vault backup: 2025-01-16 10:34:48
This commit is contained in:
parent
d7043e854a
commit
7c66c5a0c5
|
@ -13,12 +13,12 @@
|
||||||
"state": {
|
"state": {
|
||||||
"type": "markdown",
|
"type": "markdown",
|
||||||
"state": {
|
"state": {
|
||||||
"file": "組織/EverfortuneAI/{Note} Packaging Python Project.md",
|
"file": "組織/EverfortuneAI/{Note} Nox.md",
|
||||||
"mode": "source",
|
"mode": "source",
|
||||||
"source": false
|
"source": false
|
||||||
},
|
},
|
||||||
"icon": "lucide-file",
|
"icon": "lucide-file",
|
||||||
"title": "{Note} Packaging Python Project"
|
"title": "{Note} Nox"
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
]
|
]
|
||||||
|
@ -144,9 +144,10 @@
|
||||||
"obsidian-git:Open Git source control": false
|
"obsidian-git:Open Git source control": false
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"active": "e0251782d53c0c60",
|
"active": "6dd46505ac5f0579",
|
||||||
"lastOpenFiles": [
|
"lastOpenFiles": [
|
||||||
"組織/EverfortuneAI/{Note} Nox.md",
|
"組織/EverfortuneAI/{Note} Nox.md",
|
||||||
|
"Untitled.md",
|
||||||
"組織/EverfortuneAI/{Note} Packaging Python Project.md",
|
"組織/EverfortuneAI/{Note} Packaging Python Project.md",
|
||||||
"README.md",
|
"README.md",
|
||||||
"Test.md",
|
"Test.md",
|
||||||
|
|
|
@ -46,4 +46,8 @@ def test_flexibility(session, numpy, scikit_learn):
|
||||||
# Test
|
# Test
|
||||||
session.install('pytest')
|
session.install('pytest')
|
||||||
session.run('pytest', 'tests', '-v')
|
session.run('pytest', 'tests', '-v')
|
||||||
```
|
```
|
||||||
|
- Issues
|
||||||
|
1. 在某些依賴套件組合下,load_model在進行reload測試時會報錯,原因是舊模型與新模型的記憶體位置相同導致AssertionError,但為何會有這樣的差異仍未知
|
||||||
|
2. 將nox的venv_backend設為conda,並開啟reuse_venv進行測試,測試一陣子後,會開始出現conda error,猜測原因是reuse的機制出問題,如果是新創造的conda env就沒問題(沒開啟reuse會很耗時,每次都得重建測試環境,目前一個環境的容量約5GB,安裝得花不少時間)
|
||||||
|
3. 呈上,若把venv_backend改成預設的venv,在測試一段時間後仍會報錯,原因可能跟上述相同
|
Loading…
Reference in New Issue