omidrohanian
commited on
Commit
•
dbdac9b
1
Parent(s):
6dd65e3
Create README.md
Browse files
README.md
CHANGED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
---
|
4 |
+
|
5 |
+
# Model
|
6 |
+
The Pandemic PACT Advanced Categorisation Engine (PPACE) is a fine-tuned 8B LLM designed for automatically classifying research abstracts from funded biomedical projects according to WHO-aligned research priorities. Developed as part of the GLOPID-R Pandemic PACT project, PPACE assists in tracking and analysing research funding and clinical evidence for a wide range of diseases with outbreak potential.
|
7 |
+
|
8 |
+
The model leverages a human-annotated dataset expanded with rationales generated by a larger LLM. These rationales provide explanations for the chosen labels, enhancing the model's interpretability and accuracy.
|
9 |
+
|
10 |
+
# Usage
|
11 |
+
Todo
|
12 |
+
|
13 |
+
# Model Details
|
14 |
+
PPACE is fine-tuned using Low-Rank Adaptation (LoRA) to ensure efficient training while maintaining high performance. The fine-tuning process involves training the model for 4 epochs on a dataset of 5142 projects, using 8 A100 GPUs with a batch size of 1 per GPU and 4 gradient accumulation steps.
|
15 |
+
|
16 |
+
## Hyperparameters
|
17 |
+
|
18 |
+
| Hyperparameter | Value |
|
19 |
+
|---------------------------|--------|
|
20 |
+
| Total Batch Size | 2 |
|
21 |
+
| Gradient Accumulation Steps | 4 |
|
22 |
+
| Learning Rate | 2e-4 |
|
23 |
+
| LR Scheduler | Linear |
|
24 |
+
| Epochs | 2 |
|
25 |
+
| LoRA Rank | 128 |
|
26 |
+
| LoRA α | 256 |
|
27 |
+
| LoRA Dropout | 0.05 |
|