Repository navigation
docs: document tested Conda CPU installation - #231
Merged
Merged
Conversation
Refs #230. Keep installation and native-library diagnostics scoped to one environment. Signed-off-by: LauraGPT <zhifu.gzf@alibaba-inc.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Document a Conda alternative in the bilingual installation guide, based on the
actual isolated CPU installation reported in #230. Keep the same interpreter
for installation, dependency checks and startup. Explain how preconfigured
Linux images can inject foreign PyTorch shared libraries despite
pip checkpassing, and preserve other library paths needed by FFmpeg.
User impact
Users who prefer Conda no longer need to infer the environment commands from
the venv example or search issue comments. This changes documentation only;
requirements, model loading, clipping, provider routing and all runtime source
remain byte-identical. Refs #230; do not auto-close the reporter's issue.
Validation
documentation tree. Four Bash blocks pass
bash -n.points to 求一个conda虚拟环境的安装命令,不是很喜欢用uv #230 (comment).
runtime or requirements bytes changed: Conda 26.9.4, Python 3.12.16,
torch/torchaudio 2.10.0+cpu, all 132 original requirement RECORDs complete;
real native imports/resampling, CLI help, Gradio, and synthetic video
encoding/decoding/audio extraction with installed FunASR audio loading passed.
The environment and experiments were not redundantly rebuilt or rerun.
python -m pytest -qis not green: collection imports the legacyfunclip/test/imagemagick_test.pydemo, which calls MoviePy TextClip and failsbecause the configured ImageMagick executable is
unset. This unchanged demoalso references an external example video. The collection error and complete
traceback are retained; the passing focused checks are not a full-suite claim.
cached completion of only 14 missing packages, and foreign-library import
failure are preserved. The accepted child environment removes only the foreign
system
torch/libentry, not all ofLD_LIBRARY_PATH; no shared environmentor production service was changed.
paid provider call, public server startup, or Nano leaderboard work occurred.
Screenshots or clips
Not applicable to a documentation-only change. The previous synthetic CPU media
fixture verifies I/O only, not recognition-to-clipping quality.
Notes for reviewers
Review was performed inline by the author under the user's explicit autonomous
maintenance authorization, not presented as an independent GitHub approval.
No review or other protection setting is changed. Exact-head CI, a normal merge
and final-main CI/log acceptance remain required. No release/PyPI operation.
Human-facing prose has no new string-presence tests; real installation evidence,
shell syntax and the repository's existing relevant checks establish the scope.