rajistics commited on
Commit
a6072b2
1 Parent(s): 0022d33

removed a bunch of text

Browse files
Files changed (1) hide show
  1. app.py +1 -43
app.py CHANGED
@@ -30,47 +30,6 @@ FIM_SUFFIX = "<fim_suffix>"
30
 
31
  FIM_INDICATOR = "<FILL_HERE>"
32
 
33
- FORMATS = """## Model Formats
34
-
35
- The model is pretrained on code and is formatted with special tokens in addition to the pure code data,\
36
- such as prefixes specifying the source of the file or tokens separating code from a commit message.\
37
- Use these templates to explore the model's capacities:
38
-
39
- ### 1. Prefixes 🏷️
40
- For pure code files, use any combination of the following prefixes:
41
-
42
- ```
43
- <reponame>REPONAME<filename>FILENAME<gh_stars>STARS\ncode<|endoftext|>
44
- ```
45
- STARS can be one of: 0, 1-10, 10-100, 100-1000, 1000+
46
-
47
- ### 2. Commits 💾
48
- The commits data is formatted as follows:
49
-
50
- ```
51
- <commit_before>code<commit_msg>text<commit_after>code<|endoftext|>
52
- ```
53
-
54
- ### 3. Jupyter Notebooks 📓
55
- The model is trained on Jupyter notebooks as Python scripts and structured formats like:
56
-
57
- ```
58
- <start_jupyter><jupyter_text>text<jupyter_code>code<jupyter_output>output<jupyter_text>
59
- ```
60
-
61
- ### 4. Issues 🐛
62
- We also trained on GitHub issues using the following formatting:
63
- ```
64
- <issue_start><issue_comment>text<issue_comment>...<issue_closed>
65
- ```
66
-
67
- ### 5. Fill-in-the-middle 🧩
68
- Fill in the middle requires rearranging the model inputs. The playground handles this for you - all you need is to specify where to fill:
69
- ```
70
- code before<FILL_HERE>code after
71
- ```
72
- """
73
-
74
  theme = gr.themes.Monochrome(
75
  primary_hue="indigo",
76
  secondary_hue="blue",
@@ -156,7 +115,7 @@ description = """
156
  <ul>
157
  <li><a href="https://huggingface.co/smallcloudai/Refact-1_6B-fim" style='color: #e6b800;'>ReFact 1.6B</a>: An Open-Source Coding Assistant with Fine-Tuning on codebase, autocompletion, code refactoring, code analysis, integrated chat and more</li>
158
  </ul>
159
- <p><b>Please note:</b> These models are not designed for instruction purposes. If you're looking for instruction or want to chat with a fine-tuned model, you can visit the <a href="https://huggingface.co/spaces/HuggingFaceH4/starchat-playground">StarChat Playground</a>.</p>
160
  </div>
161
  """
162
 
@@ -237,7 +196,6 @@ with gr.Blocks(theme=theme, analytics_enabled=False, css=css) as demo:
237
  fn=process_example,
238
  outputs=[output],
239
  )
240
- gr.Markdown(FORMATS)
241
 
242
  submit.click(
243
  generate,
 
30
 
31
  FIM_INDICATOR = "<FILL_HERE>"
32
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  theme = gr.themes.Monochrome(
34
  primary_hue="indigo",
35
  secondary_hue="blue",
 
115
  <ul>
116
  <li><a href="https://huggingface.co/smallcloudai/Refact-1_6B-fim" style='color: #e6b800;'>ReFact 1.6B</a>: An Open-Source Coding Assistant with Fine-Tuning on codebase, autocompletion, code refactoring, code analysis, integrated chat and more</li>
117
  </ul>
118
+ <p><b>Please note:</b> This demo is based on the Big Code Playground. It is running on CPU, but can be run faster using GPU.</p>
119
  </div>
120
  """
121
 
 
196
  fn=process_example,
197
  outputs=[output],
198
  )
 
199
 
200
  submit.click(
201
  generate,