alfraser commited on
Commit
a377dbd
Β·
verified Β·
1 Parent(s): 907f936

Updated README to include project outline structure.

Browse files
Files changed (1) hide show
  1. README.md +31 -0
README.md CHANGED
@@ -14,6 +14,37 @@ license: cc-by-sa-4.0
14
 
15
  This project is a demonstration playground for the LLM-enabled architectures built as a submission for the Online MSc in Artificial Intelligence through the University of Bath. The purpose of the project is to explore "LLM-enabled architectures" where an LLM is used in conjunction with some store of private data. The goal is to provide decision support information to technical managers on the _how_ of using LLMs with their organisational data. Specifically by comparing technical architectures and assessing the organisational implications of the technical choices.
16
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
17
  # Demonstration environment
18
 
19
  The project is available as a demonstration running [here on Hugging Face Spaces](https://huggingface.co/spaces/alfraser/llm-arch). This should be the preferred method to interact with the project.
 
14
 
15
  This project is a demonstration playground for the LLM-enabled architectures built as a submission for the Online MSc in Artificial Intelligence through the University of Bath. The purpose of the project is to explore "LLM-enabled architectures" where an LLM is used in conjunction with some store of private data. The goal is to provide decision support information to technical managers on the _how_ of using LLMs with their organisational data. Specifically by comparing technical architectures and assessing the organisational implications of the technical choices.
16
 
17
+ # File Structure
18
+
19
+ ```
20
+ llm-arch
21
+ β”œβ”€β”€ config
22
+ β”‚ β”œβ”€β”€ architectures.json (configuration for the architectures under test and displayed in the UI)
23
+ β”œβ”€β”€ data
24
+ β”‚ β”œβ”€β”€ fine_tuning (data and scripts related to fine-tuning LLMs)
25
+ β”‚ β”œβ”€β”€ json (raw json files containing the synthetic private data built for the project)
26
+ β”‚ β”œβ”€β”€ sqlite
27
+ β”‚ β”‚ β”œβ”€β”€ 01_all_products_dataset.db (sqlite db containing all products generated)
28
+ β”‚ β”‚ β”œβ”€β”€ 02_baseline_dataset.db (sqlite db containing the subset of data selected to be the baseline)
29
+ β”‚ β”‚ β”œβ”€β”€ test_records.db (sqlite database containing the peristed test results)
30
+ β”‚ β”œβ”€β”€ vector_stores
31
+ β”œβ”€β”€ img
32
+ β”œβ”€β”€ pages
33
+ β”œβ”€β”€ src
34
+ β”‚ β”œβ”€β”€ data_synthesis
35
+ β”‚ β”œβ”€β”€ training
36
+ β”‚ β”œβ”€β”€ architectures.py (the core architecture pipeline code including components, and trace)
37
+ β”‚ β”œβ”€β”€ common.py (utilities for common functions, e.g. security token access, data type manipulations)
38
+ β”‚ β”œβ”€β”€ datatypes.py (object oriented representation of the test data and single point for runtime access of the product DB)
39
+ β”‚ β”œβ”€β”€ st_helpers.py (helpers specific to streamlit)
40
+ β”‚ β”œβ”€β”€ testing.py (functionality relating to running, recording and reporting on batches of tests)
41
+ β”œβ”€β”€ Home.py (main entry point for streamlit - first page in the streamlit app)
42
+ β”œβ”€β”€ local_env.yml (conda environment for running project locally)
43
+ β”œβ”€β”€ README.md (readme - this file)
44
+ β”œβ”€β”€ requirements.txt (requirements file for additional requirements in the HF spaces environment - do not use for local running of the project)
45
+ ```
46
+
47
+
48
  # Demonstration environment
49
 
50
  The project is available as a demonstration running [here on Hugging Face Spaces](https://huggingface.co/spaces/alfraser/llm-arch). This should be the preferred method to interact with the project.