Update README.md
Browse files
README.md
CHANGED
@@ -28,7 +28,7 @@ language:
|
|
28 |
- es
|
29 |
- en
|
30 |
- fr
|
31 |
-
base_model: HiTZ/Medical-
|
32 |
---
|
33 |
|
34 |
<p align="center">
|
@@ -38,7 +38,7 @@ base_model: HiTZ/Medical-mT5-xl
|
|
38 |
for the Medical Domain</h2>
|
39 |
<be>
|
40 |
|
41 |
-
# Model Card for Medical MT5-
|
42 |
|
43 |
|
44 |
<p align="justify">
|
@@ -60,9 +60,9 @@ Medical MT5-xl-multitask is a version of Medical MT5 finetuned for sequence labe
|
|
60 |
<tr>
|
61 |
<th></th>
|
62 |
<th><a href="https://huggingface.co/HiTZ/Medical-mT5-large">HiTZ/Medical-mT5-large</a></th>
|
63 |
-
<th><a href="https://huggingface.co/HiTZ/Medical-
|
64 |
<th><a href="https://huggingface.co/HiTZ/Medical-mT5-large-multitask">HiTZ/Medical-mT5-large-multitask</a></th>
|
65 |
-
<th><a href="https://huggingface.co/HiTZ/Medical-
|
66 |
</tr>
|
67 |
</thead>
|
68 |
<tbody>
|
@@ -95,8 +95,8 @@ Although you can also directly use it with 🤗 huggingface. In order to label
|
|
95 |
import torch
|
96 |
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
97 |
|
98 |
-
model = AutoModelForSeq2SeqLM.from_pretrained("Medical-
|
99 |
-
tokenizer = AutoTokenizer.from_pretrained("Medical-
|
100 |
|
101 |
input_example = "<Disease> Torsade de pointes ventricular tachycardia during low dose intermittent dobutamine treatment in a patient with dilated cardiomyopathy and congestive heart failure ."
|
102 |
|
@@ -119,7 +119,7 @@ print(tokenizer.decode(output[0], skip_special_tokens=True))
|
|
119 |
- **Model type**: text2text-generation
|
120 |
- **Language(s) (NLP)**: English, Spanish, French, Italian
|
121 |
- **License**: apache-2.0
|
122 |
-
- **Finetuned from model**: HiTZ/Medical-
|
123 |
|
124 |
|
125 |
# Ethical Statement
|
|
|
28 |
- es
|
29 |
- en
|
30 |
- fr
|
31 |
+
base_model: HiTZ/Medical-MT5-xl
|
32 |
---
|
33 |
|
34 |
<p align="center">
|
|
|
38 |
for the Medical Domain</h2>
|
39 |
<be>
|
40 |
|
41 |
+
# Model Card for Medical MT5-xl-multitask
|
42 |
|
43 |
|
44 |
<p align="justify">
|
|
|
60 |
<tr>
|
61 |
<th></th>
|
62 |
<th><a href="https://huggingface.co/HiTZ/Medical-mT5-large">HiTZ/Medical-mT5-large</a></th>
|
63 |
+
<th><a href="https://huggingface.co/HiTZ/Medical-MT5-xl">HiTZ/Medical-MT5-xl</a></th>
|
64 |
<th><a href="https://huggingface.co/HiTZ/Medical-mT5-large-multitask">HiTZ/Medical-mT5-large-multitask</a></th>
|
65 |
+
<th><a href="https://huggingface.co/HiTZ/Medical-MT5-xl-multitask">HiTZ/Medical-MT5-xl-multitask</a></th>
|
66 |
</tr>
|
67 |
</thead>
|
68 |
<tbody>
|
|
|
95 |
import torch
|
96 |
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
97 |
|
98 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("Medical-MT5-xl-multitask",torch_dtype=torch.bfloat16, device_map="auto")
|
99 |
+
tokenizer = AutoTokenizer.from_pretrained("Medical-MT5-xl-multitask")
|
100 |
|
101 |
input_example = "<Disease> Torsade de pointes ventricular tachycardia during low dose intermittent dobutamine treatment in a patient with dilated cardiomyopathy and congestive heart failure ."
|
102 |
|
|
|
119 |
- **Model type**: text2text-generation
|
120 |
- **Language(s) (NLP)**: English, Spanish, French, Italian
|
121 |
- **License**: apache-2.0
|
122 |
+
- **Finetuned from model**: HiTZ/Medical-MT5-xl
|
123 |
|
124 |
|
125 |
# Ethical Statement
|