jamie/ucsinfer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Activity

Added more helpful prompts and help

Browse Source
This commit is contained in:
Jamie Hardt
2025-08-27 14:04:46 -07:00
parent 2d233bd1f2
commit 7b3930ece0
1 changed files with 20 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
ucsinfer/__main__.py
View File

@@ -77,13 +77,17 @@ def finetune():
@ucsinfer.command('evaluate')
@click.option('--offset', type=int, default=0)
@click.option('--limit', type=int, default=-1)
@click.option('--offset', type=int, default=0, metavar="<int>",
help='Skip this many records in the dataset before processing')
@click.option('--limit', type=int, default=-1, metavar="<int>",
help='Process this many records and then exit')
@click.option('--no-foley', 'no_foley', flag_value=True, default=False,
help="Ignore any data in the set with FOLYProp or FOLYFeet "
"category")
@click.option('--model', type=str,
default="paraphrase-multilingual-mpnet-base-v2")
@click.option('--model', type=str, metavar="<model-name>",
default="paraphrase-multilingual-mpnet-base-v2",
show_default=True,
help="Select the sentence_transformer model to use")
@click.argument('dataset', type=click.File('r', encoding='utf8'),
default='dataset.csv')
def evaluate(dataset, offset, limit, model, no_foley):
@@ -110,8 +114,17 @@ def evaluate(dataset, offset, limit, model, no_foley):
ctx = InferenceContext(m, model)
reader = csv.reader(dataset)
print(f"Evaluating model {model}...")
results = []
for i, row in enumerate(tqdm.tqdm(reader)):
if offset > 0:
print(f"Skipping {offset} records...")
if limit > 0:
print(f"Will only evaluate {limit} records...")
progress_bar = tqdm.tqdm(total=limit)
for i, row in enumerate(reader):
if i < offset:
continue
@@ -131,6 +144,8 @@ def evaluate(dataset, offset, limit, model, no_foley):
results.append({'catid': cat_id, 'result': "TOP_10"})
else:
results.append({'catid': cat_id, 'result': "MISS"})
progress_bar.update(1)
total = len(results)
total_top = len([x for x in results if x['result'] == 'TOP'])