Upload new_captioner.py
Browse files- new_captioner.py +109 -0
new_captioner.py
ADDED
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import base64
|
2 |
+
import requests
|
3 |
+
import os
|
4 |
+
from openai import OpenAI
|
5 |
+
from tqdm import tqdm
|
6 |
+
import time
|
7 |
+
import sys
|
8 |
+
|
9 |
+
# Проверка наличия аргумента командной строки
|
10 |
+
if len(sys.argv) < 2:
|
11 |
+
print("Please, provide the path to image folder.")
|
12 |
+
sys.exit(1)
|
13 |
+
|
14 |
+
# Get the path to image dir from command line.
|
15 |
+
image_dir = sys.argv[1]
|
16 |
+
|
17 |
+
openai_api_key = "EMPTY"
|
18 |
+
openai_api_base = "http://localhost:8000/v1"
|
19 |
+
client = OpenAI(
|
20 |
+
api_key=openai_api_key,
|
21 |
+
base_url=openai_api_base,
|
22 |
+
)
|
23 |
+
|
24 |
+
model_type = client.models.list().data[0].id
|
25 |
+
print(f'model_type: {model_type}')
|
26 |
+
|
27 |
+
# Function to encode the image
|
28 |
+
def encode_image(image_path):
|
29 |
+
with open(image_path, "rb") as image_file:
|
30 |
+
return base64.b64encode(image_file.read()).decode('utf-8')
|
31 |
+
|
32 |
+
# Directories
|
33 |
+
#dir with tags captions from wd tagger
|
34 |
+
txt_dir = './txt/'
|
35 |
+
#dir with result captions
|
36 |
+
maintxt_dir = './maintxt/'
|
37 |
+
image_path =''
|
38 |
+
|
39 |
+
# Ensure the output directory exists
|
40 |
+
os.makedirs(maintxt_dir, exist_ok=True)
|
41 |
+
|
42 |
+
# Get list of all JPEG images in the directory
|
43 |
+
image_files = [f for f in os.listdir(image_dir) if f.lower().endswith(('.jpg', '.jpeg'))]
|
44 |
+
|
45 |
+
total_files = len(image_files)
|
46 |
+
start_time = time.time()
|
47 |
+
|
48 |
+
progress_bar = tqdm(total=total_files, unit='file', bar_format='{l_bar}{bar}| {n_fmt}/{total_fmt} [{elapsed}<{remaining}, {rate_fmt}{postfix}]')
|
49 |
+
total_elapsed_time = 0
|
50 |
+
processed_files = 0
|
51 |
+
|
52 |
+
# Process all images in the image directory
|
53 |
+
for image_file in image_files:
|
54 |
+
image_path = os.path.join(image_dir, image_file)
|
55 |
+
txt_file = os.path.join(txt_dir, os.path.splitext(image_file)[0] + '.txt')
|
56 |
+
output_file = os.path.join(maintxt_dir, os.path.splitext(image_file)[0] + '.txt')
|
57 |
+
|
58 |
+
# Read tags from the corresponding txt file
|
59 |
+
with open(txt_file, 'r') as f:
|
60 |
+
tags = f.read().strip()
|
61 |
+
|
62 |
+
base64_image = encode_image(image_path)
|
63 |
+
|
64 |
+
step_start_time = time.time()
|
65 |
+
|
66 |
+
chat_response = client.chat.completions.create(
|
67 |
+
model="./phi3_v14_800-merged",
|
68 |
+
messages=[{
|
69 |
+
"role": "user",
|
70 |
+
"content": [
|
71 |
+
{"type": "text", "text": f"Make a caption that describe this image. Here is the tags for this image: {tags}"},
|
72 |
+
{
|
73 |
+
"type": "image_url",
|
74 |
+
"image_url": {
|
75 |
+
"url": f"data:image/jpeg;base64,{base64_image}"
|
76 |
+
},
|
77 |
+
},
|
78 |
+
],
|
79 |
+
}],
|
80 |
+
extra_body={'repetition_penalty': 1.05, 'top_k': -1,'top_p': 1,'temperature': 0, 'use_beam_search': True, 'best_of':5},
|
81 |
+
)
|
82 |
+
|
83 |
+
step_end_time = time.time()
|
84 |
+
step_time = step_end_time - step_start_time
|
85 |
+
total_elapsed_time += step_time
|
86 |
+
remaining_time = (total_elapsed_time / (processed_files + 1)) * (total_files - processed_files - 1)
|
87 |
+
|
88 |
+
# Convert remaining time to hours, minutes and seconds
|
89 |
+
remaining_hours = int(remaining_time // 3600)
|
90 |
+
remaining_minutes = int((remaining_time % 3600) // 60)
|
91 |
+
remaining_seconds = int(remaining_time % 60)
|
92 |
+
|
93 |
+
# Extract the content from the response
|
94 |
+
content = chat_response.choices[0].message.content
|
95 |
+
content = content.lstrip()
|
96 |
+
# Write the content to the output file
|
97 |
+
with open(output_file, 'w', encoding='utf-8') as f:
|
98 |
+
f.write(content)
|
99 |
+
|
100 |
+
print(f"\n\nFile {image_file}\nProcessing time: {step_time:.2f} seconds\n{content}")
|
101 |
+
print(f"Response saved to file: {output_file}")
|
102 |
+
|
103 |
+
processed_files += 1
|
104 |
+
progress_bar.update(1)
|
105 |
+
progress_bar.set_postfix(remaining=f'{remaining_hours:02d}:{remaining_minutes:02d}:{remaining_seconds:02d}', refresh=True)
|
106 |
+
|
107 |
+
progress_bar.close()
|
108 |
+
print("All images processed.")
|
109 |
+
print(f"Total time: {time.time() - start_time:.2f} seconds")
|