joshnguyen
commited on
Commit
·
3ee1192
1
Parent(s):
4fc6d9e
Update README.md
Browse files
README.md
CHANGED
@@ -13,6 +13,52 @@ tags:
|
|
13 |
- astronomy
|
14 |
- astrophysics
|
15 |
---
|
16 |
-
## Training procedure
|
17 |
|
18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
- astronomy
|
14 |
- astrophysics
|
15 |
---
|
|
|
16 |
|
17 |
+
<p><h1>AstroLLaMA</h1></p>
|
18 |
+
|
19 |
+
<p align="center">
|
20 |
+
<img src="https://huggingface.co/universeTBD/astrollama/resolve/main/images/astrollama-logo.png" alt="AstroLLaMA" width="500px"/>
|
21 |
+
</p>
|
22 |
+
|
23 |
+
## Loading the model
|
24 |
+
|
25 |
+
```python
|
26 |
+
from transformers import AutoModelForCausalLM
|
27 |
+
from transformers import AutoTokenizer
|
28 |
+
|
29 |
+
tokenizer = AutoTokenizer.from_pretrained(
|
30 |
+
pretrained_model_name_or_path="universeTBD/astrollama"
|
31 |
+
)
|
32 |
+
model = AutoModelForCausalLM.from_pretrained(
|
33 |
+
pretrained_model_name_or_path="universeTBD/astrollama",
|
34 |
+
device_map="auto",
|
35 |
+
)
|
36 |
+
```
|
37 |
+
|
38 |
+
## Generating text from a prompt
|
39 |
+
|
40 |
+
```python
|
41 |
+
import torch
|
42 |
+
from transformers import pipeline
|
43 |
+
|
44 |
+
generator = pipeline(
|
45 |
+
task="text-generation",
|
46 |
+
model=model,
|
47 |
+
tokenizer="tokenizer",
|
48 |
+
device_map="auto"
|
49 |
+
)
|
50 |
+
|
51 |
+
# Taken from https://arxiv.org/abs/2308.12823
|
52 |
+
prompt = "In this letter, we report the discovery of the highest redshift, " \
|
53 |
+
"heavily obscured, radio-loud QSO candidate selected using JWST NIRCam/MIRI, " \
|
54 |
+
"mid-IR, sub-mm, and radio imaging in the COSMOS-Web field. "
|
55 |
+
|
56 |
+
# For reproducibility
|
57 |
+
torch.manual_seed(42)
|
58 |
+
|
59 |
+
generated_text = generator(
|
60 |
+
prompt,
|
61 |
+
do_sample=True,
|
62 |
+
max_length=512
|
63 |
+
)
|
64 |
+
```
|