active 2023-08-15 → 2026-07-24 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 1075 days from 2023-08-15 to 2026-07-24. Pushes: 290 total, peak 36 in a day. Pull requests: 284 total, peak 10 in a day. Issues: 178 total, peak 7 in a day. Comments: 375 total, peak 23 in a day. Stars: 7 total, peak 1 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| khatchad | 908 | 218 | 202 | 244 |
| msridhar | 147 | 5 | 2 | 40 |
| dependabot[bot] | 86 | 10 | 68 | 8 |
| juliandolby | 78 | 43 | 2 | 22 |
| Copilot | 62 | 12 | 4 | 14 |
| codecov[bot] | 36 | 0 | 0 | 36 |
| tatianacv | 8 | 0 | 3 | 4 |
| XC-Xinze | 2 | 0 | 1 | 1 |
| Latte17 | 1 | 0 | 0 | 1 |
| sifislag | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#767khatchad2026-07-24 01:15Baseline the consumer-visible evidence populations for release-scale regression detection
- Issue#740khatchad2026-07-17 03:07Trampoline cache key collides call sites with the same total arity but different positional/keyword splits
- Issue#716khatchad2026-07-10 23:17Value-shape unions silently drop unresolvable points-to members, overclaiming a single shape
- Issue#712khatchad2026-07-10 18:31Resolve `build`'s `input_shape` subscripts from the call-site input tensor shape
- Issue comment#684khatchad2026-07-04 02:59Module globals bound through a wildcard re-export keep an empty points-to set
- Issue#683khatchad2026-07-04 01:37`strategy.run` types its function argument's parameters but materializes no invoke edge under an unresolved receiver
- Issue#689khatchad2026-07-03 16:14Receiver-keyed trampoline contexts grow whole-project analysis time ~8x on layer-heavy subjects
- Issue comment#661khatchad2026-07-02 19:59Callable-object dispatch edges not materialized for self.layer(x) / self(x) / self.container[i](x) (0.52.10)
- Issue#662khatchad2026-07-02 14:29Canonicalize the dual-name `tf.keras.Model` summary for class-shell opt-in
- Issue#662khatchad2026-07-02 14:29Canonicalize the dual-name `tf.keras.Model` summary for class-shell opt-in
- Issue#659khatchad2026-07-01 02:16Regression: a parameter's tensor type is lost across an interprocedural call in whole-project analysis
- Issue#659khatchad2026-07-01 02:16Regression: a parameter's tensor type is lost across an interprocedural call in whole-project analysis
- Issue#651khatchad2026-06-30 18:36An `np.array` argument reaching only `tf.train.*` proto builders types the parameter as a tensor
- Issue#617khatchad2026-06-27 21:32Model the remaining `Input` parameters: `tensor`, `ragged`, and `type_spec`
- Issue#582khatchad2026-06-17 03:07Recover the unsorted-segment output shape when `num_segments` is static
- Issue#549khatchad2026-05-24 01:04Audit dataset-method field-duplication in `tensorflow.xml` — comment cites 1-CFA but engine uses 2-CFA
- Issue#549khatchad2026-05-24 01:04Audit dataset-method field-duplication in `tensorflow.xml` — comment cites 1-CFA but engine uses 2-CFA
- Issue#549khatchad2026-05-24 01:04Audit dataset-method field-duplication in `tensorflow.xml` — comment cites 1-CFA but engine uses 2-CFA
- Issue comment#509khatchad2026-05-23 23:07Dedicated `Relu` / `Cast` generators don't propagate result classification through chained consumers
- Issue#547khatchad2026-05-23 02:57Register `Cast`/`Relu`/`ExpandDims`/`ClipByValue` in `TensorGenerator.createManualGenerator`
- Issue#547khatchad2026-05-23 02:44Register `Cast`/`Relu`/`ExpandDims`/`ClipByValue` in `TensorGenerator.createManualGenerator`
- Issue#547khatchad2026-05-23 02:44Register `Cast`/`Relu`/`ExpandDims`/`ClipByValue` in `TensorGenerator.createManualGenerator`
- Issue#546khatchad2026-05-22 20:17Specialize ragged-tensor generators to emit `NumericDim` for statically-known integer-literal args
- Issue comment#545khatchad2026-05-22 01:49Use a typed sentinel for dynamic/batch dimensions instead of raw `null` in `TensorType.dims`
- Issue comment#545khatchad2026-05-22 01:48Use a typed sentinel for dynamic/batch dimensions instead of raw `null` in `TensorType.dims`
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 7 stars here means stars gained during the window, not the repo's star count.