File size: 22,882 Bytes

f8ba0eb

2025-08-21 01:42:04 - INFO - Loading model: Qwen/Qwen2.5-VL-3B-Instruct-AWQ
2025-08-21 01:42:09 - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
2025-08-21 01:42:40 - INFO - Model loaded in 35.77 seconds
2025-08-21 01:42:40 - INFO - GPU Memory Usage after model load: 3250.55 MB
2025-08-21 02:54:09 - INFO - [c40f2273-a9f5-4d96-82d4-990269ab9708] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_001.mp4'
2025-08-21 02:54:09 - INFO - [c40f2273-a9f5-4d96-82d4-990269ab9708] Video saved to temporary file: temp_videos/c40f2273-a9f5-4d96-82d4-990269ab9708.mp4
2025-08-21 02:54:09 - INFO - [c40f2273-a9f5-4d96-82d4-990269ab9708] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:54:13 - INFO - [c40f2273-a9f5-4d96-82d4-990269ab9708] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:54:13 - INFO - [c40f2273-a9f5-4d96-82d4-990269ab9708] 30 frames saved to temp_videos/c40f2273-a9f5-4d96-82d4-990269ab9708
2025-08-21 02:54:13 - INFO - Prompt token length: 2306
2025-08-21 02:54:23 - INFO - Tokens per second: 11.859020159952623, Peak GPU memory MB: 5350.375
2025-08-21 02:54:23 - INFO - [c40f2273-a9f5-4d96-82d4-990269ab9708] Inference time: 14.12 seconds, CPU usage: 2.0%, CPU core utilization: [2.0, 2.0, 1.9, 1.9]
2025-08-21 02:54:23 - INFO - [c40f2273-a9f5-4d96-82d4-990269ab9708] Cleaned up temporary frame directory: temp_videos/c40f2273-a9f5-4d96-82d4-990269ab9708
2025-08-21 02:54:23 - INFO - [1bbf302e-4b0b-4363-bddd-3fb826552587] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_002.mp4'
2025-08-21 02:54:23 - INFO - [1bbf302e-4b0b-4363-bddd-3fb826552587] Video saved to temporary file: temp_videos/1bbf302e-4b0b-4363-bddd-3fb826552587.mp4
2025-08-21 02:54:23 - INFO - [1bbf302e-4b0b-4363-bddd-3fb826552587] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:54:27 - INFO - [1bbf302e-4b0b-4363-bddd-3fb826552587] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:54:27 - INFO - [1bbf302e-4b0b-4363-bddd-3fb826552587] 30 frames saved to temp_videos/1bbf302e-4b0b-4363-bddd-3fb826552587
2025-08-21 02:54:27 - INFO - Prompt token length: 2306
2025-08-21 02:54:34 - INFO - Tokens per second: 12.033912631174916, Peak GPU memory MB: 5350.375
2025-08-21 02:54:34 - INFO - [1bbf302e-4b0b-4363-bddd-3fb826552587] Inference time: 10.49 seconds, CPU usage: 44.1%, CPU core utilization: [80.0, 27.5, 40.1, 29.0]
2025-08-21 02:54:34 - INFO - [1bbf302e-4b0b-4363-bddd-3fb826552587] Cleaned up temporary frame directory: temp_videos/1bbf302e-4b0b-4363-bddd-3fb826552587
2025-08-21 02:54:34 - INFO - [48b38709-fb9f-4c1d-9db6-279fea58e01f] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_003.mp4'
2025-08-21 02:54:34 - INFO - [48b38709-fb9f-4c1d-9db6-279fea58e01f] Video saved to temporary file: temp_videos/48b38709-fb9f-4c1d-9db6-279fea58e01f.mp4
2025-08-21 02:54:34 - INFO - [48b38709-fb9f-4c1d-9db6-279fea58e01f] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:54:37 - INFO - [48b38709-fb9f-4c1d-9db6-279fea58e01f] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:54:37 - INFO - [48b38709-fb9f-4c1d-9db6-279fea58e01f] 30 frames saved to temp_videos/48b38709-fb9f-4c1d-9db6-279fea58e01f
2025-08-21 02:54:37 - INFO - Prompt token length: 2306
2025-08-21 02:54:45 - INFO - Tokens per second: 11.980873759204092, Peak GPU memory MB: 5350.375
2025-08-21 02:54:45 - INFO - [48b38709-fb9f-4c1d-9db6-279fea58e01f] Inference time: 10.84 seconds, CPU usage: 43.7%, CPU core utilization: [49.1, 32.4, 65.3, 27.8]
2025-08-21 02:54:45 - INFO - [48b38709-fb9f-4c1d-9db6-279fea58e01f] Cleaned up temporary frame directory: temp_videos/48b38709-fb9f-4c1d-9db6-279fea58e01f
2025-08-21 02:54:45 - INFO - [218b6cb4-0c13-4223-b6be-fbc881774b17] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_004.mp4'
2025-08-21 02:54:45 - INFO - [218b6cb4-0c13-4223-b6be-fbc881774b17] Video saved to temporary file: temp_videos/218b6cb4-0c13-4223-b6be-fbc881774b17.mp4
2025-08-21 02:54:45 - INFO - [218b6cb4-0c13-4223-b6be-fbc881774b17] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:54:48 - INFO - [218b6cb4-0c13-4223-b6be-fbc881774b17] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:54:48 - INFO - [218b6cb4-0c13-4223-b6be-fbc881774b17] 30 frames saved to temp_videos/218b6cb4-0c13-4223-b6be-fbc881774b17
2025-08-21 02:54:48 - INFO - Prompt token length: 2306
2025-08-21 02:55:13 - INFO - Tokens per second: 11.894932301505968, Peak GPU memory MB: 5350.375
2025-08-21 02:55:13 - INFO - [218b6cb4-0c13-4223-b6be-fbc881774b17] Inference time: 27.98 seconds, CPU usage: 33.8%, CPU core utilization: [13.9, 45.9, 13.3, 61.9]
2025-08-21 02:55:13 - INFO - [218b6cb4-0c13-4223-b6be-fbc881774b17] Cleaned up temporary frame directory: temp_videos/218b6cb4-0c13-4223-b6be-fbc881774b17
2025-08-21 02:55:13 - INFO - [6550b43c-430e-4dee-8467-1a05b4c082cd] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_005.mp4'
2025-08-21 02:55:13 - INFO - [6550b43c-430e-4dee-8467-1a05b4c082cd] Video saved to temporary file: temp_videos/6550b43c-430e-4dee-8467-1a05b4c082cd.mp4
2025-08-21 02:55:13 - INFO - [6550b43c-430e-4dee-8467-1a05b4c082cd] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:55:16 - INFO - [6550b43c-430e-4dee-8467-1a05b4c082cd] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:55:16 - INFO - [6550b43c-430e-4dee-8467-1a05b4c082cd] 30 frames saved to temp_videos/6550b43c-430e-4dee-8467-1a05b4c082cd
2025-08-21 02:55:16 - INFO - Prompt token length: 2306
2025-08-21 02:55:25 - INFO - Tokens per second: 11.99842860278374, Peak GPU memory MB: 5350.375
2025-08-21 02:55:25 - INFO - [6550b43c-430e-4dee-8467-1a05b4c082cd] Inference time: 12.41 seconds, CPU usage: 40.7%, CPU core utilization: [34.0, 38.6, 64.3, 25.9]
2025-08-21 02:55:25 - INFO - [6550b43c-430e-4dee-8467-1a05b4c082cd] Cleaned up temporary frame directory: temp_videos/6550b43c-430e-4dee-8467-1a05b4c082cd
2025-08-21 02:55:25 - INFO - [172a602d-213b-41d6-b892-e7ca06e535bc] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_006.mp4'
2025-08-21 02:55:25 - INFO - [172a602d-213b-41d6-b892-e7ca06e535bc] Video saved to temporary file: temp_videos/172a602d-213b-41d6-b892-e7ca06e535bc.mp4
2025-08-21 02:55:25 - INFO - [172a602d-213b-41d6-b892-e7ca06e535bc] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:55:28 - INFO - [172a602d-213b-41d6-b892-e7ca06e535bc] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:55:28 - INFO - [172a602d-213b-41d6-b892-e7ca06e535bc] 30 frames saved to temp_videos/172a602d-213b-41d6-b892-e7ca06e535bc
2025-08-21 02:55:29 - INFO - Prompt token length: 2306
2025-08-21 02:55:40 - INFO - Tokens per second: 11.862422969421846, Peak GPU memory MB: 5350.375
2025-08-21 02:55:40 - INFO - [172a602d-213b-41d6-b892-e7ca06e535bc] Inference time: 15.04 seconds, CPU usage: 39.3%, CPU core utilization: [21.5, 43.6, 21.6, 70.6]
2025-08-21 02:55:40 - INFO - [172a602d-213b-41d6-b892-e7ca06e535bc] Cleaned up temporary frame directory: temp_videos/172a602d-213b-41d6-b892-e7ca06e535bc
2025-08-21 02:55:40 - INFO - [082b484d-e219-4cde-ac8e-8af5b8f380cd] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_007.mp4'
2025-08-21 02:55:40 - INFO - [082b484d-e219-4cde-ac8e-8af5b8f380cd] Video saved to temporary file: temp_videos/082b484d-e219-4cde-ac8e-8af5b8f380cd.mp4
2025-08-21 02:55:40 - INFO - [082b484d-e219-4cde-ac8e-8af5b8f380cd] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:55:43 - INFO - [082b484d-e219-4cde-ac8e-8af5b8f380cd] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:55:43 - INFO - [082b484d-e219-4cde-ac8e-8af5b8f380cd] 30 frames saved to temp_videos/082b484d-e219-4cde-ac8e-8af5b8f380cd
2025-08-21 02:55:44 - INFO - Prompt token length: 2306
2025-08-21 02:55:52 - INFO - Tokens per second: 12.007495276914103, Peak GPU memory MB: 5350.375
2025-08-21 02:55:52 - INFO - [082b484d-e219-4cde-ac8e-8af5b8f380cd] Inference time: 11.83 seconds, CPU usage: 42.8%, CPU core utilization: [60.5, 34.4, 49.7, 26.5]
2025-08-21 02:55:52 - INFO - [082b484d-e219-4cde-ac8e-8af5b8f380cd] Cleaned up temporary frame directory: temp_videos/082b484d-e219-4cde-ac8e-8af5b8f380cd
2025-08-21 02:55:52 - INFO - [d4aec199-0b7e-4058-b8ba-bdfbb7806fca] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_008.mp4'
2025-08-21 02:55:52 - INFO - [d4aec199-0b7e-4058-b8ba-bdfbb7806fca] Video saved to temporary file: temp_videos/d4aec199-0b7e-4058-b8ba-bdfbb7806fca.mp4
2025-08-21 02:55:52 - INFO - [d4aec199-0b7e-4058-b8ba-bdfbb7806fca] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:55:55 - INFO - [d4aec199-0b7e-4058-b8ba-bdfbb7806fca] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:55:55 - INFO - [d4aec199-0b7e-4058-b8ba-bdfbb7806fca] 30 frames saved to temp_videos/d4aec199-0b7e-4058-b8ba-bdfbb7806fca
2025-08-21 02:55:56 - INFO - Prompt token length: 2306
2025-08-21 02:56:04 - INFO - Tokens per second: 11.871294681994929, Peak GPU memory MB: 5350.375
2025-08-21 02:56:04 - INFO - [d4aec199-0b7e-4058-b8ba-bdfbb7806fca] Inference time: 12.13 seconds, CPU usage: 43.3%, CPU core utilization: [35.9, 32.5, 78.1, 26.8]
2025-08-21 02:56:04 - INFO - [d4aec199-0b7e-4058-b8ba-bdfbb7806fca] Cleaned up temporary frame directory: temp_videos/d4aec199-0b7e-4058-b8ba-bdfbb7806fca
2025-08-21 02:56:04 - INFO - [20eacc2f-2a33-4211-b488-f449c4bbc64d] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_009.mp4'
2025-08-21 02:56:04 - INFO - [20eacc2f-2a33-4211-b488-f449c4bbc64d] Video saved to temporary file: temp_videos/20eacc2f-2a33-4211-b488-f449c4bbc64d.mp4
2025-08-21 02:56:04 - INFO - [20eacc2f-2a33-4211-b488-f449c4bbc64d] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:56:07 - INFO - [20eacc2f-2a33-4211-b488-f449c4bbc64d] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:56:07 - INFO - [20eacc2f-2a33-4211-b488-f449c4bbc64d] 30 frames saved to temp_videos/20eacc2f-2a33-4211-b488-f449c4bbc64d
2025-08-21 02:56:08 - INFO - Prompt token length: 2306
2025-08-21 02:56:15 - INFO - Tokens per second: 11.63501242448262, Peak GPU memory MB: 5350.375
2025-08-21 02:56:15 - INFO - [20eacc2f-2a33-4211-b488-f449c4bbc64d] Inference time: 10.73 seconds, CPU usage: 46.3%, CPU core utilization: [38.3, 58.2, 31.7, 56.7]
2025-08-21 02:56:15 - INFO - [20eacc2f-2a33-4211-b488-f449c4bbc64d] Cleaned up temporary frame directory: temp_videos/20eacc2f-2a33-4211-b488-f449c4bbc64d
2025-08-21 02:56:15 - INFO - [7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_010.mp4'
2025-08-21 02:56:15 - INFO - [7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5] Video saved to temporary file: temp_videos/7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5.mp4
2025-08-21 02:56:15 - INFO - [7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:56:18 - INFO - [7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:56:18 - INFO - [7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5] 30 frames saved to temp_videos/7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5
2025-08-21 02:56:18 - INFO - Prompt token length: 2306
2025-08-21 02:56:31 - INFO - Tokens per second: 11.874488678953208, Peak GPU memory MB: 5350.375
2025-08-21 02:56:31 - INFO - [7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5] Inference time: 16.08 seconds, CPU usage: 37.8%, CPU core utilization: [19.6, 68.7, 18.4, 44.3]
2025-08-21 02:56:31 - INFO - [7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5] Cleaned up temporary frame directory: temp_videos/7bd61912-f2d8-49f3-a1d2-d25a5bb09ff5
2025-08-21 02:56:31 - INFO - [305ccf60-14df-466d-8565-f04265430ba1] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_011.mp4'
2025-08-21 02:56:31 - INFO - [305ccf60-14df-466d-8565-f04265430ba1] Video saved to temporary file: temp_videos/305ccf60-14df-466d-8565-f04265430ba1.mp4
2025-08-21 02:56:31 - INFO - [305ccf60-14df-466d-8565-f04265430ba1] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:56:34 - INFO - [305ccf60-14df-466d-8565-f04265430ba1] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:56:34 - INFO - [305ccf60-14df-466d-8565-f04265430ba1] 30 frames saved to temp_videos/305ccf60-14df-466d-8565-f04265430ba1
2025-08-21 02:56:35 - INFO - Prompt token length: 2306
2025-08-21 02:56:44 - INFO - Tokens per second: 11.829041430743297, Peak GPU memory MB: 5350.375
2025-08-21 02:56:44 - INFO - [305ccf60-14df-466d-8565-f04265430ba1] Inference time: 12.93 seconds, CPU usage: 42.0%, CPU core utilization: [28.8, 42.5, 25.9, 70.9]
2025-08-21 02:56:44 - INFO - [305ccf60-14df-466d-8565-f04265430ba1] Cleaned up temporary frame directory: temp_videos/305ccf60-14df-466d-8565-f04265430ba1
2025-08-21 02:56:44 - INFO - [659dc8e0-c40a-432f-887e-c9cdeefc17a4] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_012.mp4'
2025-08-21 02:56:44 - INFO - [659dc8e0-c40a-432f-887e-c9cdeefc17a4] Video saved to temporary file: temp_videos/659dc8e0-c40a-432f-887e-c9cdeefc17a4.mp4
2025-08-21 02:56:44 - INFO - [659dc8e0-c40a-432f-887e-c9cdeefc17a4] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:56:47 - INFO - [659dc8e0-c40a-432f-887e-c9cdeefc17a4] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:56:47 - INFO - [659dc8e0-c40a-432f-887e-c9cdeefc17a4] 30 frames saved to temp_videos/659dc8e0-c40a-432f-887e-c9cdeefc17a4
2025-08-21 02:56:48 - INFO - Prompt token length: 2306
2025-08-21 02:56:58 - INFO - Tokens per second: 11.928726359703456, Peak GPU memory MB: 5350.375
2025-08-21 02:56:58 - INFO - [659dc8e0-c40a-432f-887e-c9cdeefc17a4] Inference time: 13.75 seconds, CPU usage: 39.8%, CPU core utilization: [31.6, 62.3, 41.3, 23.8]
2025-08-21 02:56:58 - INFO - [659dc8e0-c40a-432f-887e-c9cdeefc17a4] Cleaned up temporary frame directory: temp_videos/659dc8e0-c40a-432f-887e-c9cdeefc17a4
2025-08-21 02:56:58 - INFO - [05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_013.mp4'
2025-08-21 02:56:58 - INFO - [05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989] Video saved to temporary file: temp_videos/05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989.mp4
2025-08-21 02:56:58 - INFO - [05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:57:01 - INFO - [05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:57:01 - INFO - [05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989] 30 frames saved to temp_videos/05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989
2025-08-21 02:57:01 - INFO - Prompt token length: 2306
2025-08-21 02:57:07 - INFO - Tokens per second: 12.014726651428436, Peak GPU memory MB: 5350.375
2025-08-21 02:57:07 - INFO - [05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989] Inference time: 9.37 seconds, CPU usage: 43.8%, CPU core utilization: [29.4, 29.6, 88.1, 27.9]
2025-08-21 02:57:07 - INFO - [05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989] Cleaned up temporary frame directory: temp_videos/05a4c1b9-d6d6-4e4e-a0f2-f58a0664c989
2025-08-21 02:57:07 - INFO - [0f5076d3-96af-4d28-be73-0db23c76eaf4] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_014.mp4'
2025-08-21 02:57:07 - INFO - [0f5076d3-96af-4d28-be73-0db23c76eaf4] Video saved to temporary file: temp_videos/0f5076d3-96af-4d28-be73-0db23c76eaf4.mp4
2025-08-21 02:57:07 - INFO - [0f5076d3-96af-4d28-be73-0db23c76eaf4] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:57:10 - INFO - [0f5076d3-96af-4d28-be73-0db23c76eaf4] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:57:10 - INFO - [0f5076d3-96af-4d28-be73-0db23c76eaf4] 30 frames saved to temp_videos/0f5076d3-96af-4d28-be73-0db23c76eaf4
2025-08-21 02:57:11 - INFO - Prompt token length: 2306
2025-08-21 02:57:19 - INFO - Tokens per second: 11.861972979079045, Peak GPU memory MB: 5350.375
2025-08-21 02:57:19 - INFO - [0f5076d3-96af-4d28-be73-0db23c76eaf4] Inference time: 11.61 seconds, CPU usage: 41.6%, CPU core utilization: [42.9, 26.4, 27.0, 69.9]
2025-08-21 02:57:19 - INFO - [0f5076d3-96af-4d28-be73-0db23c76eaf4] Cleaned up temporary frame directory: temp_videos/0f5076d3-96af-4d28-be73-0db23c76eaf4
2025-08-21 02:57:19 - INFO - [a60e4adc-5a10-496e-8dba-e95fa8204801] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_015.mp4'
2025-08-21 02:57:19 - INFO - [a60e4adc-5a10-496e-8dba-e95fa8204801] Video saved to temporary file: temp_videos/a60e4adc-5a10-496e-8dba-e95fa8204801.mp4
2025-08-21 02:57:19 - INFO - [a60e4adc-5a10-496e-8dba-e95fa8204801] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:57:22 - INFO - [a60e4adc-5a10-496e-8dba-e95fa8204801] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:57:22 - INFO - [a60e4adc-5a10-496e-8dba-e95fa8204801] 30 frames saved to temp_videos/a60e4adc-5a10-496e-8dba-e95fa8204801
2025-08-21 02:57:22 - INFO - Prompt token length: 2306
2025-08-21 02:57:31 - INFO - Tokens per second: 12.034885208983422, Peak GPU memory MB: 5350.375
2025-08-21 02:57:31 - INFO - [a60e4adc-5a10-496e-8dba-e95fa8204801] Inference time: 12.68 seconds, CPU usage: 39.5%, CPU core utilization: [59.8, 22.7, 53.7, 22.1]
2025-08-21 02:57:31 - INFO - [a60e4adc-5a10-496e-8dba-e95fa8204801] Cleaned up temporary frame directory: temp_videos/a60e4adc-5a10-496e-8dba-e95fa8204801
2025-08-21 02:57:31 - INFO - [262c15ae-e353-4d00-b508-c4d77d75300a] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/new/Clips_60s/video_part_016.mp4'
2025-08-21 02:57:31 - INFO - [262c15ae-e353-4d00-b508-c4d77d75300a] Video saved to temporary file: temp_videos/262c15ae-e353-4d00-b508-c4d77d75300a.mp4
2025-08-21 02:57:31 - INFO - [262c15ae-e353-4d00-b508-c4d77d75300a] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 02:57:35 - INFO - [262c15ae-e353-4d00-b508-c4d77d75300a] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 02:57:35 - INFO - [262c15ae-e353-4d00-b508-c4d77d75300a] 30 frames saved to temp_videos/262c15ae-e353-4d00-b508-c4d77d75300a
2025-08-21 02:57:35 - INFO - Prompt token length: 2306
2025-08-21 02:57:59 - INFO - Tokens per second: 12.052444962168167, Peak GPU memory MB: 5350.375
2025-08-21 02:57:59 - INFO - [262c15ae-e353-4d00-b508-c4d77d75300a] Inference time: 27.71 seconds, CPU usage: 33.2%, CPU core utilization: [31.4, 17.1, 70.7, 13.4]
2025-08-21 02:57:59 - INFO - [262c15ae-e353-4d00-b508-c4d77d75300a] Cleaned up temporary frame directory: temp_videos/262c15ae-e353-4d00-b508-c4d77d75300a