tuneconfig 
Hyperparameter config file generator and experiment runner.
Quickstart
$ pip install -U tuneconfig
Usage
Config file generator
import pprint
import tuneconfig
# Define parameter formatting function
def format_fn(param):
fmt = {
"batch_size": "batch",
"horizon": "hr",
"learning_rate": "lr",
"optimizer": "opt",
"epochs": None,
"num_samples": None,
}
return fmt.get(param, param)
# Define a configuration template for grid search
config_iterator = tuneconfig.ConfigFactory({
"batch_size": tuneconfig.grid_search([32, 128]),
"horizon": 40,
"learning_rate": tuneconfig.grid_search([0.01, 0.1]),
"epochs": 1000,
"optimizer": tuneconfig.grid_search(["Adam", "RMSProp"]),
"num_samples": 10
},
format_fn=format_fn
)
# Iterate over config dicts
for idx, config in enumerate(config_iterator):
name = config_iterator._trial_id(config)
print(f"config {idx} ({name}):")
pprint.pprint(config)
print()
# Dump config dicts as JSON files
tmp = "/tmp/tuneconfig"
json_config_files = config_iterator.dump(tmp)
print(">> Saved config files:")
pprint.pprint(json_config_files)
config 0 (batch=32/hr=40/lr=0.01/opt=Adam):
{'batch_size': 32,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.01,
'num_samples': 10,
'optimizer': 'Adam'}
config 1 (batch=32/hr=40/lr=0.01/opt=RMSProp):
{'batch_size': 32,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.01,
'num_samples': 10,
'optimizer': 'RMSProp'}
config 2 (batch=32/hr=40/lr=0.1/opt=Adam):
{'batch_size': 32,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.1,
'num_samples': 10,
'optimizer': 'Adam'}
config 3 (batch=32/hr=40/lr=0.1/opt=RMSProp):
{'batch_size': 32,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.1,
'num_samples': 10,
'optimizer': 'RMSProp'}
config 4 (batch=128/hr=40/lr=0.01/opt=Adam):
{'batch_size': 128,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.01,
'num_samples': 10,
'optimizer': 'Adam'}
config 5 (batch=128/hr=40/lr=0.01/opt=RMSProp):
{'batch_size': 128,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.01,
'num_samples': 10,
'optimizer': 'RMSProp'}
config 6 (batch=128/hr=40/lr=0.1/opt=Adam):
{'batch_size': 128,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.1,
'num_samples': 10,
'optimizer': 'Adam'}
config 7 (batch=128/hr=40/lr=0.1/opt=RMSProp):
{'batch_size': 128,
'epochs': 1000,
'horizon': 40,
'learning_rate': 0.1,
'num_samples': 10,
'optimizer': 'RMSProp'}
>> Saved config files:
['/tmp/tuneconfig/batch=32/hr=40/lr=0.01/opt=Adam/config.json',
'/tmp/tuneconfig/batch=32/hr=40/lr=0.01/opt=RMSProp/config.json',
'/tmp/tuneconfig/batch=32/hr=40/lr=0.1/opt=Adam/config.json',
'/tmp/tuneconfig/batch=32/hr=40/lr=0.1/opt=RMSProp/config.json',
'/tmp/tuneconfig/batch=128/hr=40/lr=0.01/opt=Adam/config.json',
'/tmp/tuneconfig/batch=128/hr=40/lr=0.01/opt=RMSProp/config.json',
'/tmp/tuneconfig/batch=128/hr=40/lr=0.1/opt=Adam/config.json',
'/tmp/tuneconfig/batch=128/hr=40/lr=0.1/opt=RMSProp/config.json']
License
Copyright (c) 2020 Thiago Pereira Bueno All Rights Reserved.
tuneconfig is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
tuneconfig is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with tuneconfig. If not, see http://www.gnu.org/licenses/.
1 maintainer
Metadata
1 maintainer
Release files for tuneconfig 0.14.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| tuneconfig-0.14.1.tar.gz | 15.1 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| tuneconfig-0.14.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 48.6 kB
Release files / tuneconfig-0.14.1.tar.gz
| Download URL | tuneconfig-0.14.1.tar.gz |
|---|---|
| Size | 15.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
4dfa98ea021718d87e49d26fc103785cb0e1745f855203a872ff7cd938562405
|
|
BLAKE2b-256 checksum How to use checksums |
d5c0eee2728ff105cccb56e04a50d0b4e0c42527c351964506d0ffe55946d8af
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/46.1.3 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.6.9
|
Release files / tuneconfig-0.14.1-py3-none-any.whl
| Download URL | tuneconfig-0.14.1-py3-none-any.whl |
|---|---|
| Size | 33.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
e7977401be53038b3952331e69fa47da6897b035daeed095853474d35455e04c
|
|
BLAKE2b-256 checksum How to use checksums |
c819296c2b6c83c8c48bf115d18446894703745e10b5cff163975c7664c93136
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/46.1.3 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.6.9
|
Security for tuneconfig 0.14.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
