MattStammers
commited on
Commit
•
bba32dc
1
Parent(s):
692974d
Create README.md
Browse filesA nearly useless finetuned version of Mixtral7b
README.md
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
language: en
|
2 |
+
tags:
|
3 |
+
- snomed-ct
|
4 |
+
- text-generation
|
5 |
+
---
|
6 |
+
|
7 |
+
# My Model Name
|
8 |
+
|
9 |
+
## Model description
|
10 |
+
This is a text generation model for SNOMED-CT. As it is text-generation, it is prone to hallucination and should not be used for any kind of production purpose but it was fun to build. It is based on Mixtral7b and was fine-tuned on a part of the SNOMED-CT corpus then tested against a gold-standard.
|
11 |
+
|
12 |
+
## How to use
|
13 |
+
Provide code snippets on how to use your model.
|
14 |
+
|
15 |
+
```python
|
16 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
17 |
+
|
18 |
+
model_name = "{username}/{model_id}"
|
19 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
20 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
21 |
+
```
|
22 |
+
|
23 |
+
Model Performance
|
24 |
+
Accuracy: {metrics["0"]}
|
25 |
+
Precision: {metrics["0"]}
|
26 |
+
Recall: {metrics["0"]}
|
27 |
+
|
28 |
+
ParameterName SNOMEDCode ExtractedSNOMEDNumbers CorrectPrediction
|
29 |
+
*Heart rate 364075005 3222222 False
|
30 |
+
|
31 |
+
Limitations and bias
|
32 |
+
It is really low grade but I had fun building it so have pushed it up
|
33 |
+
|
34 |
+
Acknowledgments
|
35 |
+
Thanks to the Mixtral AI team for creating the base model for this one.
|