Jensen-holm commited on
Commit
b471057
β€’
1 Parent(s): e236c9a

changing name to Numpy-Neuron

Browse files
Files changed (2) hide show
  1. README.md +33 -3
  2. app.py +2 -2
README.md CHANGED
@@ -1,5 +1,5 @@
1
  ---
2
- title: Backprop Playground
3
  emoji: πŸ”™
4
  colorFrom: yellow
5
  colorTo: blue
@@ -10,5 +10,35 @@ pinned: false
10
  license: mit
11
  ---
12
 
13
- This web app uses a neural network framework that I built from scratch in <br>
14
- python, using numpy as the only 3rd party library in the framework itself. <br>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ title: Numpy-Neuron
3
  emoji: πŸ”™
4
  colorFrom: yellow
5
  colorTo: blue
 
10
  license: mit
11
  ---
12
 
13
+ ## What is this? <br>
14
+
15
+ The Numpy-Neuron is a GUI built around a neural network framework that I have built from scratch
16
+ in [numpy](https://numpy.org/). In this GUI, you can test different hyper parameters that will be fed to this framework and used
17
+ to train a neural network on the [MNIST](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_digits.html) dataset of 8x8 pixel images.
18
+
19
+ ## ⚠️ PLEASE READ ⚠️
20
+ This application is impossibly slow on the HuggingFace CPU instance that it is running on. It is advised to clone the
21
+ repository and run it locally.
22
+
23
+ In order to get a decent classification score on the validation set of the MNIST data (hard coded to 20%), you will have to
24
+ do somewhere between 15,000 epochs and 50,000 epochs with a learning rate around 0.001, and a hidden layer size
25
+ over 10. (roughly the example that I have provided). Running this many epochs with a hidden layer of that size
26
+ is pretty expensive on 2 cpu cores that this space has. So if you are actually curious, you might want to clone
27
+ this and run it locally because it will be much much faster.
28
+
29
+ `git clone https://huggingface.co/spaces/Jensen-holm/Numpy-Neuron`
30
+
31
+ After cloning, you will have to install the dependencies from requirements.txt into your environment. (venv reccommended)
32
+
33
+ `pip3 install -r requirements.txt`
34
+
35
+ Then, you can run the application on local host with the following command.
36
+
37
+ `python3 app.py`
38
+
39
+
40
+ ## Development
41
+
42
+ In order to push from this GitHub repo to the hugging face space:
43
+
44
+ `git push --force space main`
app.py CHANGED
@@ -75,7 +75,7 @@ def classification(
75
 
76
  if __name__ == "__main__":
77
  with gr.Blocks() as interface:
78
- gr.Markdown("# Backpropagation Playground")
79
  gr.Markdown(
80
  """
81
  ## What is this? <br>
@@ -94,7 +94,7 @@ if __name__ == "__main__":
94
  is pretty expensive on 2 cpu cores that this space has. So if you are actually curious, you might want to clone
95
  this and run it locally because it will be much much faster.
96
 
97
- `git clone https://huggingface.co/spaces/Jensen-holm/Backprop-Playground`
98
 
99
  After cloning, you will have to install the dependencies from requirements.txt into your environment. (venv reccommended)
100
 
 
75
 
76
  if __name__ == "__main__":
77
  with gr.Blocks() as interface:
78
+ gr.Markdown("# Numpy Neuron")
79
  gr.Markdown(
80
  """
81
  ## What is this? <br>
 
94
  is pretty expensive on 2 cpu cores that this space has. So if you are actually curious, you might want to clone
95
  this and run it locally because it will be much much faster.
96
 
97
+ `git clone https://huggingface.co/spaces/Jensen-holm/Numpy-Neuron`
98
 
99
  After cloning, you will have to install the dependencies from requirements.txt into your environment. (venv reccommended)
100