awacke1 commited on
Commit
21496d1
·
verified ·
1 Parent(s): 5c7c87f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +47 -0
README.md CHANGED
@@ -14,3 +14,50 @@ Check out the configuration reference at https://huggingface.co/docs/hub/spaces-
14
 
15
  Inspired by this video which discusses how Anthropic is studying how their model works using Dictionary Learning.
16
  https://youtu.be/AYTYk-kmXuI?t=168
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
 
15
  Inspired by this video which discusses how Anthropic is studying how their model works using Dictionary Learning.
16
  https://youtu.be/AYTYk-kmXuI?t=168
17
+
18
+
19
+ Importing Libraries:
20
+
21
+ streamlit for creating the app.
22
+ Scikit-learn (sklearn.decomposition.DictionaryLearning and datasets.load_digits) for dictionary learning and loading the dataset.
23
+ matplotlib for plotting.
24
+ numpy for numerical operations.
25
+ Streamlit App Initialization:
26
+
27
+ The app sets a title and a description with st.title and st.write.
28
+ Loading Data:
29
+
30
+ Digit data is loaded using load_digits(). This dataset contains 8x8 images of digits (0-9).
31
+ Sample Image Display:
32
+
33
+ A slider is created with st.slider to select an index for displaying a sample image from the dataset using st.image.
34
+ Selecting Number of Components:
35
+
36
+ Another slider allows users to choose the number of dictionary components (n_components) for dictionary learning.
37
+ Performing Dictionary Learning:
38
+
39
+ DictionaryLearning is initialized with the selected number of components and fitted to the data.
40
+ The dictionary elements learned are stored in dictionary.
41
+ Displaying Dictionary Components:
42
+
43
+ The learned dictionary components are displayed as images using Matplotlib. Each component is reshaped into an 8x8 image.
44
+ Displaying Sparsity:
45
+
46
+ Sparsity is calculated as the mean of zero entries in the transformed data (X_transformed) and displayed with st.write.
47
+
48
+
49
+ Assistant: Sure! The requirements.txt file is used to specify the dependencies needed for your Python project. For the Streamlit application demonstrating dictionary learning, you'll need the following libraries:
50
+
51
+ streamlit to create the interactive web app.
52
+ scikit-learn for performing dictionary learning and loading the dataset.
53
+ matplotlib for plotting the images.
54
+ numpy for numerical operations.
55
+ Here is a requirements.txt file that includes all these dependencies:
56
+
57
+ streamlit>=1.0.0
58
+ scikit-learn>=0.24.0
59
+ matplotlib>=3.0.0
60
+ numpy>=1.19.0
61
+
62
+
63
+