apunkt/doc-to-lora
1
0
Fork 0
mirror of https://github.com/SakanaAI/doc-to-lora.git synced 2026-07-23 17:01:04 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

include hotpot_qa + fix ui display

Browse source
This commit is contained in:
51616 2025-05-22 07:17:21 +00:00
parent 6c60f6bbaf
commit e3012c1d78
2 changed files with 6 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

1
src/ctx_to_lora/eval.py
View file

@ -93,6 +93,7 @@ CLOSED_QA_DATASETS = {
"longbench/hotpotqa",
"longbench/2wikimqa",
"longbench/musique",
"hotpot_qa",
"squad",
"triviaqa_retrieved",
"negative_nq",

11
webui/app.py
View file

@ -144,7 +144,7 @@ def get_generated_text_data(run_path: str) -> dict:
split = os.path.basename(filename).replace("_generated_text.jsonl", "")
try:
with open(filename) as f:
lines = f.readlines(100_000)
lines = f.readlines(10_000_000)
data = [json.loads(line) for line in lines]
generated_data[split] = data
except FileNotFoundError:
@ -171,10 +171,10 @@ def get_generated_text_data(run_path: str) -> dict:
for filename in subdir_files:
base_split = os.path.basename(filename).replace("_generated_text.jsonl", "")
# Create a unique split key including the subdirectory name
split = f"{subdir_name}_{base_split}"
split = f"{subdir_name}/{base_split}"
try:
with open(filename) as f:
lines = f.readlines(100_000)
lines = f.readlines(10_000_000)
data = [json.loads(line) for line in lines]
generated_data[split] = data
except FileNotFoundError:
@ -222,7 +222,7 @@ def get_base_model_generated_text(model_name, generated_data):
try:
with open(base_model_path) as f:
lines = f.readlines(100_000)
lines = f.readlines(10_000_000)
data = [json.loads(line) for line in lines]
base_model_data[split] = data
except FileNotFoundError:
@ -232,7 +232,7 @@ def get_base_model_generated_text(model_name, generated_data):
try:
with open(base_model_no_context_path) as f:
lines = f.readlines(100_000)
lines = f.readlines(10_000_000)
data = [json.loads(line) for line in lines]
base_model_no_context_data[split] = data
except FileNotFoundError:
@ -394,7 +394,6 @@ def visualize(run):
# Get base model outputs if we have the model name and modulated outputs
base_model_data = {}
base_model_no_context_data = {}
print(generated_data.keys())
if model_name and generated_data:
base_model_data, base_model_no_context_data = get_base_model_generated_text(
model_name, generated_data