Skip to content

Fix convert-lora-to-ggml.py #2738

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Aug 23, 2023
Merged

Fix convert-lora-to-ggml.py #2738

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
33 changes: 17 additions & 16 deletions convert-lora-to-ggml.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,23 +6,22 @@
import sys
from typing import Any, Dict, Sequence, TextIO

import numpy as np
import torch

from convert import DATA_TYPE_TO_FTYPE, NUMPY_TYPE_TO_DATA_TYPE, DataType
NUMPY_TYPE_TO_FTYPE: Dict[str, int] = {"float32": 0, "float16": 1}


HF_SUBLAYER_TO_GGML = {
"self_attn.q_proj": "attention.wq",
"self_attn.k_proj": "attention.wk",
"self_attn.v_proj": "attention.wv",
"self_attn.o_proj": "attention.wo",
"mlp.gate_proj": "feed_forward.w1",
"mlp.down_proj": "feed_forward.w2",
"mlp.up_proj": "feed_forward.w3",
"input_layernorm": "attention_norm",
"self_attn.q_proj": "attn_q",
"self_attn.k_proj": "attn_k",
"self_attn.v_proj": "attn_v",
"self_attn.o_proj": "attn_output",
"mlp.gate_proj": "ffn_gate",
"mlp.down_proj": "ffn_down",
"mlp.up_proj": "ffn_up",
"input_layernorm": "attn_norm",
"post_attention_layernorm": "ffn_norm",
# "norm": "norm",
# "embed_tokens": "tok_embeddings",
# "lm_head": "output",
}


Expand All @@ -39,7 +38,7 @@ def translate_tensor_name(t: str) -> str:
sys.exit(1)

output_string = (
f"layers.{nn}.{HF_SUBLAYER_TO_GGML[sub_layer]}.weight.lora{lora_type}"
f"blk.{nn}.{HF_SUBLAYER_TO_GGML[sub_layer]}.weight.lora{lora_type}"
)
return output_string
else:
Expand All @@ -54,20 +53,22 @@ def write_file_header(fout: TextIO, params: Dict[str, Any]) -> None:
# https://opendelta.readthedocs.io/en/latest/modules/deltas.html says that `lora_alpha` is an int
# but some models ship a float value instead
# let's convert to int, but fail if lossless conversion is not possible
assert int(params["lora_alpha"]) == params["lora_alpha"], "cannot convert float to int losslessly"
assert (
int(params["lora_alpha"]) == params["lora_alpha"]
), "cannot convert float to int losslessly"
fout.write(struct.pack("i", int(params["lora_alpha"])))


def write_tensor_header(
self, name: str, shape: Sequence[int], data_type: DataType
self, name: str, shape: Sequence[int], data_type: np.dtype
) -> None:
sname = name.encode("utf-8")
fout.write(
struct.pack(
"iii",
len(shape),
len(sname),
DATA_TYPE_TO_FTYPE[NUMPY_TYPE_TO_DATA_TYPE[data_type]],
NUMPY_TYPE_TO_FTYPE[data_type.name],
)
)
fout.write(struct.pack("i" * len(shape), *shape[::-1]))
Expand Down