expon
Experiment tool for deep learning (PyTorch).
features
-
Git check
expon could automatically check the git status of current working directory.
In each experiment, it first check the git status (raise expception if the working tree is not clean), and save the current git commit id for code restore and experiment reproduce.
-
Experiment save.
expon save the all the experment information including the defined metrics, config, experiment seed and perhaps metric visualization (like loss-epochs) in one place. The output supports markdown and html form.
demo
from expon import EXP, Params, Metric
# init. The save directory will be './EXP/run/demo/'
exp = EXP(workspace = 'run', exp_name='demo', exp_description='this is a demo')
# experiment params
params = expon.Params()
params.lamb = 1
params.learning_rate = 0.001
params.batch_size = 512
exp.set_params(params)
# experiment metrics
loss = expon.Metric('loss', 'epoch', draw=True)
acc = expon.Metric('acc', draw=True)
exp.add_metric(loss)
exp.add_metric(acc)
# expon will randomly set the random, numpy and torch seed in [0, 999].
exp.set_seed()
# assume 100 epochs
for i in range(0, 100):
loss.update(1-0.01*i)
acc.update(0.01*i)
# add addition information
exp.add_info({'final acc': 0.91})
# save the experiment to one place
exp.save(output_format='md', show_metric=False)
Metadata
Release files for expon 0.0.7
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 | Actions |
|---|---|---|---|---|
| expon-0.0.7.tar.gz | 9.7 kB | Without provenance | Details for expon-0.0.7.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| expon-0.0.7-py3-none-any.whl | Python 3 | none | any | Details for expon-0.0.7-py3-none-any.whl |
Total release size: 22.3 kB
Release files / expon-0.0.7.tar.gz
| Download URL | expon-0.0.7.tar.gz |
|---|---|
| Size | 9.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
233d80508d7fbf6eb0f45690127aea624a7b6c8b9ea59edf838ca915418b595c
|
|
BLAKE2b-256 checksum How to use checksums |
79ba2e7641badbbf4ae99f650eeec1dfd1180aa876dc42d0bd30d7ff9554736e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.21.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.3
|
Release files / expon-0.0.7-py3-none-any.whl
| Download URL | expon-0.0.7-py3-none-any.whl |
|---|---|
| Size | 12.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
d0be8b9485522f248301508e3203064202145c257bcd62f12c428c1939a0a11b
|
|
BLAKE2b-256 checksum How to use checksums |
74205a2b5384701fb0c4d38297a391706d01485652d1ec3b92050634f1951eac
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.21.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.3
|
Security for expon 0.0.7
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.