ultimate
Project description
# ultimate
A very simpe neural network implemention for python
## Installation
pip install ultimate
## Why Ultimate?
+ Super tiny and super easy
+ Support feature importance
+ Support missing values
+ Support am2/a2m2/am2l/a2m2l activation functions
+ Support hardmse/hardmax loss functions
## How To Use?
<pre>
from ultimate.mlp import MLP
mlp = MLP(
mi=0,
dtype='float64',
activation=[], # am2/a2m2/am2l/a2m2l
layer_size=[],
input_type='pointwise',
loss_type='mse', # mse/softmax/hardmse/hardmax
output_range=[0, 1],
output_shrink=0.001,
importance_mul=0.001,
leaky=-0.001,
dropout=0,
bias_rate=[0.005],
weight_rate=[],
regularization=1
)
mlp.train(
in_arr,
target_arr,
epoch_train=5,
epoch_decay=1,
iteration_log=100,
rate_init=0.06,
rate_decay=0.9,
importance_out=False,
loss_mul=0.001,
verbose=1,
shuffle=True
)
mlp.predict(
in_arr,
out_arr=None,
verbose=0,
iteration_log=100
)
</pre>
## Examples
+ [Feature Importance](https://www.kaggle.com/anycode/feature-importance-using-nn)
+ [Image Regression](https://www.kaggle.com/anycode/image-regression)
+ [Iris Classification](https://www.kaggle.com/anycode/image-regression)
+ [MNIST Recognition](https://www.kaggle.com/anycode/mnist-recognition)
A very simpe neural network implemention for python
## Installation
pip install ultimate
## Why Ultimate?
+ Super tiny and super easy
+ Support feature importance
+ Support missing values
+ Support am2/a2m2/am2l/a2m2l activation functions
+ Support hardmse/hardmax loss functions
## How To Use?
<pre>
from ultimate.mlp import MLP
mlp = MLP(
mi=0,
dtype='float64',
activation=[], # am2/a2m2/am2l/a2m2l
layer_size=[],
input_type='pointwise',
loss_type='mse', # mse/softmax/hardmse/hardmax
output_range=[0, 1],
output_shrink=0.001,
importance_mul=0.001,
leaky=-0.001,
dropout=0,
bias_rate=[0.005],
weight_rate=[],
regularization=1
)
mlp.train(
in_arr,
target_arr,
epoch_train=5,
epoch_decay=1,
iteration_log=100,
rate_init=0.06,
rate_decay=0.9,
importance_out=False,
loss_mul=0.001,
verbose=1,
shuffle=True
)
mlp.predict(
in_arr,
out_arr=None,
verbose=0,
iteration_log=100
)
</pre>
## Examples
+ [Feature Importance](https://www.kaggle.com/anycode/feature-importance-using-nn)
+ [Image Regression](https://www.kaggle.com/anycode/image-regression)
+ [Iris Classification](https://www.kaggle.com/anycode/image-regression)
+ [MNIST Recognition](https://www.kaggle.com/anycode/mnist-recognition)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
No source distribution files available for this release.See tutorial on generating distribution archives.
Built Distribution
Close
Hashes for ultimate-1.15.1-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 327b8dec0c15ab2de7df018730de1d4e2ee43169c483aa1651c7084e599f2ec7 |
|
MD5 | 977f45219d4de7b89b0ee5fb3243849b |
|
BLAKE2b-256 | 1bd7845918c5b42a02291042a4ae243080d27987bd298728f9dfaf2668242f1f |