Will Lin
1fe50c0092
The v2 top level had grown to ~27 dirs + 13 loose files — half of them
single-concept abstraction shells (memory/ 86 LOC, transport/ 222, parity/ 178,
extend/ 226) and half vendored fastvideo code sitting as peers to the actual v2
design. Regroup to mirror README section 3 "Planes & dependency order":
core/ enums+types, card, loop, program, parity, request, parallel
(the model-native contracts; no kernels)
runtime/ + folded-in substrate: cache, memory, transport, extend
(the import graph shows only runtime consumes them; compile/cudagraph
already lived here)
serving/ + deploy/ (products / fleet)
_vendor/ all copied fastvideo: models, layers, attention, configs, distributed,
platforms, api, hooks, logging_utils, third_party + fastvideo_args/
utils/logger/envs/forward_context — internal layout unchanged, still
mirrors upstream for diffing
28 dirs + 13 root files -> 8 dirs + 6 root files. Pure mechanical move: 947
absolute-import paths rewritten (v2.X -> v2.{core,runtime,serving,_vendor}.X),
boundary-anchored so platform/ (native dispatch) and platforms/ (vendored CUDA
detect) no longer collide and neither does hooks/ vs extend/. Deleted the empty
v2/loader/. README section 3 + 16 updated; stale test count corrected
(34 files/216 tests -> 22 files/143 tests).
Validated: `import v2` stays torch-free; `v2/run_tests.py` and `pytest v2/tests/`
-> 143 passed, 2 skipped on the numpy toy backend. (On a GPU box force the toy
backend with CUDA_VISIBLE_DEVICES="" or detect() picks cuda.) No external importer
changed — examples use the re-exported `from v2 import VideoGenerator`.
33 lines
1.1 KiB
Python
33 lines
1.1 KiB
Python
"""TaskType — declared, never inferred.
|
|
|
|
The task is part of the typed request and the program branches on it. Heuristics
|
|
may only suggest a default at the product boundary; the runtime never infers it.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
from enum import Enum
|
|
|
|
|
|
class TaskType(str, Enum):
|
|
# --- wired in phase 1 (Wan2.1 / LTX2) ---
|
|
T2V = "t2v" # text -> video
|
|
I2V = "i2v" # image -> video
|
|
TI2V = "ti2v" # text+image -> video
|
|
V2V = "v2v" # video -> video
|
|
T2I = "t2i" # text -> image (batchable diffusion)
|
|
|
|
# --- declared now for omni-readiness; implemented in phase 2 ---
|
|
T2A = "t2a" # text -> audio (diffusion)
|
|
T2VS = "t2vs" # text -> video+sound (joint denoise)
|
|
V2W = "v2w" # video -> world rollout (chunked, causal)
|
|
A2W = "a2w" # action -> world rollout
|
|
REASON = "reason" # AR text reasoning (Cosmos3 reasoner / omni thinker)
|
|
|
|
@property
|
|
def is_world_rollout(self) -> bool:
|
|
return self in (TaskType.V2W, TaskType.A2W)
|
|
|
|
@property
|
|
def is_autoregressive(self) -> bool:
|
|
return self in (TaskType.REASON, )
|