Spaces:
Build error
Build error
Commit History
-
Update requirements.txt
Verified signature599dfe7
-
changing the example to what is actually in the example
9c1370a
-
Merge pull request #2 from Jensen-holm/dev
Unverified signaturecleaning up exampled6cc61a -
making the example code cleaner
Jensen Holm498c4e0 -
loss functions now just have staticmethods
Jensen Holmd7ea050 -
adding unit tests for activation functions (using pytest)
Jensen Holmff1254a
-
adding example documentation
Jensen Holm9249567 -
format
Jensen Holm05819f2 -
adding example program 'example.py'
Jensen Holm284dbf2
-
found the bug in setup.py that was stopping me from getting the package
to work. Now it works!adc3f91
-
readme updates
2cac6be -
pypi package works!
a094ad4 -
fixed bug where I forgot to change the name from nn to numpyneuron
4dc8296 -
stuck on getting this package to work on pypi
f7ea12e -
python package not really working, but it is all set up okay I think
e53a40c -
better about and readme docs
7855334 -
Merge pull request #1 from Jensen-holm/package
Unverified signaturemaking it a python package on PyPI84ec87c -
making it a python package on PyPI
75481dd -
Moving towards making this a python package for simple neural network
creation. This means making it an option to make this a gradio app or not, that logic is handled in __post_init__ and it not expensive. Also I have implemented a forward_fn thing so we can handle non logits loss functions if that makes sense03f2b37
-
getting rid of regression tab, thinking about having a custom data tab
where users can upload their own dataset and hyper parameter tune on that. and making the classification tab the Classification example tab.9117cbd -
fixing bug that happened with a type that I made, I think huggingface
operates on an older version of python??de85694 -
features added:
- batch size argument - new example that is more performant and better actuallye11b37a
-
adding Sigmoid as an option for output layer activation
28a7ac6
-
changing name to Numpy-Neuron
b471057 -
getting rid of unessesary imports
e236c9a -
latest update, finishing classification with MNIST! More details on
huggingface repo commits.6d0453c
-
switching to gradio and a complete rewrite that I did over in the
hugging face repo!6307b4f
-
fixed bug that returned same number twice in response
d04aaf5
-
also returning accuracy score from test data from the api
bff0f22
-
ready for prod
336625e -
working on readme and adding ml vis submodule
3ca18fe
-
adding accuracy score tracking
303df47
-
adding CORS support
8ea5c07 -
added loss history in the return to the user so they can make graphs and
stuff2fc4a94 -
must specify werkzwug for flask to work on render
ceca02c -
forgot to add gunicorn
843007e -
ignore and removing dockerfile
abcc2e0 -
adding dockerfile
a4c4ba0 -
adding accuracy score metric
03176c2 -
moving on to testing other datasets!
9e506b7 -
currently debugging mismatched shape error when computing weights and
bias derivatives.880505a -
init weights and biases, and getting through epochs
84bbd7d -
init weights and biases
38e3b7b -
successful example
14841f9 -
handling errors well with the neural netork api
8c348c5 -
working on new python implementation with cleaner code
29cce3f -
initial dumb api working
2c781f8 -
stuck on working with computeOutput function, getting dim error every
time0987346 -
mapping activation arg to the actual function
0a04cd7 -
initializing weights and biases
59ae052