johngoad commited on
Commit
ed1a677
1 Parent(s): 70af6b0

Upload app2.py with huggingface_hub

Browse files
Files changed (1) hide show
  1. app2.py +50 -0
app2.py ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import torch
2
+ import gradio as gr
3
+ import re
4
+ from transformers import AutoTokenizer, ViTFeatureExtractor, VisionEncoderDecoderModel
5
+
6
+ device='cpu'
7
+ encoder_checkpoint = "nlpconnect/vit-gpt2-image-captioning"
8
+ decoder_checkpoint = "nlpconnect/vit-gpt2-image-captioning"
9
+ model_checkpoint = "nlpconnect/vit-gpt2-image-captioning"
10
+ feature_extractor = ViTFeatureExtractor.from_pretrained(encoder_checkpoint)
11
+ tokenizer = AutoTokenizer.from_pretrained(decoder_checkpoint)
12
+ model = VisionEncoderDecoderModel.from_pretrained(model_checkpoint).to(device)
13
+
14
+ def predict(image,max_length=64, num_beams=4):
15
+ image = image.convert('RGB')
16
+ image = feature_extractor(image, return_tensors="pt").pixel_values.to(device)
17
+ clean_text = lambda x: x.replace('<|endoftext|>','').split('\n')[0]
18
+ caption_ids = model.generate(image, max_length = max_length)[0]
19
+ caption_text = clean_text(tokenizer.decode(caption_ids))
20
+ return caption_text
21
+
22
+ def set_example_image(example: list) -> dict:
23
+ return gr.Image.update(value=example[0])
24
+ css = '''
25
+ h1#title {
26
+ text-align: center;
27
+ }
28
+ h3#header {
29
+ text-align: center;
30
+ }
31
+ img#overview {
32
+ max-width: 800px;
33
+ max-height: 600px;
34
+ }
35
+ img#style-image {
36
+ max-width: 1000px;
37
+ max-height: 600px;
38
+ }
39
+ '''
40
+ demo = gr.Blocks(css=css)
41
+ with demo:
42
+ gr.Markdown('''<h1 id="title">Image Caption 🖼️</h1>''')
43
+ gr.Markdown('''Made by : Shreyas Dixit''')
44
+ with gr.Column():
45
+ input = gr.inputs.Image(label="Upload your Image", type = 'pil', optional=True)
46
+ output = gr.outputs.Textbox(type="auto",label="Captions")
47
+ btn = gr.Button("Genrate Caption")
48
+ btn.click(fn=predict, inputs=input, outputs=output)
49
+
50
+ demo.launch()