Translation
Transformers
PyTorch
Safetensors
English
t5
text2text-generation
united states air force
united states space force
department of defense
dod
usaf
ussf
afi
air force
space force
bullets
performance reports
evaluations
awards
opr
epr
narratives
interpreter
mbzuai
lamini-flan-t5-783m
flan-t5
google
opera
justinthelaw
text-generation-inference
Inference Endpoints
justinthelaw
commited on
Commit
•
45b57c5
1
Parent(s):
73156f3
Docs(Widget): Added 3 Bullet Examples
Browse files
README.md
CHANGED
@@ -22,6 +22,13 @@ tags:
|
|
22 |
- google
|
23 |
- opera
|
24 |
- justinthelaw
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
25 |
---
|
26 |
|
27 |
# Model Card for Opera Bullet Interpreter
|
@@ -43,7 +50,7 @@ To learn more about this project, please visit the [Opera GitHub Repository](htt
|
|
43 |
- [Evaluation](#evaluation)
|
44 |
- [Model Examination](#model-examination)
|
45 |
- [Environmental Impact](#environmental-impact)
|
46 |
-
- [Technical Specifications
|
47 |
- [Citation](#citation)
|
48 |
- [Model Card Authors](#model-card-authors-optional)
|
49 |
- [Model Card Contact](#model-card-contact)
|
@@ -72,7 +79,13 @@ This is a fine-tuned version of the LaMini-Flan-T5-783M, using the justinthelaw/
|
|
72 |
|
73 |
Used to programmatically produce training data for Opera's Bullet Forge (see GitHub repository for details).
|
74 |
|
75 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
76 |
|
77 |
Used to quickly interpret bullets written by Airman (Air Force) or Guardians (Space Force), into long-form, plain English sentences.
|
78 |
|
@@ -92,8 +105,6 @@ Look-up acronyms to ensure the correct narrative is being formed. Double-check (
|
|
92 |
|
93 |
## Training Data
|
94 |
|
95 |
-
pre-processing or additional filtering. -->
|
96 |
-
|
97 |
The model was fine-tuned on the justinthelaw/opera-bullet-completions dataset, which can be partially found at the GitHub repository.
|
98 |
|
99 |
## Training Procedure
|
@@ -170,8 +181,6 @@ VSCode, Jupyter Notebook, Python3, PyTorch, Transformers, Pandas, Asyncio, Logur
|
|
170 |
|
171 |
# Model Card Authors
|
172 |
|
173 |
-
construction? Etc. -->
|
174 |
-
|
175 |
Justin Law, Alden Davidson, Christopher Kodama, My Tran
|
176 |
|
177 |
# Model Card Contact
|
@@ -191,7 +200,8 @@ from transformers import T5ForConditionalGeneration, T5Tokenizer
|
|
191 |
|
192 |
bullet_data_creation_prefix = (
|
193 |
"Using upwards of 3 sentences, expand upon the following Air and Space Force bullet statement by "
|
194 |
-
+ "spelling-out acronyms and adding additional context that is not already included in the
|
|
|
195 |
)
|
196 |
|
197 |
# Path of the pre-trained model that will be used
|
|
|
22 |
- google
|
23 |
- opera
|
24 |
- justinthelaw
|
25 |
+
widget:
|
26 |
+
- text: "Using upwards of 3 sentences, expand upon the following Air and Space Force bullet statement by spelling-out acronyms and adding additional context that is not already included in the Air and Space Force bullet statement: - EFMP SME; dir'd 57 enrollments/194 incoming/101 outgoing inquiries--beat pkg processing time 50%/on-time rt 99%"
|
27 |
+
example_title: "Bullet Interpretation 1"
|
28 |
+
- text: "Using upwards of 3 sentences, expand upon the following Air and Space Force bullet statement by spelling-out acronyms and adding additional context that is not already included in the Air and Space Force bullet statement: - Steered F22 DMS ops; led 9 mbrs/provided 24 hr MXS spt/returned 54 AFREP items--saved $258K/nailed 2K sorties"
|
29 |
+
example_title: "Bullet Interpretation 2"
|
30 |
+
- text: "Using upwards of 3 sentences, expand upon the following Air and Space Force bullet statement by spelling-out acronyms and adding additional context that is not already included in the Air and Space Force bullet statement: - Setup Ex KEY RESOLVE '10 comms; installed 35 linesran 500ft cabling--supported multi-national operations"
|
31 |
+
example_title: "Bullet Interpretation 3"
|
32 |
---
|
33 |
|
34 |
# Model Card for Opera Bullet Interpreter
|
|
|
50 |
- [Evaluation](#evaluation)
|
51 |
- [Model Examination](#model-examination)
|
52 |
- [Environmental Impact](#environmental-impact)
|
53 |
+
- [Technical Specifications](#technical-specifications-optional)
|
54 |
- [Citation](#citation)
|
55 |
- [Model Card Authors](#model-card-authors-optional)
|
56 |
- [Model Card Contact](#model-card-contact)
|
|
|
79 |
|
80 |
Used to programmatically produce training data for Opera's Bullet Forge (see GitHub repository for details).
|
81 |
|
82 |
+
The exact prompt to achieve the desired result is:
|
83 |
+
|
84 |
+
```
|
85 |
+
Using upwards of 3 sentences, expand upon the following Air and Space Force bullet statement by spelling-out acronyms and adding additional context that is not already included in the Air and Space Force bullet statement: <INSERT BULLET HERE>"
|
86 |
+
```
|
87 |
+
|
88 |
+
## Downstream Use
|
89 |
|
90 |
Used to quickly interpret bullets written by Airman (Air Force) or Guardians (Space Force), into long-form, plain English sentences.
|
91 |
|
|
|
105 |
|
106 |
## Training Data
|
107 |
|
|
|
|
|
108 |
The model was fine-tuned on the justinthelaw/opera-bullet-completions dataset, which can be partially found at the GitHub repository.
|
109 |
|
110 |
## Training Procedure
|
|
|
181 |
|
182 |
# Model Card Authors
|
183 |
|
|
|
|
|
184 |
Justin Law, Alden Davidson, Christopher Kodama, My Tran
|
185 |
|
186 |
# Model Card Contact
|
|
|
200 |
|
201 |
bullet_data_creation_prefix = (
|
202 |
"Using upwards of 3 sentences, expand upon the following Air and Space Force bullet statement by "
|
203 |
+
+ "spelling-out acronyms and adding additional context that is not already included in the " +
|
204 |
+
"Air and Space Force bullet statement: "
|
205 |
)
|
206 |
|
207 |
# Path of the pre-trained model that will be used
|