root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-27 03:44:35 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

gguf : single pass for writing tensors + refactoring writer

Browse Source
This commit is contained in:
M. Yusuf Sarıgöz 2023-08-17 16:57:50 +03:00
parent dce07c3121
commit 5f97a48fc1
1 changed files with 0 additions and 59 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

59
convert-llama-h5-to-gguf.py
View File

@ -279,65 +279,6 @@ gguf_writer.write_kv_data_to_file()
print("gguf: write tensors") print("gguf: write tensors")
gguf_writer.write_tensors_to_file() gguf_writer.write_tensors_to_file()
# tensor data
print("gguf: convert and write tensor data")
if num_parts == 0:
part_names = ("pytorch_model.bin",)
else:
part_names = (
f"pytorch_model-{n:05}-of-{num_parts:05}.bin" for n in range(1, num_parts + 1)
)
for part_name in part_names:
print("gguf: loading model part '" + part_name + "'")
model_part = torch.load(f"{dir_model}/{part_name}", map_location="cpu")
for name in model_part.keys():
data = model_part[name]
old_dtype = data.dtype
# we don't need these
if name.endswith(".rotary_emb.inv_freq"):
continue
# convert any unsupported data types to float32
if data.dtype != torch.float16 and data.dtype != torch.float32:
data = data.to(torch.float32)
data = data.squeeze().numpy()
# reverse permute these
if name.endswith(".q_proj.weight") or name.endswith(".k_proj.weight"):
data = reverse_hf_permute(data, head_count, head_count_kv)
# map tensor names
if name.endswith(".weight") and name[:-7] in tensor_map:
name = tensor_map[name[:-7]] + ".weight"
elif name.endswith(".bias") and name[:-5] in tensor_map:
name = tensor_map[name[:-5]] + ".bias"
else:
print("Can not map tensor '" + name + "'")
sys.exit()
n_dims = len(data.shape)
data_dtype = data.dtype
# if f32 desired, convert any float16 to float32
if ftype == 0 and data.dtype == np.float16:
data = data.astype(np.float32)
# TODO: Why cant we use these float16 as-is? There should be not reason to store float16 as float32
if ftype == 1 and data_dtype == np.float16 and n_dims == 1:
data = data.astype(np.float32)
# if f16 desired, convert any float32 2-dim weight tensors to float16
if ftype == 1 and data_dtype == np.float32 and name.endswith(".weight") and n_dims == 2:
data = data.astype(np.float16)
gguf_writer.write_tensor_to_file(data)
gguf_writer.close() gguf_writer.close()