Spaces:
Running
Running
Alex Stern
commited on
Commit
·
460c08a
1
Parent(s):
d4fb664
Update README.md
Browse files
README.md
CHANGED
@@ -10,3 +10,48 @@ pinned: false
|
|
10 |
---
|
11 |
|
12 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
---
|
11 |
|
12 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
13 |
+
|
14 |
+
---
|
15 |
+
# TTV Email Draft Generator - ReadMe
|
16 |
+
|
17 |
+
## Overview
|
18 |
+
The TTV Email Draft Generator is a tool designed to streamline email response processes for TTV. Leveraging OpenAI's GPT models, it assists in crafting coherent, contextually relevant email drafts quickly.
|
19 |
+
|
20 |
+
## Key Features
|
21 |
+
- AI-generated email responses.
|
22 |
+
- Customizable response tone and length.
|
23 |
+
- Integration with GPT-3.5 and GPT-4 models.
|
24 |
+
|
25 |
+
## Functions Description
|
26 |
+
|
27 |
+
### `generate_ttv_email_response(email_content, bullet_points, tone, length, model_choice)`
|
28 |
+
This function generates an email response based on provided content, bullet points, tone, and length, interfacing with the chosen AI model (GPT-3.5 or GPT-4) to create a draft that aligns with TTV's communication style.
|
29 |
+
|
30 |
+
### `sig_figs(x, precision)`
|
31 |
+
Rounds a number to a specified number of significant figures, useful for displaying cost estimates with appropriate precision.
|
32 |
+
|
33 |
+
### `calculate_token_usage_price(response, model_choice)`
|
34 |
+
Calculates the cost of using the OpenAI API, based on the number of tokens used in the request and response, adjusting for the selected model.
|
35 |
+
|
36 |
+
### `word_count(word)`
|
37 |
+
Counts the number of words in a given string, used to set the 'Matching Client's Email Length' option.
|
38 |
+
|
39 |
+
### `get_email_length(email_content, length)`
|
40 |
+
Determines the desired length of the email response based on user input and the original email's length, key for customizing the AI-generated response's length.
|
41 |
+
|
42 |
+
### `gradio_interface()`
|
43 |
+
Defines and launches the Gradio interface, setting up input fields, output displays, and the overall application layout for intuitive use.
|
44 |
+
|
45 |
+
## Deployment
|
46 |
+
Hosted on Hugging Face Spaces for easy access and usability.
|
47 |
+
|
48 |
+
## Usage
|
49 |
+
Access via the Hugging Face Spaces link, input original email content, specify bullet points, select desired tone and length, and choose an AI model. Submit to receive a generated response and cost estimate.
|
50 |
+
|
51 |
+
## Future Enhancements
|
52 |
+
Potential updates include direct email client integration, a customized user interface, and further AI response quality refinements.
|
53 |
+
|
54 |
+
---
|
55 |
+
|
56 |
+
*For detailed instructions and support, refer to the user manual or contact the development team.*
|
57 |
+
|