| Name | expyro JSON |
| Version |
0.0.7
JSON |
| download |
| home_page | None |
| Summary | A lightweight library to manage reproducible experiments. |
| upload_time | 2025-09-18 11:39:21 |
| maintainer | None |
| docs_url | None |
| author | Lukas Haverbeck |
| requires_python | >=3.12 |
| license | MIT License
Copyright (c) 2024 Lukas Haverbeck
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
|
| keywords |
experiment-tracking
reproducibility
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# expyro 🧪✨
A lightweight Python library to stop your experiments from being a hot mess. Because "it worked on my machine" is not a valid scientific publication.
`expyro` is your new lab assistant 🧑🔬 that automatically organizes your chaos: configurations, results, plots, and even that random log file you swear you'll look at later.
## Features at a Glance 👀
* **🗂️ Structured Experiment Tracking:** No more `final_final_v2_test.py` files. Each run gets its own fancy, timestamped folder. Look organized, even if you're not.
* **🎯 Type Safety:** Your config isn't just a bunch of random numbers. It's a *well-defined* bunch of random numbers! Thanks, type hints!
* **♻️ Reproducibility:** Relive the magic (or the horror) of any past run. Perfect for when your advisor asks "can we get the results from last Tuesday?".
* **📊 Artifact Generation:** Automatically save your beautiful plots and tables. Make your future thesis-writing self cry tears of joy.
* **💾 Data Capture:** Easily dump any other file (models, logs, a screenshot of your error) right into the experiment's folder.
## Installation 💻
Get the core package and become 10x more organized instantly:
```bash
pip install expyro
```
### Want More? We Got More! 🍟
Level up your experiment-fu with optional extras:
```bash
# For making pretty, pretty plots (matplotlib)
pip install "expyro[matplotlib]"
# For turning results into sweet, sweet tables (pandas)
pip install "expyro[pandas]"
# I want it ALL! 🤑
pip install "expyro[all]"
```
## Quickstart: From Chaos to Clarity in 60 Seconds ⏱️
### 1. Define Your Experiment 🧪
Decorate your experiment function. It's like putting a lab coat on it.
```python
from dataclasses import dataclass
from pathlib import Path
import expyro
# Step 1: Define your config. This is your recipe.
@dataclass
class TrainConfig:
learning_rate: float = 0.01 # The spice of life
batch_size: int = 32 # The bigger, the better (until it crashes)
epochs: int = 10 # The "are we there yet?" parameter
# Step 2: Declare your experiment. Give it a home ("runs/") and a name.
# Your experiment must take exactly one argument as a config.
# The input and output must be typed.
@expyro.experiment(root=Path("runs"), name="my_awesome_experiment")
def train_model(config: TrainConfig) -> dict[str, float]:
# Your brilliant (or "it should work") experiment logic goes here.
final_loss = 0.1 * config.learning_rate
final_accuracy = 0.9
# Return whatever you want to remember
return {"final_loss": final_loss, "final_accuracy": final_accuracy}
```
### 2. Run It! 🏃♂️
Call your experiment. Watch the magic happen.
```python
if __name__ == "__main__":
cfg = TrainConfig(learning_rate=0.01, batch_size=32, epochs=10)
run = train_model(cfg) # This saves everything! You're welcome.
print(f"Run completed! Data is chilling in: {run.path}")
```
### 3. Make It Fancy! 🎨
Automatically save plots and tables. Impress everyone.
```python
import matplotlib.pyplot as plt
import pandas as pd
# Artist function: Takes config & result, returns a masterpiece (figure) or even a nested string dict of masterpieces
def create_plot(config: TrainConfig, result: dict) -> plt.Figure:
fig, ax = plt.subplots()
ax.bar(["Loss", "Accuracy"], [result["final_loss"], result["final_accuracy"]])
ax.set_title("How Did We Do?")
return fig
# Analyst function: Takes config & result, returns a sweet, sweet table (or a nested string dict of tables)
def create_table(config: TrainConfig, result: dict) -> pd.DataFrame:
return pd.DataFrame([{"metric": k, "value": v} for k, v in result.items()])
# Stack decorators like a pro! The order is bottom-up.
@expyro.plot(create_plot, file_format="pdf") # Save a high-res PDF
@expyro.table(create_table) # Save a CSV table
@expyro.experiment(root=Path("runs"), name="fancy_experiment")
def train_and_analyze(config: TrainConfig) -> dict:
# ... your code ...
return {"final_loss": 0.1, "final_accuracy": 0.9}
```
### 4. Pre-Bake Configs 🍱
Got favorite settings you keep typing over and over? Stash them as defaults and
summon them later from the command line (see below).
```python
@expyro.defaults({
"config-1": TrainConfig(learning_rate=0.1, batch_size=32, epochs=5),
"config-2": TrainConfig(learning_rate=0.001, batch_size=64, epochs=20),
})
@expyro.experiment(root=Path("runs"), name="experiment_with_defaults")
def train_with_defaults(config: TrainConfig) -> dict:
# ... your code ...
return {"final_loss": 0.1}
```
#### Launch Defaults from the CLI (With Overrides!) 🕹️
Stop editing Python files just to try a new seed. Each stored default becomes its
own subcommand under `default`, so you can mix and match directly from the terminal:
```bash
# Use config-1 exactly as declared above
expyro experiment_with_defaults default config-1
# Tweak a couple of fields on the fly
expyro experiment_with_defaults default config-1 --learning-rate=0.001 --epochs=
```
### 5. Save ALL THE THINGS! 💾
Use `hook` to save anything else right into the experiment's folder.
```python
@expyro.experiment(root=Path("runs"), name="experiment_with_everything")
def train_with_logging(config: TrainConfig) -> dict:
# Save a log file
with expyro.hook("logs/training_log.txt", "w") as f:
f.write(f"Let's hope this LR {config.learning_rate} works...\n")
f.write("Epoch 1: Loss=0.5 😬\n")
f.write("Epoch 2: Loss=0.2 😊\n")
# Save a model file (pytorch example)
# with expyro.hook("best_model.pt", "wb") as f:
# torch.save(model.state_dict(), f)
return {"final_loss": 0.1}
```
### 6. Analyze Your Glory (or Mistakes) 🔍
Iterate over past runs like a data archaeologist.
```python
# Your experiment is now also a container for all its runs!
my_experiment = train_model # This is your decorated function
print("Behold, all my past runs:")
for run in my_experiment: # 🚀 Iterate over everything!
print(f"Run {run.path.name}: Config={run.config}, Result={run.result}")
# Load a specific run from its path
that_one_run = my_experiment["2024-05-27/12:30:45.123 abcdef00"]
print(f"Ah yes, the run where loss was: {that_one_run.result['final_loss']}")
```
## What's In The Box? 📦 (The Project Structure)
Here’s how `expyro` organizes your brilliance:
```
runs/
└── my_awesome_experiment/ # Your experiment name
└── 2024-05-27/ # The date (so you know when you did the work)
├── 12:30:45.123 abcdef00/ # Time & unique ID (so you can find it)
│ ├── config.pickle # 🗃️ Your configuration, pickled.
│ ├── result.pickle # 📊 Your results, also pickled.
│ ├── artifacts/
│ │ ├── plots/ # 🎨 Home for your beautiful graphs
│ │ │ └── create_plot.pdf
│ │ └── tables/ # 📋 Home for your elegant tables
│ │ └── create_table.csv
│ └── data/ # 💾 Your custom files live here (from `hook`)
│ └── logs/
│ └── training_log.txt
└── 14:22:10.456 1a2b3c4d/ # Another run! You've been busy!
├── config.pickle
└── result.pickle
```
## CLI Time Travel Machine ⏳💻
Prefer the command line life? `expyro` scans your project for decorated experiments and hands each one its own
subcommand. It's like giving every lab rat a keyboard. 🐀
```
# Run a fresh experiment
expyro my_awesome_experiment run --learning-rate 0.01 --batch-size 32
# Kick off a run using a pre-baked config
expyro my_awesome_experiment default config-1
# Reproduce an old run with the exact same config
expyro my_awesome_experiment reproduce "2024-05-27/12:30:45.123 abcdef00"
# Redo an artifact when you forgot to save that plot 🎨
expyro my_awesome_experiment redo plots "2024-05-27/12:30:45.123 abcdef00"
```
Why so many verbs? Because reproducibility is king 👑:
* **`run`** starts a brand-new adventure and saves everything.
* **`default`** grabs a config registered with `@expyro.defaults` and runs it - no flags needed.
* **`reproduce`** reruns an experiment with the original config, giving you a carbon-copy run for free.
* **`redo`** regenerates plots or tables for an existing run, so you can tweak your visuals without touching the
science.
All from the shell, all consistent, all reproducible. 🔁
For detailed information for your specific setup, run
```bash
expyro --help
```
from the root directory of your project.
## License 📄
MIT License. Go forth and experiment! Just maybe use this library first.
---
**Now go forth and reproduce!** 🚀
Raw data
{
"_id": null,
"home_page": null,
"name": "expyro",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.12",
"maintainer_email": null,
"keywords": "experiment-tracking, reproducibility",
"author": "Lukas Haverbeck",
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/8a/68/18db580a91426f160009c985fda8fddf7c1860d5e8557de2c8e0e49e9e6c/expyro-0.0.7.tar.gz",
"platform": null,
"description": "# expyro \ud83e\uddea\u2728\n\nA lightweight Python library to stop your experiments from being a hot mess. Because \"it worked on my machine\" is not a valid scientific publication.\n\n`expyro` is your new lab assistant \ud83e\uddd1\u200d\ud83d\udd2c that automatically organizes your chaos: configurations, results, plots, and even that random log file you swear you'll look at later.\n\n## Features at a Glance \ud83d\udc40\n\n* **\ud83d\uddc2\ufe0f Structured Experiment Tracking:** No more `final_final_v2_test.py` files. Each run gets its own fancy, timestamped folder. Look organized, even if you're not.\n* **\ud83c\udfaf Type Safety:** Your config isn't just a bunch of random numbers. It's a *well-defined* bunch of random numbers! Thanks, type hints!\n* **\u267b\ufe0f Reproducibility:** Relive the magic (or the horror) of any past run. Perfect for when your advisor asks \"can we get the results from last Tuesday?\".\n* **\ud83d\udcca Artifact Generation:** Automatically save your beautiful plots and tables. Make your future thesis-writing self cry tears of joy.\n* **\ud83d\udcbe Data Capture:** Easily dump any other file (models, logs, a screenshot of your error) right into the experiment's folder.\n\n## Installation \ud83d\udcbb\n\nGet the core package and become 10x more organized instantly:\n\n```bash\npip install expyro\n```\n\n### Want More? We Got More! \ud83c\udf5f\n\nLevel up your experiment-fu with optional extras:\n\n```bash\n# For making pretty, pretty plots (matplotlib)\npip install \"expyro[matplotlib]\"\n\n# For turning results into sweet, sweet tables (pandas)\npip install \"expyro[pandas]\"\n\n# I want it ALL! \ud83e\udd11\npip install \"expyro[all]\"\n```\n\n## Quickstart: From Chaos to Clarity in 60 Seconds \u23f1\ufe0f\n\n### 1. Define Your Experiment \ud83e\uddea\n\nDecorate your experiment function. It's like putting a lab coat on it.\n\n```python\nfrom dataclasses import dataclass\nfrom pathlib import Path\nimport expyro\n\n# Step 1: Define your config. This is your recipe.\n@dataclass\nclass TrainConfig:\n learning_rate: float = 0.01 # The spice of life\n batch_size: int = 32 # The bigger, the better (until it crashes)\n epochs: int = 10 # The \"are we there yet?\" parameter\n\n# Step 2: Declare your experiment. Give it a home (\"runs/\") and a name.\n# Your experiment must take exactly one argument as a config.\n# The input and output must be typed. \n@expyro.experiment(root=Path(\"runs\"), name=\"my_awesome_experiment\")\ndef train_model(config: TrainConfig) -> dict[str, float]:\n # Your brilliant (or \"it should work\") experiment logic goes here.\n final_loss = 0.1 * config.learning_rate\n final_accuracy = 0.9\n\n # Return whatever you want to remember\n return {\"final_loss\": final_loss, \"final_accuracy\": final_accuracy}\n```\n\n### 2. Run It! \ud83c\udfc3\u200d\u2642\ufe0f\n\nCall your experiment. Watch the magic happen.\n\n```python\nif __name__ == \"__main__\":\n cfg = TrainConfig(learning_rate=0.01, batch_size=32, epochs=10)\n run = train_model(cfg) # This saves everything! You're welcome.\n print(f\"Run completed! Data is chilling in: {run.path}\")\n```\n\n### 3. Make It Fancy! \ud83c\udfa8\n\nAutomatically save plots and tables. Impress everyone.\n\n```python\nimport matplotlib.pyplot as plt\nimport pandas as pd\n\n# Artist function: Takes config & result, returns a masterpiece (figure) or even a nested string dict of masterpieces\ndef create_plot(config: TrainConfig, result: dict) -> plt.Figure:\n fig, ax = plt.subplots()\n ax.bar([\"Loss\", \"Accuracy\"], [result[\"final_loss\"], result[\"final_accuracy\"]])\n ax.set_title(\"How Did We Do?\")\n return fig\n\n# Analyst function: Takes config & result, returns a sweet, sweet table (or a nested string dict of tables)\ndef create_table(config: TrainConfig, result: dict) -> pd.DataFrame:\n return pd.DataFrame([{\"metric\": k, \"value\": v} for k, v in result.items()])\n\n# Stack decorators like a pro! The order is bottom-up.\n@expyro.plot(create_plot, file_format=\"pdf\") # Save a high-res PDF\n@expyro.table(create_table) # Save a CSV table\n@expyro.experiment(root=Path(\"runs\"), name=\"fancy_experiment\")\ndef train_and_analyze(config: TrainConfig) -> dict:\n # ... your code ...\n return {\"final_loss\": 0.1, \"final_accuracy\": 0.9}\n```\n\n### 4. Pre-Bake Configs \ud83c\udf71\n\nGot favorite settings you keep typing over and over? Stash them as defaults and\nsummon them later from the command line (see below).\n\n```python\n@expyro.defaults({\n \"config-1\": TrainConfig(learning_rate=0.1, batch_size=32, epochs=5),\n \"config-2\": TrainConfig(learning_rate=0.001, batch_size=64, epochs=20),\n})\n@expyro.experiment(root=Path(\"runs\"), name=\"experiment_with_defaults\")\ndef train_with_defaults(config: TrainConfig) -> dict:\n # ... your code ...\n return {\"final_loss\": 0.1}\n```\n\n#### Launch Defaults from the CLI (With Overrides!) \ud83d\udd79\ufe0f\n\nStop editing Python files just to try a new seed. Each stored default becomes its\nown subcommand under `default`, so you can mix and match directly from the terminal:\n\n```bash\n# Use config-1 exactly as declared above\nexpyro experiment_with_defaults default config-1\n\n# Tweak a couple of fields on the fly\nexpyro experiment_with_defaults default config-1 --learning-rate=0.001 --epochs=\n```\n\n### 5. Save ALL THE THINGS! \ud83d\udcbe\n\nUse `hook` to save anything else right into the experiment's folder.\n\n```python\n@expyro.experiment(root=Path(\"runs\"), name=\"experiment_with_everything\")\ndef train_with_logging(config: TrainConfig) -> dict:\n # Save a log file\n with expyro.hook(\"logs/training_log.txt\", \"w\") as f:\n f.write(f\"Let's hope this LR {config.learning_rate} works...\\n\")\n f.write(\"Epoch 1: Loss=0.5 \ud83d\ude2c\\n\")\n f.write(\"Epoch 2: Loss=0.2 \ud83d\ude0a\\n\")\n\n # Save a model file (pytorch example)\n # with expyro.hook(\"best_model.pt\", \"wb\") as f:\n # torch.save(model.state_dict(), f)\n\n return {\"final_loss\": 0.1}\n```\n\n### 6. Analyze Your Glory (or Mistakes) \ud83d\udd0d\n\nIterate over past runs like a data archaeologist.\n\n```python\n# Your experiment is now also a container for all its runs!\nmy_experiment = train_model # This is your decorated function\n\nprint(\"Behold, all my past runs:\")\nfor run in my_experiment: # \ud83d\ude80 Iterate over everything!\n print(f\"Run {run.path.name}: Config={run.config}, Result={run.result}\")\n\n# Load a specific run from its path\nthat_one_run = my_experiment[\"2024-05-27/12:30:45.123 abcdef00\"]\nprint(f\"Ah yes, the run where loss was: {that_one_run.result['final_loss']}\")\n```\n\n## What's In The Box? \ud83d\udce6 (The Project Structure)\n\nHere\u2019s how `expyro` organizes your brilliance:\n\n```\nruns/\n\u2514\u2500\u2500 my_awesome_experiment/ # Your experiment name\n \u2514\u2500\u2500 2024-05-27/ # The date (so you know when you did the work)\n \u251c\u2500\u2500 12:30:45.123 abcdef00/ # Time & unique ID (so you can find it)\n \u2502 \u251c\u2500\u2500 config.pickle # \ud83d\uddc3\ufe0f Your configuration, pickled.\n \u2502 \u251c\u2500\u2500 result.pickle # \ud83d\udcca Your results, also pickled.\n \u2502 \u251c\u2500\u2500 artifacts/\n \u2502 \u2502 \u251c\u2500\u2500 plots/ # \ud83c\udfa8 Home for your beautiful graphs\n \u2502 \u2502 \u2502 \u2514\u2500\u2500 create_plot.pdf\n \u2502 \u2502 \u2514\u2500\u2500 tables/ # \ud83d\udccb Home for your elegant tables\n \u2502 \u2502 \u2514\u2500\u2500 create_table.csv\n \u2502 \u2514\u2500\u2500 data/ # \ud83d\udcbe Your custom files live here (from `hook`)\n \u2502 \u2514\u2500\u2500 logs/\n \u2502 \u2514\u2500\u2500 training_log.txt\n \u2514\u2500\u2500 14:22:10.456 1a2b3c4d/ # Another run! You've been busy!\n \u251c\u2500\u2500 config.pickle\n \u2514\u2500\u2500 result.pickle\n```\n\n## CLI Time Travel Machine \u23f3\ud83d\udcbb\n\nPrefer the command line life? `expyro` scans your project for decorated experiments and hands each one its own\nsubcommand. It's like giving every lab rat a keyboard. \ud83d\udc00\n\n```\n# Run a fresh experiment\nexpyro my_awesome_experiment run --learning-rate 0.01 --batch-size 32\n\n# Kick off a run using a pre-baked config\nexpyro my_awesome_experiment default config-1\n\n# Reproduce an old run with the exact same config\nexpyro my_awesome_experiment reproduce \"2024-05-27/12:30:45.123 abcdef00\"\n\n# Redo an artifact when you forgot to save that plot \ud83c\udfa8\nexpyro my_awesome_experiment redo plots \"2024-05-27/12:30:45.123 abcdef00\"\n```\n\nWhy so many verbs? Because reproducibility is king \ud83d\udc51:\n\n* **`run`** starts a brand-new adventure and saves everything.\n* **`default`** grabs a config registered with `@expyro.defaults` and runs it - no flags needed.\n* **`reproduce`** reruns an experiment with the original config, giving you a carbon-copy run for free.\n* **`redo`** regenerates plots or tables for an existing run, so you can tweak your visuals without touching the \nscience.\n\nAll from the shell, all consistent, all reproducible. \ud83d\udd01\n\nFor detailed information for your specific setup, run\n\n```bash\nexpyro --help\n```\n\nfrom the root directory of your project.\n\n## License \ud83d\udcc4\n\nMIT License. Go forth and experiment! Just maybe use this library first.\n\n---\n\n**Now go forth and reproduce!** \ud83d\ude80\n\n",
"bugtrack_url": null,
"license": "MIT License\n \n Copyright (c) 2024 Lukas Haverbeck\n \n Permission is hereby granted, free of charge, to any person obtaining a copy\n of this software and associated documentation files (the \"Software\"), to deal\n in the Software without restriction, including without limitation the rights\n to use, copy, modify, merge, publish, distribute, sublicense, and/or sell\n copies of the Software, and to permit persons to whom the Software is\n furnished to do so, subject to the following conditions:\n \n The above copyright notice and this permission notice shall be included in all\n copies or substantial portions of the Software.\n \n THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\n IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\n FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\n AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\n LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\n OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\n SOFTWARE.\n ",
"summary": "A lightweight library to manage reproducible experiments.",
"version": "0.0.7",
"project_urls": {
"Homepage": "https://github.com/lukashaverbeck/expyro",
"Issues": "https://github.com/lukashaverbeck/expyro/issues"
},
"split_keywords": [
"experiment-tracking",
" reproducibility"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "e01cc463a2e0f9ecc60b62c27735362172c3d982879a24c0f041799da0c4e5d3",
"md5": "04d6132413ffbe0a6b8bebfa4a01ed35",
"sha256": "61ee103d610421c48904f1513a3a09794c9be8ab76038dfe0ad1137f17481b8c"
},
"downloads": -1,
"filename": "expyro-0.0.7-py3-none-any.whl",
"has_sig": false,
"md5_digest": "04d6132413ffbe0a6b8bebfa4a01ed35",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.12",
"size": 14951,
"upload_time": "2025-09-18T11:39:20",
"upload_time_iso_8601": "2025-09-18T11:39:20.086940Z",
"url": "https://files.pythonhosted.org/packages/e0/1c/c463a2e0f9ecc60b62c27735362172c3d982879a24c0f041799da0c4e5d3/expyro-0.0.7-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "8a6818db580a91426f160009c985fda8fddf7c1860d5e8557de2c8e0e49e9e6c",
"md5": "309194175f14c41d35c11e4c739d47f7",
"sha256": "58f29ff529d421bd05c46efd01d3b74b4c64448953501871a39a67b64b748738"
},
"downloads": -1,
"filename": "expyro-0.0.7.tar.gz",
"has_sig": false,
"md5_digest": "309194175f14c41d35c11e4c739d47f7",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.12",
"size": 18065,
"upload_time": "2025-09-18T11:39:21",
"upload_time_iso_8601": "2025-09-18T11:39:21.481353Z",
"url": "https://files.pythonhosted.org/packages/8a/68/18db580a91426f160009c985fda8fddf7c1860d5e8557de2c8e0e49e9e6c/expyro-0.0.7.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-09-18 11:39:21",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "lukashaverbeck",
"github_project": "expyro",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "expyro"
}
