Hongrui CHEN

Will the implementation of SSFL be included in FedML?

1

As the title asked, will the implementation of SSFL be included in FedML? I only see this paper in the list.

logging.info("test_dl_global number = " + str(len(train_data_global))) should it be 'test_data_global' instead of 'train_data_global' ?

https://github.com/FedML-AI/FedML/tree/master/data/lending_club_loan need to download dataset from the link https://www.kaggle.com/wendykan/lending-club-loan-data, but this link is not valid anymore. Can someone has downloaded the files to share it again? Thank you !

Where is the ABE and TB ?

2

I read your paper, and saw the table 2. And I found BN(BadNets), ESB(TrojanNet), TNN(TrojanNN), RB(reflection backdoor), LB(latent backdoor), IMC.  But where is the code for TB (Blended attack...

Could you please provide the code for train and neuron selection ?

7

we are replicating you results, but the trigger generation part always return high loss trigger pattern. Could you please provide the code for train and neuron selection ? this is...

about txt

1

Could you please provide those 2 txt file in this screenshot?

Do you use test data in neuron selection part (training)?

1

Here the sunrise_test is the data you use for finally test the ASR.  Do you use the data for test in neuron selection process? (as below)  

BUG in inference.py

1

This line, assert ckpt_path in MODELS, f"Choose from {list(MODELS.keys())}" and "e4t-diffusion-ffhq-celebahq-v1" is the only key of MODELS. So, in function load_e4t_unet, if os.path.exists(ckpt_path) is False, you WILL get a assert...

Need help... OOM with 2 RTX3090 (bs=2)

1

the accelerate yaml file ``` compute_environment: LOCAL_MACHINE distributed_type: MULTI_GPU downcast_bf16: 'no' gpu_ids: all machine_rank: 0 main_training_function: main mixed_precision: fp8 num_machines: 1 num_processes: 2 rdzv_backend: static same_network: true tpu_env: [] tpu_use_cluster:...

Question about Flux attention implementation

1

Hi, When I read the code at https://github.com/kohya-ss/sd-scripts/blob/5dff02a65da38c411ee679821504ce947d2abd7d/library/sd3_models.py#L545-L565 I see 3 different optimization approach for attention block in sd3. But for Flux model, which also have similar MMDiT block, I...