ericyq

The demo does not generate the same id

4

 The demo does not generate the same id

RuntimeError: Input type (c10::Half) and bias type (float) should be the same

1

image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0]

>>>latents = net(inputs.to("cuda").float(), randomize_noise=False, return_latents=True) RuntimeError: Cannot re-initialize CUDA in forked subprocess. To use CUDA with multiprocessing, you must use the 'spawn' start method 1st method torch.multiprocessing.set_start_method("spawn", force=True) or 2nd...

The cuda version is too high

5

cuda 12.1 Can it be reduced to version 11.7?

demo error: bert-base-uncased

2

>>>dynrefer_inference.py model = task.build_model(cfg) tokenizer = BertTokenizer.from_pretrained("bert-base-uncased") OSError: Can't load tokenizer for 'bert-base-uncased'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory...

demo error

2

>>>python demo/dynrefer_inference.py DynRefer/dynrefer/common/evaluation/eval_densecap.py", line 9, in from .meteor.meteor import Meteor ModuleNotFoundError: No module named 'dynrefer.common.evaluation.meteor'

gradio_demo.py huggingface/modules/transformers_modules/mPLUG-Owl3-7B-240728/x_sdpa.py", line 53, in forward o = F.scaled_dot_product_attention(q, k, v,

what's mean to n_ctx = 4?

1

what's mean to n_ctx = 4? n_cls_ctx = 4 ctx_init = "A photo of a X X X X person." n_ctx = 4 means person index is 4 bits, is...

how to load clip pretrained model

1

CUDA_VISIBLE_DEVICES=0 python train.py --config_file configs/person/cnn_base.yml

RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED

1

def stem(x): for conv, bn in [(self.conv1, self.bn1), (self.conv2, self.bn2), (self.conv3, self.bn3)]: x = self.relu(bn(conv(x))) x = self.avgpool(x) return x error: self.bn1 = nn.BatchNorm2d(width // 2) RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED