Zeras
A package of deep-learning utilities.
Description
Common to tf and torch:
from Zeras.data_parallelism import DataParallelism, get_files_with_ext
from Zeras.data_batcher import DataBatcher
from Zeras.vocab import Vocab
from Zeras.model_settings_baseboard import ModelSettingsBaseboard
Specific to tf:
from Zeras.model_baseboard import ModelBaseboard
from Zeras.model_baseboard import initialize_from_ckpt
from Zeras.model_baseboard import remove_from_trainable_variables
from Zeras.optim import linear_warmup_and_exp_decayed_lr
from Zeras.optim import linear_warmup_and_polynomial_decayed_lr
from Zeras.optim import adam_optimizer
from Zeras.optim import adam_wd_optimizer
from Zeras.optim import AdamWeightDecayOptimizer
import Zeras.zoo_nn as zrs_nn
import Zeras.zoo_layers as zrs_layers
Installation
From PyPI distribution system:
pip install Zeras
Or from the code:
git clone https://github.com/Li-Ming-Fan/Zeras
cd Zeras
make clean install
Or just incorporate the directory Zeras/ to your project.
Usage
For usage of this package, please refer to the repositories:
https://github.com/Li-Ming-Fan/text_classification
https://github.com/Li-Ming-Fan/transformer-tensorflow
https://github.com/Li-Ming-Fan/pointer-generator-refactored
Acknowledgement
The name "Zeras" is a variant from Keras. We use "Z" because it makes the direcory Zeras/ always display in the last (or first) and do not intersect other directories.
1 maintainer
Release files for Zeras 0.4.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| Zeras-0.4.4.tar.gz | 22.4 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| Zeras-0.4.4-py3.6.egg | Legacy Egg format | - | - | Details |
Total release size: 79.4 kB
Release files / Zeras-0.4.4.tar.gz
| Download URL | Zeras-0.4.4.tar.gz |
|---|---|
| Size | 22.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
8c40c6e450469e35b5c8b7cf169d26622297bce29142dd13e7a8f6ffeae99d25
|
|
BLAKE2b-256 checksum How to use checksums |
7260737ab84a16bd49cad7ca1d0bff9d1f09172ae78e254e580863704fda5a1f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.0 pkginfo/1.4.2 requests/2.23.0 setuptools/45.2.0.post20200210 requests-toolbelt/0.9.1 tqdm/4.42.1 CPython/3.6.10
|
Release files / Zeras-0.4.4-py3.6.egg
| Download URL | Zeras-0.4.4-py3.6.egg |
|---|---|
| Size | 57.0 kB |
| Tags | Egg |
|
SHA-256 checksum How to use checksums |
162840e11c480e57c8bc8c29f93200db005d5b48636c75eec04752cd0122ea74
|
|
BLAKE2b-256 checksum How to use checksums |
692029378954091697914b4a050edb76bd481dca6a88ff988d8b79cf63e53a1b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.0 pkginfo/1.4.2 requests/2.23.0 setuptools/45.2.0.post20200210 requests-toolbelt/0.9.1 tqdm/4.42.1 CPython/3.6.10
|
