Docker/mealie
Docker/mealie
1
0
Fork 0
mirror of https://github.com/mealie-recipes/mealie.git synced 2025-08-05 21:45:25 +02:00
Code Issues Releases Activity

fix: pre-commit autoupdate && pre-commit run --all-files (#3780)

Browse source
This commit is contained in:
Christian Clauss 2024-06-26 12:41:46 +02:00 • committed by GitHub
parent 5e0f8a4bf7
commit e2033b2d67
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
82 changed files with 94 additions and 98 deletions
Download patch file Download diff file Expand all files Collapse all files

2
docs/docs/assets/other/n8n/n8n-mealie-backup.json
View file

@ -383,4 +383,4 @@
},
"id": "whloxeXkdBWWi2Uj",
"tags": []
}
}

2
docs/docs/assets/svg/open-in-new.svg
View file

@ -1,3 +1,3 @@
<svg xmlns="http://www.w3.org/2000/svg" width="12" height="12" viewBox="0 0 24 24">
<path fill="rgb(229,131,37)" d="M14,3V5H17.59L7.76,14.83L9.17,16.24L19,6.41V10H21V3M19,19H5V5H12V3H5C3.89,3 3,3.9 3,5V19A2,2 0 0,0 5,21H19A2,2 0 0,0 21,19V12H19V19Z" />
</svg>
</svg>

Side by side Swipe Overlay

Before

Width:  |  Height:  |  Size: 259 B

After

Width:  |  Height:  |  Size: 260 B

Before After
Before After

9
docs/docs/contributors/guides/ingredient-parser.md
View file

@ -1,20 +1,19 @@
# Improving the Ingredient Parser
Mealie uses Conditional Random Fields (CRFs) for parsing and processing ingredients. The model used for ingredients is based off a data set of over 100,000 ingredients from a dataset compiled by the New York Times. I believe that the model used is sufficient enough to handle most of the ingredients, therefore, more data to train the model won't necessarily help improve the model.
Mealie uses Conditional Random Fields (CRFs) for parsing and processing ingredients. The model used for ingredients is based off a data set of over 100,000 ingredients from a dataset compiled by the New York Times. I believe that the model used is sufficient enough to handle most of the ingredients, therefore, more data to train the model won't necessarily help improve the model.
## Improving The CRF Parser
To improve results with the model, you'll likely need to focus on improving the tokenization and parsing of the original string to aid the model in determine what the ingredient is. Data science is not my forte, but I have done some tokenization to improve the model. You can find that code under `/mealie/services/parser_services/crfpp` along with some other utility functions to aid in the tokenization and processing of ingredient strings.
To improve results with the model, you'll likely need to focus on improving the tokenization and parsing of the original string to aid the model in determine what the ingredient is. Data science is not my forte, but I have done some tokenization to improve the model. You can find that code under `/mealie/services/parser_services/crfpp` along with some other utility functions to aid in the tokenization and processing of ingredient strings.
The best way to test on improving the parser is to register additional test cases in `/mealie/tests/unit_tests/test_crfpp_parser.py` and run the test after making changes to the tokenizer. Note that the test cases DO NOT run in the CI environment, therefore you will need to have CRF++ installed on your machine. If you're using a Mac the easiest way to do this is through brew.
When submitting a PR to improve the parser it is important to provide your test cases, the problem you were trying to solve, and the results of the changes you made. As the tests don't run in CI, not providing these details may delay your PR from being merged.
When submitting a PR to improve the parser it is important to provide your test cases, the problem you were trying to solve, and the results of the changes you made. As the tests don't run in CI, not providing these details may delay your PR from being merged.
## Alternative Parsers
Alternatively, you can register a new parser by fulfilling the `ABCIngredientParser` interface. Satisfying this single method interface allows us to register additional parsing strategies at runtime and gives the user several options when trying to parse a recipe.
Alternatively, you can register a new parser by fulfilling the `ABCIngredientParser` interface. Satisfying this single method interface allows us to register additional parsing strategies at runtime and gives the user several options when trying to parse a recipe.
## Links
- [Pretrained Model](https://github.com/mealie-recipes/mealie-nlp-model)
- [CRF++ (Forked)](https://github.com/hay-kot/crfpp)

6
docs/docs/contributors/non-coders.md
View file

@ -9,8 +9,8 @@ We love your input! We want to make contributing to this project as easy and tra
- Becoming a maintainer
- Help translate to a new language or improve current translations
[Remember to join the Discord and stay in touch with other developers working on the project](https://discord.gg/QuStdQGSGK)!
[Remember to join the Discord and stay in touch with other developers working on the project](https://discord.gg/QuStdQGSGK)!
Additionally, you can buy me a coffee and support the project. When I get financial support it helps me know that there's real interest in the project and that it's worth the time to keep developing.
Additionally, you can buy me a coffee and support the project. When I get financial support it helps me know that there's real interest in the project and that it's worth the time to keep developing.
<a href="https://www.buymeacoffee.com/haykot" target="_blank"><img src="https://cdn.buymeacoffee.com/buttons/v2/default-green.png" alt="Buy Me A Coffee" style="height: 60px !important;width: 217px !important;" ></a>
<a href="https://www.buymeacoffee.com/haykot" target="_blank"><img src="https://cdn.buymeacoffee.com/buttons/v2/default-green.png" alt="Buy Me A Coffee" style="height: 60px !important;width: 217px !important;" ></a>

1
docs/docs/documentation/community-guide/bulk-url-import.md
View file

@ -93,4 +93,3 @@ mealie_url="http://localhost:9000"
token = authentication(mail, password, mealie_url)
import_from_file(input_file, token, mealie_url)
```