quoccuongle/MultimodalOCR
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update MiniMonkey.py

Browse Source
This commit is contained in:
Yuliang Liu
2024-08-24 10:39:31 +08:00
committed by GitHub
parent 907e6a0388
commit 6b6fcbba72
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
scripts/MiniMonkey.py
View File

@@ -195,9 +195,9 @@ def eval_worker(args, data, eval_id, output_queue):
image_path = os.path.join(args.image_folder, data[i]['image_path'])
qs = data[i]['question']
pixel_values, target_aspect_ratio = load_image(image_path, min_num=9, max_num=24)
pixel_values, target_aspect_ratio = load_image(image_path, min_num=12, max_num=24)
pixel_values = pixel_values.to(f'cuda:{eval_id}').to(torch.bfloat16)
pixel_values2 = load_image2(image_path, target_aspect_ratio=target_aspect_ratio, min_num=5, max_num=8)
pixel_values2 = load_image2(image_path, target_aspect_ratio=target_aspect_ratio, min_num=3, max_num=11)
pixel_values2 = pixel_values2.to(f'cuda:{eval_id}').to(torch.bfloat16)
pixel_values = torch.cat((pixel_values[:-1], pixel_values2[:-1], pixel_values[-1:]), 0)