Skip to content
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

Jump to bottom

offshoot gen_test_data() from very long generate_data() #15

Merged
merged 2 commits into from
Jun 24, 2024
Merged

offshoot gen_test_data() from very long generate_data() #15

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
883bb54
offshoot gen_test_data() from very long generate_data()
makelinux Jun 20, 2024
c0f7320
offshoot gen_train_data() from very long generate_data()
makelinux Jun 20, 2024
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
171 changes: 97 additions & 74 deletions src/instructlab/sdg/generate_data.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -362,53 +362,19 @@ def read_taxonomy(*args, **kwargs):
return instructlab.utils.read_taxonomy(*args, **kwargs)


def generate_data(
logger,
api_base,
tls_insecure,
model_family: str,
yaml_rules: Optional[str] = None,
output_dir: Optional[str] = None,
taxonomy: Optional[str] = None,
taxonomy_base: Optional[str] = None,
prompt_file_path: Optional[str] = None,
model_name: Optional[str] = None,
num_cpus: Optional[int] = None,
num_instructions_to_generate: Optional[int] = None,
num_prompt_instructions=2,
request_batch_size=5,
temperature=1.0,
top_p=1.0,
rouge_threshold: Optional[float] = None,
console_output=True,
api_key: Optional[str] = None,
chunk_word_count=None,
server_ctx_size=None,
tls_client_cert: Optional[str] = None,
tls_client_key: Optional[str] = None,
tls_client_passwd: Optional[str] = None,
):
seed_instruction_data = []
machine_seed_instruction_data = []
generate_start = time.time()
def unescape(s):
return bytes(s, "utf-8").decode("utf-8")

if not os.path.exists(output_dir):
os.mkdir(output_dir)

# check taxonomy first then seed_tasks_path
# throw an error if both not found
# pylint: disable=broad-exception-caught,raise-missing-from
if taxonomy and os.path.exists(taxonomy):
seed_instruction_data = read_taxonomy(
logger, taxonomy, taxonomy_base, yaml_rules
)
else:
raise SystemExit(f"Error: taxonomy ({taxonomy}) does not exist.")

prompt_template = check_prompt_file(
prompt_file_path, get_model_family(model_family, model_name)
)
max_seed_tokens = max_seed_example_tokens(server_ctx_size, len(prompt_template))
def _gen_test_data(
logger,
seed_instruction_data,
max_seed_tokens,
taxonomy,
chunk_word_count,
server_ctx_size,
output_file_test,
):
max_seed_chars = num_chars_from_tokens(max_seed_tokens)
for seed_example in seed_instruction_data:
if (
Expand All @@ -426,9 +392,6 @@ def generate_data(
if not seeds:
raise SystemExit("Nothing to generate. Exiting.")

def unescape(s):
return bytes(s, "utf-8").decode("utf-8")

test_data = []
for seed_example in seed_instruction_data:
user = seed_example["instruction"]
Expand Down Expand Up @@ -457,6 +420,80 @@ def unescape(s):
fg="red",
)
raise click.exceptions.Exit(1)
# utils.jdump(test_data, os.path.join(output_dir, output_file_test))
with open(output_file_test, "w", encoding="utf-8") as outfile:
for entry in test_data:
json.dump(entry, outfile, ensure_ascii=False)
outfile.write("\n")


def _gen_train_data(machine_instruction_data, output_file_train):
train_data = []
for synth_example in machine_instruction_data:
user = synth_example["instruction"]
if len(synth_example["input"]) > 0:
user += "\n" + synth_example["input"]
train_data.append(
{
"system": utils.get_sysprompt(),
"user": unescape(user),
"assistant": unescape(synth_example["output"]),
}
)
# utils.jdump(train_data, output_file_train)
with open(output_file_train, "w", encoding="utf-8") as outfile:
for entry in train_data:
json.dump(entry, outfile, ensure_ascii=False)
outfile.write("\n")


def generate_data(
logger,
api_base,
tls_insecure,
model_family: str,
yaml_rules: Optional[str] = None,
output_dir: Optional[str] = None,
taxonomy: Optional[str] = None,
taxonomy_base: Optional[str] = None,
prompt_file_path: Optional[str] = None,
model_name: Optional[str] = None,
num_cpus: Optional[int] = None,
num_instructions_to_generate: Optional[int] = None,
num_prompt_instructions=2,
request_batch_size=5,
temperature=1.0,
top_p=1.0,
rouge_threshold: Optional[float] = None,
console_output=True,
api_key: Optional[str] = None,
chunk_word_count=None,
server_ctx_size=None,
tls_client_cert: Optional[str] = None,
tls_client_key: Optional[str] = None,
tls_client_passwd: Optional[str] = None,
):
seed_instruction_data = []
machine_seed_instruction_data = []
generate_start = time.time()

if not os.path.exists(output_dir):
os.mkdir(output_dir)

# check taxonomy first then seed_tasks_path
# throw an error if both not found
# pylint: disable=broad-exception-caught,raise-missing-from
if taxonomy and os.path.exists(taxonomy):
seed_instruction_data = read_taxonomy(
logger, taxonomy, taxonomy_base, yaml_rules
)
else:
raise SystemExit(f"Error: taxonomy ({taxonomy}) does not exist.")

prompt_template = check_prompt_file(
prompt_file_path, get_model_family(model_family, model_name)
)
max_seed_tokens = max_seed_example_tokens(server_ctx_size, len(prompt_template))

name = Path(model_name).stem # Just in case it is a file path
date_suffix = datetime.now().replace(microsecond=0).isoformat().replace(":", "_")
Expand All @@ -466,6 +503,15 @@ def unescape(s):
output_file_discarded = os.path.join(
output_dir, f"discarded_{name}_{date_suffix}.log"
)
_gen_test_data(
logger,
seed_instruction_data,
max_seed_tokens,
taxonomy,
chunk_word_count,
server_ctx_size,
os.path.join(output_dir, output_file_test),
)
logger.debug(f"Generating to: {os.path.join(output_dir, output_file)}")

request_idx = 0
Expand Down Expand Up @@ -580,32 +626,9 @@ def unescape(s):
f"Generated {total} instructions(discarded {discarded}), rouged {total - keep}, kept {keep} instructions"
)
utils.jdump(machine_instruction_data, os.path.join(output_dir, output_file))
train_data = []
for synth_example in machine_instruction_data:
user = synth_example["instruction"]
if len(synth_example["input"]) > 0:
user += "\n" + synth_example["input"]
train_data.append(
{
"system": utils.get_sysprompt(),
"user": unescape(user),
"assistant": unescape(synth_example["output"]),
}
)
# utils.jdump(train_data, os.path.join(output_dir, output_file_train))
with open(
os.path.join(output_dir, output_file_train), "w", encoding="utf-8"
) as outfile:
for entry in train_data:
json.dump(entry, outfile, ensure_ascii=False)
outfile.write("\n")
# utils.jdump(test_data, os.path.join(output_dir, output_file_test))
with open(
os.path.join(output_dir, output_file_test), "w", encoding="utf-8"
) as outfile:
for entry in test_data:
json.dump(entry, outfile, ensure_ascii=False)
outfile.write("\n")
_gen_train_data(
machine_instruction_data, os.path.join(output_dir, output_file_train)
)

progress_bar.close()

Expand Down
Loading