SoulFireMage commited on
Commit
4b8efea
·
verified ·
1 Parent(s): 11fb68b

Space config and readme

Browse files
Files changed (1) hide show
  1. README.md +41 -5
README.md CHANGED
@@ -1,13 +1,49 @@
1
  ---
2
- title: Math Sight Reading
3
- emoji: 🏃
4
- colorFrom: purple
5
  colorTo: purple
6
  sdk: gradio
7
  sdk_version: 6.29.0
8
- python_version: '3.13'
9
  app_file: app.py
10
  pinned: false
 
 
11
  ---
12
 
13
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ title: Mathematical Sight-Reading
3
+ emoji: 🎼
4
+ colorFrom: indigo
5
  colorTo: purple
6
  sdk: gradio
7
  sdk_version: 6.29.0
8
+ python_version: "3.11"
9
  app_file: app.py
10
  pinned: false
11
+ hf_oauth: true
12
+ short_description: Drill reading dense maths notation like sheet music
13
  ---
14
 
15
+ # Mathematical Sight-Reading Framework (MSRF)
16
+
17
+ Train mathematical literacy by treating notation like sheet music: decode the symbols,
18
+ explain the structure, then ground it in more than one domain.
19
+
20
+ You get a **naked formula** (no name, no context). You answer in three parts
21
+ (formal sight-reading, concept and mechanics, application and metaphor). A judge model scores
22
+ **Precision** (0-10), **Concept** (Superficial / Competent / Exemplary) and **Transfer** (A / B / C),
23
+ then shows what you missed and an exemplary sight-read.
24
+
25
+ ## How it works
26
+
27
+ - **Formula bank** (`bank.py`): 16 formulas over four tiers, each with a hidden reference
28
+ sight-read. The judge grades against that reference, not its own memory, which is what keeps
29
+ scoring stable. Add formulas by appending to the list.
30
+ - **Adaptive tiers** (`engine.py`): rolling mean of your composite score moves you up or down a tier.
31
+ Plain code, no LLM involved.
32
+ - **Spaced repetition**: poorly answered formulas return after a few drills (Leitner boxes).
33
+ - **Hint ladder**: hint 1 = types, hint 2 = mechanics. Each costs 0.1 of the composite.
34
+ - **Weak-spot report and export/import** on the Progress tab.
35
+ - **Judge** (`judge.py`): your own OpenRouter key, any model id. Default set with the
36
+ `JUDGE_MODEL` variable; dropdown choices with `JUDGE_MODELS` (comma separated).
37
+
38
+ ## Privacy
39
+
40
+ Your key, progress and model choice live in your browser's localStorage. The key passes through
41
+ the Space server in memory only to make the OpenRouter call; it is not logged or stored.
42
+
43
+ ## Tests
44
+
45
+ ```
46
+ pip install gradio pytest playwright itsdangerous authlib
47
+ python -m pytest tests/test_core.py
48
+ python tests/test_ui.py # browser test against a fake OpenRouter
49
+ ```