Skip to content

Commit 6188602

Browse files
derjogiderjogi
committed
Update README & add rough limits
1 parent 085f7a4 commit 6188602

File tree

2 files changed

+189
-3
lines changed

2 files changed

+189
-3
lines changed

‎README.md‎

Lines changed: 183 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,185 @@
1-
## Backend for SimScore
1+
# SimScore API
22

3-
Demo: https://sim-score.vercel.app/
3+
An API for semantic similarity analysis and idea ranking.
44

5-
Interactive Jupyter: [![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/derjogi/simscore-api/HEAD?labpath=%2Fcentroid_plots.ipynb)
5+
Demo UI: https://simscore.xyz/
6+
API URL: https://simscore-api.fly.dev
7+
8+
## Features
9+
10+
- Semantic similarity analysis for large sets of ideas
11+
- Cluster analysis with automatic category naming
12+
- Relationship graph generation
13+
- Pairwise similarity matrix
14+
15+
## API Usage
16+
17+
### Basic Analysis
18+
19+
#### Guest Access
20+
SimScore API can be used without authentication, allowing you to try out the basic features with a limited daily quota. Simply make requests without an Authorization header:
21+
22+
```bash
23+
curl -X POST "{{api-url}}/v1/rank-ideas" \
24+
-H "Content-Type: application/json" \
25+
-d '{
26+
"ideas": [
27+
{"id": "1", "idea": "Implement AI chatbot support"},
28+
{"id": "2", "idea": "Add voice recognition features"},
29+
{"id": "3", "idea": "Create automated customer service"}
30+
]
31+
}'
32+
```
33+
34+
(Example) Response:
35+
```json
36+
{
37+
"ranked_ideas": [
38+
{
39+
"id": "1",
40+
"idea": "Implement AI chatbot support",
41+
"similarity_score": 0.89,
42+
"cluster_id": 0
43+
},
44+
{
45+
"id": "3",
46+
"idea": "Create automated customer service",
47+
"similarity_score": 0.85,
48+
"cluster_id": 0
49+
},
50+
{
51+
"id": "2",
52+
"idea": "Add voice recognition features",
53+
"similarity_score": 0.72,
54+
"cluster_id": 1
55+
}
56+
],
57+
"relationship_graph": null,
58+
"pairwise_similarity_matrix": null,
59+
"cluster_names": null
60+
}
61+
```
62+
63+
#### Registered Users
64+
Get higher daily quotas by registering for an account. Registration is straightforward:
65+
66+
Create an account:
67+
```bash
68+
curl -X POST "{{api-url}}/v1/auth/signup" \
69+
-H "Content-Type: application/json" \
70+
-d '{
71+
"email": "[email protected]",
72+
"password": "your_password"
73+
}'
74+
```
75+
76+
This will return your bearer token:
77+
78+
{
79+
"access_token": "your-bearer-token"
80+
}
81+
82+
Once you have registered and need to access your token again, you can retrieve it with
83+
```bash
84+
curl -X POST "{{api-url}}/v1/auth/login" \
85+
-H "Content-Type: application/json" \
86+
-d '{
87+
"email": "[email protected]",
88+
"password": "your_password"
89+
}'
90+
```
91+
92+
Use your token in requests:
93+
```bash
94+
curl -X POST "{{api-url}}/v1/rank-ideas" \
95+
-H "Authorization: Bearer {{your-bearer-token here}}" \
96+
-H "Content-Type: application/json" \
97+
-d '{"ideas": [...]}'
98+
```
99+
100+
#### Enterprise Usage
101+
Need higher limits? Contact the maintainer for custom quotas tailored to your needs.
102+
103+
104+
### Advanced Analysis
105+
```bash
106+
curl -X POST "{{api-url}}/v1/rank-ideas" \
107+
-H "Authorization: Bearer [your-bearer-token here]" \
108+
-H "Content-Type: application/json" \
109+
-d '{
110+
"ideas": [
111+
{"id": "1", "idea": "Implement AI chatbot support"},
112+
{"id": "2", "idea": "Add voice recognition features"}
113+
],
114+
"advanced_features": {
115+
"relationship_graph": true,
116+
"cluster_names": true,
117+
"pairwise_similarity_matrix": true
118+
}
119+
}'
120+
```
121+
122+
(Example) Response:
123+
```json
124+
{
125+
"ranked_ideas": [
126+
{
127+
"id": "1",
128+
"idea": "Implement AI chatbot support",
129+
"similarity_score": 0.89,
130+
"cluster_id": 0
131+
},
132+
{
133+
"id": "2",
134+
"idea": "Add voice recognition features",
135+
"similarity_score": 0.72,
136+
"cluster_id": 1
137+
}
138+
],
139+
"relationship_graph": {
140+
"nodes": [
141+
{
142+
"id": "1",
143+
"coordinates": {"x": 0.8, "y": 0.2}
144+
},
145+
{
146+
"id": "2",
147+
"coordinates": {"x": 0.3, "y": 0.7}
148+
},
149+
{
150+
"id": "Centroid",
151+
"coordinates": {"x": 0.5, "y": 0.5}
152+
}
153+
],
154+
"edges": [
155+
{
156+
"from_id": "1",
157+
"to_id": "2",
158+
"similarity": 0.65
159+
},
160+
{
161+
"from_id": "1",
162+
"to_id": "Centroid",
163+
"similarity": 0.89
164+
},
165+
{
166+
"from_id": "2",
167+
"to_id": "Centroid",
168+
"similarity": 0.72
169+
}
170+
]
171+
},
172+
"cluster_names": {
173+
"0": "AI Customer Support",
174+
"1": "Voice Technologies"
175+
},
176+
"pairwise_similarity_matrix": [
177+
[1.0, 0.65],
178+
[0.65, 1.0]
179+
]
180+
}
181+
```
182+
183+
### Limits
184+
* At least 4 ideas need to be submitted in order for the analysis to run successful
185+
* A maxiumum of 10'000 ideas or 10mb of data (whichever is smaller) will be enforced. Should you require higher limits, please get in touch.

‎app/api/v1/routes/ideas.py‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -53,6 +53,12 @@ async def rank_ideas(
5353
if num_ideas < 4:
5454
return Response(status_code=400, content='Please provide at least 4 items to analyze')
5555

56+
if num_ideas > 10_000:
57+
return Response(status_code=400, content='Please provide less than 10000 items to analyze')
58+
59+
if total_bytes > 10_000_000:
60+
return Response(status_code=400, content='Please provide less than 10MB of data to analyze')
61+
5662
# Check credits for basic analysis
5763
user_id = user_info["user_id"]
5864

0 commit comments

Comments
 (0)