay=10 ** -5)

Nice try!

Kenny-Li2023 commented 3 years ago

May I know your specific parameters in NUS Wide？I use your library's parameters to run out the result is only 0.8188 using resnet model and 64 bit hash code in NUS Wide dataset

Can you share the NUS_WIDE dataset with me

swuxyj commented 3 years ago

May I know your specific parameters in NUS Wide？I use your library's parameters to run out the result is only 0.8188 using resnet model and 64 bit hash code in NUS Wide dataset

Can you share the NUS_WIDE dataset with me

NUS_WIDE dataset can be downloaded in HashNet or DSDH_PyTorch

Kenny-Li2023 commented 3 years ago

May I know your specific parameters in NUS Wide？I use your library's parameters to run out the result is only 0.8188 using resnet model and 64 bit hash code in NUS Wide dataset

Can you share the NUS_WIDE dataset with me

NUS_WIDE dataset can be downloaded in HashNet or DSDH_PyTorch

Thanks！！The format of the NUS_WIDS dataset is different from the script format in the code. How do I generate my own script please?

swuxyj commented 3 years ago

May I know your specific parameters in NUS Wide？I use your library's parameters to run out the result is only 0.8188 using resnet model and 64 bit hash code in NUS Wide dataset

Can you share the NUS_WIDE dataset with me

NUS_WIDE dataset can be downloaded in HashNet or DSDH_PyTorch

Thanks！！The format of the NUS_WIDS dataset is different from the script format in the code. How do I generate my own script please?

You can refer to this code to divide your own dataset

Kenny-Li2023 commented 3 years ago

Why is the accuracy I trained is only 0.7146 on the NUS_WIDE dataset?

Kenny-Li2023 commented 3 years ago

I get mAP 0.844 when use optimizer = torch.optim.RMSprop(params_list, lr = 1e-5, weight_decay=10 ** -5)

你好，请问你具体改了哪些参数？我直接用作者的代码但是精确度只有0.7146，论文用的NUS_WIDE不是原始NUS_WIDE数据集吧？