land14 commited on
Commit
0cabb57
·
1 Parent(s): 0a67a24

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +59 -0
README.md CHANGED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ tags:
4
+ - Explain code
5
+ - Code Summarization
6
+
7
+ license: mit
8
+ ---
9
+
10
+
11
+ # Gemini
12
+
13
+ ## Model description
14
+
15
+ Gemini is a transformer based on Google's T5 model. The model is pre-trained on approximately 800k code/description pairs and then fine-tuned on 10k higher-level explanations that were synthetically generated. Gemini is capable of summarization/explaining short to medium code snippets in:
16
+
17
+ - Python
18
+ - Javascript (mostly vanilla JS, however, it can handle frameworks like React as well)
19
+ - Java
20
+ - Ruby
21
+ - Go
22
+
23
+ And outputs a description in English.
24
+
25
+ ## Intended uses & limitations
26
+
27
+ Gemini without any additional fine-tuning is capable of explaining code in a sentence or two and typically performs best in Python and Javascript. We recommend using Gemini for either simple code explanation, documentation or producing more synthetic data to improve its explanations.
28
+
29
+ ### How to use
30
+
31
+ You can use this model directly with a pipeline for Text2Text generation, as shown below:
32
+
33
+ ```python
34
+ from transformers import pipeline, set_seed
35
+
36
+ summarizer = pipeline('text2text-generation', model='describeai/gemini')
37
+ code = "print('hello world!')"
38
+
39
+ response = "Summarized code: "+ summarizer(code, max_length=100, num_beams=3)
40
+ print(response)
41
+
42
+ ```
43
+
44
+ Which should yield something along the lines of:
45
+
46
+ ```
47
+ Summarized code: The following code is greeting the world.
48
+ ```
49
+
50
+
51
+
52
+ ### Limitations
53
+
54
+ Typically, Gemini may produce overly simplistic descriptions that don't encompass the entire code snippet. We suspect with more training data, this could be circumvented and will produce better results.
55
+
56
+
57
+ ### About Us
58
+
59
+ A Describe.ai, we are focused on building Artificial Intelligence systems that can understand language as well as humans. While a long path, we plan to contribute our findings to our API to the Open Source community.