flashattention

app.py

@@ -473,7 +473,7 @@ def create_demo(model):
 
 def parse_args():
     parser = argparse.ArgumentParser()
-    parser.add_argument("--model_path", type=str, default=None)
+    parser.add_argument("--model_path", type=str, default="models/visualcloze-384-lora.pth")
     parser.add_argument("--precision", type=str, choices=["fp32", "bf16", "fp16"], default="bf16")
     parser.add_argument("--resolution", type=int, default=384)
     return parser.parse_args()

flux/math.py

@@ -2,8 +2,8 @@ from einops import rearrange
 import torch
 from torch import Tensor
 import torch.nn.functional as F
-from flash_attn import flash_attn_varlen_func
-from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input  # noqa
+# from flash_attn import flash_attn_varlen_func
+# from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input  # noqa
 
 
 def _upad_input(query_layer, key_layer, value_layer, query_mask, key_mask, query_length):