haraldwolff/llama.cpp
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

py : cleanup the code

Browse source 
- use f-strings where possible
- drop first param of encode/decode functions since "utf-8" is the default
This commit is contained in:
Pavol Rusnak 2023-03-29 21:31:24 +02:00
parent 9733104be5
commit cbef542879
6 changed files with 27 additions and 29 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
convert-ggml-to-pth.py
View file

@ -27,9 +27,9 @@ def read_tokens(fin, vocab_size):
text_len = struct.unpack("i", fin.read(4))[0]
text_bytes = fin.read(text_len)
try:
text = text_bytes.decode("utf-8")
text = text_bytes.decode()
except UnicodeDecodeError:
text = text_bytes.decode("utf-8", "replace")
text = text_bytes.decode(errors="replace")
score = struct.unpack("f", fin.read(4))[0]
tokens.append((text, score))
return tokens
@ -82,7 +82,7 @@ def read_variables(fin):
shape = tuple(struct.unpack("i" * n_dims, fin.read(4 * n_dims)))
shape = shape[::-1]
name = fin.read(name_length).decode("utf-8")
name = fin.read(name_length).decode()
# ensure tensor data is aligned
tensor_data_offset = fin.tell()
@ -207,11 +207,11 @@ AI: Hello! How can I assist you today?
print(ctx.rstrip("\n"))
while True:
print("-" * 60)
prompt = input(f"User: ")
prompt = input("User: ")
if ctx != "":
ctx = ctx + "User: " + prompt + "\n"
ctx = f"{ctx}User: {prompt}\n"
else:
ctx = prompt + "\nAI:"
ctx = f"{prompt}\nAI:"
ctx = (ctx[-1920:]) if len(ctx) >= 2048 else ctx
@ -236,7 +236,7 @@ AI: Hello! How can I assist you today?
)
s = generation_output.sequences[0]
decoded = tokenizer.decode(s)
ctx = decoded + "\n"
ctx = f"{decoded}\n"
def main():

6
convert-gpt4all-to-ggml.py
View file

@ -49,7 +49,7 @@ def write_header(f_out, header):
def write_tokens(fout, tokenizer):
for i in range(tokenizer.vocab_size()):
if tokenizer.is_unknown(i):
text = " \u2047 ".encode("utf-8")
text = " \u2047 ".encode()
elif tokenizer.is_control(i):
text = b""
elif tokenizer.is_byte(i):
@ -60,13 +60,13 @@ def write_tokens(fout, tokenizer):
byte_value = int(piece[3:-1], 16)
text = struct.pack("B", byte_value)
else:
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode("utf-8")
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode()
fout.write(struct.pack("i", len(text)))
fout.write(text)
fout.write(struct.pack("f", tokenizer.get_score(i)))
# TODO: GPT4All - add extra <pad> token
text = "<pad>".encode("utf-8")
text = "<pad>".encode()
fout.write(struct.pack("i", len(text)))
fout.write(text)
fout.write(struct.pack("f", 0.0))

14
convert-gptq-to-ggml.py
View file

@ -50,7 +50,7 @@ fout.write(struct.pack("i", 4))
# This loop unchanged from convert-pth-to-ggml.py:
for i in range(tokenizer.vocab_size()):
if tokenizer.is_unknown(i):
text = " \u2047 ".encode("utf-8")
text = " \u2047 ".encode()
elif tokenizer.is_control(i):
text = b""
elif tokenizer.is_byte(i):
@ -61,13 +61,13 @@ for i in range(tokenizer.vocab_size()):
byte_value = int(piece[3:-1], 16)
text = struct.pack("B", byte_value)
else:
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode("utf-8")
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode()
fout.write(struct.pack("i", len(text)))
fout.write(text)
fout.write(struct.pack("f", tokenizer.get_score(i)))
def write_header(shape, dst_name, ftype_cur):
sname = dst_name.encode('utf-8')
sname = dst_name.encode()
fout.write(struct.pack("iii", len(shape), len(sname), ftype_cur))
fout.write(struct.pack("i" * len(shape), *shape[::-1]))
fout.write(sname)
@ -80,7 +80,7 @@ def write_header(shape, dst_name, ftype_cur):
def convert_non_q4(src_name, dst_name):
v = model[src_name]
shape = v.shape
print("Processing non-Q4 variable: " + src_name + " with shape: ", shape, " and type: ", v.dtype)
print(f"Processing non-Q4 variable: {src_name} with shape: {shape} and type: {v.dtype}")
if len(shape) == 1:
print(" Converting to float32")
v = v.to(torch.float32)
@ -105,7 +105,7 @@ def convert_q4(src_name, dst_name, permute=False):
# Each int32 item is actually 8 int4 items packed together, and it's transposed.
shape = (qweight.shape[0], qweight.shape[1] * 8)
print("Processing Q4 variable: " + src_name + " with shape: ", shape)
print(f"Processing Q4 variable: {src_name} with shape: {shape}")
# The output format has the int4 weights in groups of 32 rather than 8.
# It looks like this:
@ -168,5 +168,5 @@ for i in range(n_layer):
fout.close()
print("Done. Output file: " + fname_out)
print("")
print(f"Done. Output file: {fname_out}")
print()

6
convert-pth-to-ggml.py
View file

@ -120,7 +120,7 @@ def write_header(fout, hparams, ftype):
def write_tokens(fout, tokenizer):
for i in range(tokenizer.vocab_size()):
if tokenizer.is_unknown(i):
text = " \u2047 ".encode("utf-8")
text = " \u2047 ".encode()
elif tokenizer.is_control(i):
text = b""
elif tokenizer.is_byte(i):
@ -131,7 +131,7 @@ def write_tokens(fout, tokenizer):
byte_value = int(piece[3:-1], 16)
text = struct.pack("B", byte_value)
else:
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode("utf-8")
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode()
fout.write(struct.pack("i", len(text)))
fout.write(text)
fout.write(struct.pack("f", tokenizer.get_score(i)))
@ -191,7 +191,7 @@ def process_and_write_variables(fout, model, ftype, part_id, n_parts):
fullshape = list(partshape)
if n_dims > 1:
fullshape[split_dim] *= n_parts
sname = name.encode('utf-8')
sname = name.encode()
fout.write(struct.pack("iii", n_dims, len(sname), ftype_cur))
for dim in reversed(fullshape):
fout.write(struct.pack("i", dim))

4
convert-unversioned-ggml-to-ggml.py
View file

@ -44,7 +44,7 @@ def write_header(f_out, header):
def write_tokens(fout, tokenizer):
for i in range(tokenizer.vocab_size()):
if tokenizer.is_unknown(i):
text = " \u2047 ".encode("utf-8")
text = " \u2047 ".encode()
elif tokenizer.is_control(i):
text = b""
elif tokenizer.is_byte(i):
@ -55,7 +55,7 @@ def write_tokens(fout, tokenizer):
byte_value = int(piece[3:-1], 16)
text = struct.pack("B", byte_value)
else:
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode("utf-8")
text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode()
fout.write(struct.pack("i", len(text)))
fout.write(text)
fout.write(struct.pack("f", tokenizer.get_score(i)))

10
migrate-ggml-2023-03-30-pr613.py
View file

@ -272,13 +272,11 @@ def main():
tokens = read_tokens(fin, hparams)
if hparams['magic'] == 0x67676a74: # ggjt
print("%s: input ggml has already been converted to 'ggjt' magic\n" %
(args.fin_path))
print(f"{args.fin_path}: input ggml has already been converted to 'ggjt' magic\n")
sys.exit(1)
if hparams['magic'] != 0x67676d66: # ggmf
print("%s: input ggml file doesn't have expected 'ggmf' magic: %#x\n" %
(args.fin_path, hparams['magic']))
print(f"{args.fin_path}: input ggml file doesn't have expected 'ggmf' magic: {hparams['magic']:#x}\n")
sys.exit(1)
hparams['magic'] = 0x67676a74 # ggjt
@ -286,7 +284,7 @@ def main():
# count number of multipart files by convention
n_parts = 1
while True:
if os.path.exists("%s.%d" % (args.fin_path, n_parts)):
if os.path.exists(f"{args.fin_path}.{n_parts}"):
n_parts += 1
else:
break
@ -302,7 +300,7 @@ def main():
print(f"Processing part {part_id+1} of {n_parts}\n")
fin_path = args.fin_path
if part_id > 0:
fin_path += ".%d" % (part_id)
fin_path += f".{part_id}"
with open(fin_path, "rb") as fin:
read_tokens(fin, read_hparams(fin))
copy_tensors(fin, fout, part_id, n_parts)