initial commit
Browse files
app.py
CHANGED
@@ -1,10 +1,12 @@
|
|
1 |
import gradio as gr
|
2 |
-
from transformers import
|
3 |
from PIL import Image
|
4 |
import torch
|
5 |
|
|
|
6 |
model_path = "microsoft/git-base-vqav2"
|
7 |
dataset_name = "Multimodal-Fatima/OK-VQA_train"
|
|
|
8 |
|
9 |
questions = ["What can happen the objects shown are thrown on the ground?",
|
10 |
"What was the machine beside the bowl used for?",
|
@@ -17,7 +19,8 @@ questions = ["What can happen the objects shown are thrown on the ground?",
|
|
17 |
"Where can that toilet seat be bought?",
|
18 |
"What do you call the kind of pants that the man on the right is wearing?"]
|
19 |
|
20 |
-
|
|
|
21 |
|
22 |
|
23 |
def main(select_exemple_num):
|
|
|
1 |
import gradio as gr
|
2 |
+
from transformers import AutoProcessor, AutoModelForVisualQuestionAnswering, AutoModelForCausalLM, AutoTokenizer
|
3 |
from PIL import Image
|
4 |
import torch
|
5 |
|
6 |
+
model = AutoModelForCausalLM.from_pretrained("microsoft/git-base-vqav2")
|
7 |
model_path = "microsoft/git-base-vqav2"
|
8 |
dataset_name = "Multimodal-Fatima/OK-VQA_train"
|
9 |
+
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
10 |
|
11 |
questions = ["What can happen the objects shown are thrown on the ground?",
|
12 |
"What was the machine beside the bowl used for?",
|
|
|
19 |
"Where can that toilet seat be bought?",
|
20 |
"What do you call the kind of pants that the man on the right is wearing?"]
|
21 |
|
22 |
+
processor = AutoProcessor.from_pretrained(model_path)
|
23 |
+
model = AutoModelForVisualQuestionAnswering.from_pretrained(model_path)
|
24 |
|
25 |
|
26 |
def main(select_exemple_num):
|