pkiage commited on
Commit
37c732e
1 Parent(s): 7e99cca

docs: add setup details

Browse files
Files changed (1) hide show
  1. README.md +59 -2
README.md CHANGED
@@ -13,6 +13,61 @@ Emphasis on:
13
 
14
  - [Streamlit](https://streamlit.io/)
15
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  # Roadmap
17
  Models:
18
  - [ ] Add LightGBM
@@ -23,7 +78,7 @@ Visualization:
23
  - [ ] Add decision surface plot(s)
24
 
25
  Documentation:
26
- - [ ] Add getting started and usage documentation
27
  - [ ] Add documentation evaluating models
28
  - [ ] Add design rationale(s)
29
 
@@ -34,6 +89,8 @@ Other:
34
  - [ ] Add test/code coverage badge
35
  - [ ] Add continuous integration badge
36
 
 
 
37
  # References
38
 
39
  ## Inspiration:
@@ -52,7 +109,7 @@ Other:
52
  - Project structure
53
 
54
  [GraphViz Buildpack](https://github.com/weibeld/heroku-buildpack-graphviz)
55
- - On deployment
56
 
57
  ## Political, Economic, Social, Technological, Legal and Environmental(PESTLE):
58
 
 
13
 
14
  - [Streamlit](https://streamlit.io/)
15
 
16
+ ### Hardware initially built on:
17
+ Processor: 11th Gen Intel(R) Core(TM) i7-1165G7 @2.80Ghz, 2803 Mhz, 4 Core(s), 8 Logical Processor(s)
18
+
19
+ Memory (RAM): 16GB
20
+
21
+ ## Local setup
22
+ ### Obtain the repo locally and open its root folder
23
+ #### To potentially contribute
24
+ git clone https://github.com/pkiage/tool-credit-risk-modelling.git
25
+
26
+ or
27
+
28
+ gh repo clone pkiage/tool-credit-risk-modelling
29
+
30
+ #### Just to deploy locally
31
+ Download ZIP
32
+
33
+ ### (optional) Setup virtual environment:
34
+ python -m venv venv
35
+
36
+ ### (optional) Activate virtual environment:
37
+ #### If using Unix based OS run the following in terminal:
38
+ .\venv\bin\activate
39
+
40
+ #### If using Windows run the following in terminal:
41
+ .\venv\Scripts\activate
42
+
43
+ ### Install requirements by running the following in terminal:
44
+ #### Required packages
45
+ pip install -r requirements.txt
46
+
47
+ #### Complete graphviz installation
48
+ https://graphviz.org/download/
49
+
50
+ ### Run the streamlit app (app.py) by running the following in terminal:
51
+
52
+ streamlit run app.py
53
+
54
+ ## Deployed setup details
55
+ For faster model building and testing (particularly XGBoost) a local setup is recommended.
56
+
57
+ Free Heroku dyno type was used to deploy the app
58
+ https://devcenter.heroku.com/articles/dyno-types
59
+
60
+
61
+ Memory (RAM): 512 MB
62
+
63
+ CPU Share: 1x
64
+
65
+ Compute: 1x-4x
66
+
67
+ Dedicated: no
68
+
69
+ Sleeps: yes
70
+
71
  # Roadmap
72
  Models:
73
  - [ ] Add LightGBM
 
78
  - [ ] Add decision surface plot(s)
79
 
80
  Documentation:
81
+ - [x] Add getting started and usage documentation
82
  - [ ] Add documentation evaluating models
83
  - [ ] Add design rationale(s)
84
 
 
89
  - [ ] Add test/code coverage badge
90
  - [ ] Add continuous integration badge
91
 
92
+
93
+
94
  # References
95
 
96
  ## Inspiration:
 
109
  - Project structure
110
 
111
  [GraphViz Buildpack](https://github.com/weibeld/heroku-buildpack-graphviz)
112
+ - Buildpack used for Heroku deployment
113
 
114
  ## Political, Economic, Social, Technological, Legal and Environmental(PESTLE):
115