theluyuan/RWKV-Runner
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update convert_safetensors.py

Browse Source
This commit is contained in:
josc146 2023-12-06 23:08:40 +08:00
parent f7227cd1c1
commit 053a08f5b7
1 changed files with 16 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
backend-python/convert_safetensors.py vendored
View File

@ -25,7 +25,7 @@ def rename_key(rename, name):
return name return name
def convert_file(pt_filename: str, sf_filename: str, transpose_names=[], rename={}): def convert_file(pt_filename: str, sf_filename: str, rename={}, transpose_names=[]):
loaded = torch.load(pt_filename, map_location="cpu") loaded = torch.load(pt_filename, map_location="cpu")
if "state_dict" in loaded: if "state_dict" in loaded:
loaded = loaded["state_dict"] loaded = loaded["state_dict"]
@ -34,12 +34,14 @@ def convert_file(pt_filename: str, sf_filename: str, transpose_names=[], rename=
# for k, v in loaded.items(): # for k, v in loaded.items():
# print(f'{k}\t{v.shape}\t{v.dtype}') # print(f'{k}\t{v.shape}\t{v.dtype}')
loaded = {rename_key(rename, k).lower(): v.contiguous() for k, v in loaded.items()}
# For tensors to be contiguous # For tensors to be contiguous
for k, v in loaded.items(): for k, v in loaded.items():
for transpose_name in transpose_names: for transpose_name in transpose_names:
if transpose_name in k: if transpose_name in k:
loaded[k] = v.transpose(0, 1) loaded[k] = v.transpose(0, 1)
loaded = {rename_key(rename, k).lower(): v.contiguous() for k, v in loaded.items()}
loaded = {k: v.clone().half().contiguous() for k, v in loaded.items()}
for k, v in loaded.items(): for k, v in loaded.items():
print(f"{k}\t{v.shape}\t{v.dtype}") print(f"{k}\t{v.shape}\t{v.dtype}")
@ -60,8 +62,18 @@ if __name__ == "__main__":
convert_file( convert_file(
args.input, args.input,
args.output, args.output,
["lora_A"], rename={
{"time_faaaa": "time_first", "lora_A": "lora.0", "lora_B": "lora.1"}, "time_faaaa": "time_first",
"time_maa": "time_mix",
"lora_A": "lora.0",
"lora_B": "lora.1",
},
transpose_names=[
"time_mix_w1",
"time_mix_w2",
"time_decay_w1",
"time_decay_w2",
],
) )
print(f"Saved to {args.output}") print(f"Saved to {args.output}")
except Exception as e: except Exception as e: