训练报告: 01180043_e5_w60

训练环境与参数(train)
变量名 含义
time_start 开始时间 2026-01-17 23:42:40
platform 平台 Linux-5.10.134-18.0.6.lifsea8.x86_64-x86_64-with-glibc2.35
processor CPU架构 x86_64
python_version Python版本 3.10.6
torch_version Torch版本 2.0.0+cu118
cuda_available CUDA可用 False
cuda_device_count GPU数量 0
cuda_device_name_0 GPU0名称 NA
device_used 使用设备 cpu
epochs 训练轮数 5
train_seconds 训练时长 1小时0分钟15秒
threshold_default 默认阈值 0.5
pos_weight_used pos_weight 60.0
lr_init 初始学习率 1e-3
optimizer 优化器 Adam
report_folder_name 报告名 01180043_e5_w60
测试环境与参数(cut_test)
变量名 含义
time 记录时间 2026-01-18 00:44:53
platform 平台 Linux-5.10.134-18.0.6.lifsea8.x86_64-x86_64-with-glibc2.35
processor CPU架构 x86_64
python_version Python版本 3.10.6
torch_version Torch版本 2.0.0+cu118
cuda_available CUDA可用 False
device_used 使用设备 cpu
threshold_default 默认阈值 0.5
batch_size batch_size 64
report_folder_name 报告名 01180043_e5_w60
训练集曲线(epoch_metrics)
Loss
Precision / Recall / F1
AP(PR-AUC) / AUC(ROC)
Accuracy(若无则为空)
正类预测率 pos_pred_rate
学习率 lr(若无则为空)
建议阅读顺序:先看 pos_pred_rate 判断“乱报切点”程度,再看 PRF 是否平衡,最后结合 AP/AUC 评估整体质量。
训练集最终指标(train / val 汇总)
阈值 0.5
精确率 0.139344262295082
召回率 0.9444444444444444
F1 0.2428571428571428
准确率 0.8827433628318584
AP(PR-AUC) 0.2618542940229625
AUC(ROC) 0.9322799097065463
TP 17
FP 105
TN 781
FN 1
正类预测率 0.1349557522123894
测试集最终指标(cut_test)
阈值 0.5
精确率 0.09740259740259741
召回率 0.75
F1 0.1724137931034483
准确率 0.8937269372693727
AP(PR-AUC) 0.2215022460351199
AUC(ROC) 0.932434456928839
TP 15
FP 139
TN 1196
FN 5
正类预测率 0.1136531365313653
测试集帧轴可视化(每个视频:TP / FP / FN 的帧位置)
TP(预测=切点 且 GT=切点) FP(预测=切点 但 GT=非切点) FN(GT=切点 但 预测=非切点) 提示:鼠标悬停点可看帧号
V001.mp4
total_frames: 43  |  TP 1 FP 8 FN 0
GT cuts: 1 Pred cuts: 9
轴左侧≈frame 0 轴右侧≈frame 42
V002.mp4
total_frames: 161  |  TP 1 FP 12 FN 1
GT cuts: 2 Pred cuts: 13
轴左侧≈frame 0 轴右侧≈frame 160
V003.mp4
total_frames: 122  |  TP 5 FP 35 FN 0
GT cuts: 5 Pred cuts: 40
轴左侧≈frame 0 轴右侧≈frame 121
V004.mp4
total_frames: 700  |  TP 2 FP 0 FN 4
GT cuts: 6 Pred cuts: 2
轴左侧≈frame 0 轴右侧≈frame 699
V005.mp4
total_frames: 334  |  TP 6 FP 84 FN 0
GT cuts: 6 Pred cuts: 90
轴左侧≈frame 0 轴右侧≈frame 333
测试集数据明细(Excel 其它 Sheets 预览)
dataset_summary(显示前 11 行 / 共 11 行)
item value
num_videos 5
num_pairs 1355
num_cuts 20
num_non_cuts 1335
pos_ratio 0.01476014760147601
per_video_frame_stats
min_frames 43
max_frames 700
mean_frames 272
median_frames 161
per_video(显示前 5 行 / 共 5 行)
vid vid_idx total_frames gt_cut_count pred_cut_count tp fp fn gt_cuts pred_cuts
V001.mp4 0 43 1 9 1 8 0 26 18,19,20,21,22,23,24,25,26
V002.mp4 1 161 2 13 1 12 1 86,131 84,86,96,140,151,152,153,154,155,156,157,158,159
V003.mp4 2 122 5 40 5 35 0 18,35,56,72,105 18,21,23,25,29,34,35,36,37,38,39,40,41,42,43,44,45,47,48,49,50,51,52,53,54,55,56,61,62,63,72,84,86,90,92,95,97,99,101,105
V004.mp4 3 700 6 2 2 0 4 46,419,457,504,600,643 46,643
V005.mp4 4 334 6 90 6 84 0 20,69,134,169,238,263 12,13,14,15,16,17,18,19,20,21,29,30,31,32,34,35,36,37,44,45,46,55,57,58,59,60,61,62,69,70,78,79,80,81,83,84,97,99,100,113,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,143,155,156,167,169,171,173,174,176,178,183,184,186,198,199,201,213,219,221,223,226,238,263,265,269,272,274,276,277,280,281,282,284,329,330
classification_report(显示前 5 行 / 共 5 行)
text
precision recall f1-score support
Non-cut 0.9958 0.8959 0.9432 1335
Cut 0.0974 0.7500 0.1724 20
accuracy 0.8937 1355
下载