|
| 1 | +# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. |
| 2 | +# |
| 3 | +# Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 | +# you may not use this file except in compliance with the License. |
| 5 | +# You may obtain a copy of the License at |
| 6 | +# |
| 7 | +# http://www.apache.org/licenses/LICENSE-2.0 |
| 8 | +# |
| 9 | +# Unless required by applicable law or agreed to in writing, software |
| 10 | +# distributed under the License is distributed on an "AS IS" BASIS, |
| 11 | +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 12 | +# See the License for the specific language governing permissions and |
| 13 | +# limitations under the License. |
| 14 | + |
| 15 | +import os |
| 16 | +from pathlib import Path |
| 17 | +import json |
| 18 | +import pytest |
| 19 | + |
| 20 | +from nemo_automodel.components.datasets.llm import retrieval_dataset_inline as rdi |
| 21 | + |
| 22 | +def load_jsonl_one_line(path): |
| 23 | + ans = [] |
| 24 | + with open(path, "r") as f: |
| 25 | + for line in f: |
| 26 | + ans.append(json.loads(line)) |
| 27 | + break # only load the first line |
| 28 | + return ans |
| 29 | + |
| 30 | +def _embedding_testdata_training_file() -> Path: |
| 31 | + return Path(os.environ["TEST_DATA_DIR"]) / "embedding_testdata" / "training.jsonl" |
| 32 | + |
| 33 | + |
| 34 | +def test_retrieval_dataset_inline_embedding_testdata_smoke(): |
| 35 | + data_file = _embedding_testdata_training_file() |
| 36 | + if not data_file.exists(): |
| 37 | + pytest.skip(f"Missing embedding test data file: {data_file}") |
| 38 | + |
| 39 | + ds = rdi.make_retrieval_dataset( |
| 40 | + data_dir_list=str(data_file), |
| 41 | + data_type="train", |
| 42 | + train_n_passages=2, # 1 positive + 1 negative |
| 43 | + do_shuffle=False, |
| 44 | + max_train_samples=1, |
| 45 | + ) |
| 46 | + |
| 47 | + assert len(ds) >= 1 |
| 48 | + |
| 49 | + ex = ds[0] |
| 50 | + assert isinstance(ex.get("question"), str) and ex["question"] |
| 51 | + assert isinstance(ex.get("doc_text"), list) and len(ex["doc_text"]) == 2 |
| 52 | + assert isinstance(ex["doc_text"][0], str) |
| 53 | + assert isinstance(ex.get("doc_image"), list) and len(ex["doc_image"]) == 2 |
| 54 | + assert isinstance(ex.get("query_instruction"), str) |
| 55 | + assert isinstance(ex.get("passage_instruction"), str) |
| 56 | + |
| 57 | + payload = load_jsonl_one_line(data_file) |
| 58 | + assert ex['doc_text'][0] == payload[0]['pos_doc'] |
| 59 | + assert ex['doc_text'][1] == payload[0]['neg_doc'][0] |
| 60 | + assert ex['question'] == payload[0]['query'] |
| 61 | + |
0 commit comments