streamft 0.2.4__tar.gz → 0.2.5__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (24) hide show
  1. {streamft-0.2.4 → streamft-0.2.5}/PKG-INFO +1 -1
  2. {streamft-0.2.4 → streamft-0.2.5}/pyproject.toml +1 -1
  3. {streamft-0.2.4 → streamft-0.2.5}/streamft/cli.py +13 -2
  4. {streamft-0.2.4 → streamft-0.2.5}/streamft.egg-info/PKG-INFO +1 -1
  5. {streamft-0.2.4 → streamft-0.2.5}/tests/test_streamft_smoke.py +39 -0
  6. {streamft-0.2.4 → streamft-0.2.5}/LICENSE +0 -0
  7. {streamft-0.2.4 → streamft-0.2.5}/README.md +0 -0
  8. {streamft-0.2.4 → streamft-0.2.5}/setup.cfg +0 -0
  9. {streamft-0.2.4 → streamft-0.2.5}/streamft/__init__.py +0 -0
  10. {streamft-0.2.4 → streamft-0.2.5}/streamft/__main__.py +0 -0
  11. {streamft-0.2.4 → streamft-0.2.5}/streamft/autotune.py +0 -0
  12. {streamft-0.2.4 → streamft-0.2.5}/streamft/data.py +0 -0
  13. {streamft-0.2.4 → streamft-0.2.5}/streamft/disk_quant_store.py +0 -0
  14. {streamft-0.2.4 → streamft-0.2.5}/streamft/doctor.py +0 -0
  15. {streamft-0.2.4 → streamft-0.2.5}/streamft/fit.py +0 -0
  16. {streamft-0.2.4 → streamft-0.2.5}/streamft/lora.py +0 -0
  17. {streamft-0.2.4 → streamft-0.2.5}/streamft/metrics.py +0 -0
  18. {streamft-0.2.4 → streamft-0.2.5}/streamft/stream_quant_lora.py +0 -0
  19. {streamft-0.2.4 → streamft-0.2.5}/streamft/stream_train.py +0 -0
  20. {streamft-0.2.4 → streamft-0.2.5}/streamft.egg-info/SOURCES.txt +0 -0
  21. {streamft-0.2.4 → streamft-0.2.5}/streamft.egg-info/dependency_links.txt +0 -0
  22. {streamft-0.2.4 → streamft-0.2.5}/streamft.egg-info/entry_points.txt +0 -0
  23. {streamft-0.2.4 → streamft-0.2.5}/streamft.egg-info/requires.txt +0 -0
  24. {streamft-0.2.4 → streamft-0.2.5}/streamft.egg-info/top_level.txt +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: streamft
3
- Version: 0.2.4
3
+ Version: 0.2.5
4
4
  Summary: Fine-tune models that do not fit: stream NF4 weights from disk, one transformer block at a time.
5
5
  Author-email: Gokhan Cava <byjameson@users.noreply.github.com>
6
6
  License: Apache-2.0
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
4
4
 
5
5
  [project]
6
6
  name = "streamft"
7
- version = "0.2.4"
7
+ version = "0.2.5"
8
8
  description = "Fine-tune models that do not fit: stream NF4 weights from disk, one transformer block at a time."
9
9
  readme = "README.md"
10
10
  requires-python = ">=3.10"
@@ -133,6 +133,17 @@ def _format_messages(row: dict) -> list[dict]:
133
133
  ]
134
134
 
135
135
 
136
+ def sft_schedule(rows, encode):
137
+ """Split rows into trainable ones and a count of dropped ones.
138
+
139
+ The caller loops `for i in range(steps)` over the returned rows, so `--steps N`
140
+ always means N optimizer updates regardless of how many rows the prompt mask
141
+ rejects. Kept separate from the training loop so it can be tested without a GPU.
142
+ """
143
+ valid = [r for r in rows if encode(r) is not None]
144
+ return valid, len(rows) - len(valid)
145
+
146
+
136
147
  def _encode_chat(tokenizer, row: dict, seq: int):
137
148
  """Chat-template encode with the prompt tokens masked out of the loss."""
138
149
  messages = _format_messages(row)
@@ -376,8 +387,8 @@ def main() -> int:
376
387
  print(f"[eval] before loss={ce0:.4f}", flush=True)
377
388
  # Filter once so `--steps N` means N optimizer updates, not "N attempts minus
378
389
  # whatever the mask rejected". A silently short run is a lie about the budget.
379
- valid_rows = [r for r in train_rows if _encode_chat(tok, r, args.seq) is not None]
380
- skipped_train = len(train_rows) - len(valid_rows)
390
+ valid_rows, skipped_train = sft_schedule(
391
+ train_rows, lambda r: _encode_chat(tok, r, args.seq))
381
392
  if not valid_rows:
382
393
  raise SystemExit("[sft] no training row has a valid prompt mask; refusing to run")
383
394
  if skipped_train:
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: streamft
3
- Version: 0.2.4
3
+ Version: 0.2.5
4
4
  Summary: Fine-tune models that do not fit: stream NF4 weights from disk, one transformer block at a time.
5
5
  Author-email: Gokhan Cava <byjameson@users.noreply.github.com>
6
6
  License: Apache-2.0
@@ -8,6 +8,7 @@ They pin the things that broke silently during development:
8
8
 
9
9
  import math
10
10
 
11
+ import pytest
11
12
  import torch
12
13
  import torch.nn as nn
13
14
  import torch.nn.functional as F
@@ -209,3 +210,41 @@ def test_encode_chat_drops_rows_whose_prompt_is_not_a_prefix():
209
210
  ids, labels = _encode_chat(_StubTokenizer([1, 2, 3, 4], [1, 2]), row, 32)
210
211
  assert ids.tolist() == [[1, 2, 3, 4]]
211
212
  assert labels.tolist() == [[-100, -100, 3, 4]]
213
+
214
+
215
+ # ------------------------------------------------ SFT step budget & empty eval
216
+
217
+ def test_sft_schedule_keeps_the_step_budget_with_mixed_rows():
218
+ from streamft.cli import sft_schedule
219
+
220
+ rows = list(range(10))
221
+
222
+ def encode(r):
223
+ return None if r % 3 == 0 else [r] # 4 of 10 are rejected
224
+
225
+ valid, dropped = sft_schedule(rows, encode)
226
+ assert dropped == 4 and len(valid) == 6
227
+
228
+ steps = 25
229
+ schedule = [valid[i % len(valid)] for i in range(steps)]
230
+ assert len(schedule) == steps # N updates even though 4 rows were dropped
231
+ assert all(v % 3 != 0 for v in schedule)
232
+
233
+
234
+ def test_masked_eval_raises_when_every_row_is_skipped(monkeypatch):
235
+ import types
236
+
237
+ from streamft import cli
238
+
239
+ class FakeTrainer:
240
+ def __init__(self):
241
+ self.model = types.SimpleNamespace(eval=lambda: None, train=lambda: None)
242
+ self.norm = nn.Identity()
243
+ self.lm_head = nn.Linear(4, 7, bias=False)
244
+
245
+ def _forward_cached(self, ids):
246
+ return torch.zeros(1, ids.shape[1], 4), None, None
247
+
248
+ monkeypatch.setattr(cli, "_encode_chat", lambda *a, **k: None)
249
+ with pytest.raises(RuntimeError, match="every row was skipped"):
250
+ cli._masked_eval(FakeTrainer(), [{"instruction": "x", "response": "y"}], object(), 8)
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes
File without changes