Skip to content

Commit 07fa58b

Browse files
qgallouedecLuisVasquezBSC
authored andcommitted
⚰️ Remove deprecated (huggingface#3704)
1 parent 61e07aa commit 07fa58b

11 files changed

Lines changed: 204 additions & 1210 deletions

tests/test_data_collator_completion_only.py

Lines changed: 0 additions & 169 deletions
This file was deleted.

tests/test_data_utils.py

Lines changed: 0 additions & 43 deletions
Original file line numberDiff line numberDiff line change
@@ -29,7 +29,6 @@
2929
maybe_extract_prompt,
3030
maybe_unpair_preference_dataset,
3131
pack_dataset,
32-
pack_examples,
3332
truncate_dataset,
3433
unpair_preference_dataset,
3534
)
@@ -424,48 +423,6 @@ def test_maybe_extract_prompt_standard_already_explicit(self):
424423
)
425424

426425

427-
class TestPackExamples(unittest.TestCase):
428-
def test_larger_chunks(self):
429-
examples = {
430-
"input_ids": [[1, 2, 3], [4, 5, 6, 7], [8]],
431-
"attention_mask": [[0, 1, 1], [0, 0, 1, 1], [1]],
432-
}
433-
seq_length = 5
434-
expected_output = {
435-
"input_ids": [[1, 2, 3, 4, 5], [6, 7, 8]],
436-
"attention_mask": [[0, 1, 1, 0, 0], [1, 1, 1]],
437-
}
438-
result = pack_examples(examples, seq_length)
439-
self.assertEqual(result, expected_output)
440-
441-
def test_smaller_chunks(self):
442-
examples = {
443-
"input_ids": [[1, 2, 3], [4, 5, 6, 7], [8]],
444-
"attention_mask": [[0, 1, 1], [0, 0, 1, 1], [1]],
445-
}
446-
seq_length = 2
447-
expected_output = {
448-
"input_ids": [[1, 2], [3, 4], [5, 6], [7, 8]],
449-
"attention_mask": [[0, 1], [1, 0], [0, 1], [1, 1]],
450-
}
451-
result = pack_examples(examples, seq_length)
452-
self.assertEqual(result, expected_output)
453-
454-
def test_with_dataset(self):
455-
examples = {
456-
"input_ids": [[1, 2, 3], [4, 5, 6, 7], [8]],
457-
"attention_mask": [[0, 1, 1], [0, 0, 1, 1], [1]],
458-
}
459-
dataset = Dataset.from_dict(examples)
460-
seq_length = 3
461-
expected_output = {
462-
"input_ids": [[1, 2, 3], [4, 5, 6], [7, 8]],
463-
"attention_mask": [[0, 1, 1], [0, 0, 1], [1, 1]],
464-
}
465-
dataset = dataset.map(pack_examples, batched=True, fn_kwargs={"seq_length": seq_length})
466-
self.assertEqual(dataset.to_dict(), expected_output)
467-
468-
469426
class TestPackDatasetWrapped(unittest.TestCase):
470427
def test_with_dataset(self):
471428
examples = {

0 commit comments

Comments
 (0)