diff --git a/.gitignore b/.gitignore index 82f9275..f7d3395 100644 --- a/.gitignore +++ b/.gitignore @@ -160,3 +160,7 @@ cython_debug/ # and can be added to the global gitignore or merged into this file. For a more nuclear # option (not recommended) you can uncomment the following to ignore the entire idea folder. #.idea/ +dataset/train/images +dataset/train/labels +app_logs.txt +.aider* diff --git a/README.md b/README.md index 53131d3..f98ac50 100644 --- a/README.md +++ b/README.md @@ -53,6 +53,7 @@ SearchVision is designed to provide a seamless experience for users to perform i ## Contributing We welcome contributions! If you're interested in contributing to SearchVision, please read our [CONTRIBUTING.md](CONTRIBUTING.md) file for guidelines on how to get started. +You can also give feedback [here](https://feedbackflux.vercel.app/brandonshen123@gmail.com/b/SearchVision) ## License diff --git a/dataset/data.yaml b/dataset/data.yaml new file mode 100644 index 0000000..4ff8642 --- /dev/null +++ b/dataset/data.yaml @@ -0,0 +1,5 @@ +path: C:\Users\Brandon Shen\Documents\SearchVision\dataset +train: C:\Users\Brandon Shen\Documents\SearchVision\dataset\train\images +val: C:\Users\Brandon Shen\Documents\SearchVision\dataset\train\images +names: + 0: basketball diff --git a/dataset/train/labels.cache b/dataset/train/labels.cache new file mode 100644 index 0000000..f835e11 Binary files /dev/null and b/dataset/train/labels.cache differ diff --git a/requirements.txt b/requirements.txt index 2be2fb1..8cbd9f3 100644 --- a/requirements.txt +++ b/requirements.txt @@ -4,3 +4,9 @@ requests beautifulsoup4 ultralytics python-multipart +torch +torchvision +numpy +scikit-learn +pillow +python-dotenv diff --git a/runs/detect/train/args.yaml b/runs/detect/train/args.yaml new file mode 100644 index 0000000..315f5b0 --- /dev/null +++ b/runs/detect/train/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train diff --git a/runs/detect/train10/args.yaml b/runs/detect/train10/args.yaml new file mode 100644 index 0000000..e46c174 --- /dev/null +++ b/runs/detect/train10/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train10 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train10 diff --git a/runs/detect/train11/args.yaml b/runs/detect/train11/args.yaml new file mode 100644 index 0000000..51a4b2a --- /dev/null +++ b/runs/detect/train11/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train11 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train11 diff --git a/runs/detect/train12/args.yaml b/runs/detect/train12/args.yaml new file mode 100644 index 0000000..7bfc360 --- /dev/null +++ b/runs/detect/train12/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train12 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train12 diff --git a/runs/detect/train13/args.yaml b/runs/detect/train13/args.yaml new file mode 100644 index 0000000..6322fa5 --- /dev/null +++ b/runs/detect/train13/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train13 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train13 diff --git a/runs/detect/train14/args.yaml b/runs/detect/train14/args.yaml new file mode 100644 index 0000000..9c1277a --- /dev/null +++ b/runs/detect/train14/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train14 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train14 diff --git a/runs/detect/train15/args.yaml b/runs/detect/train15/args.yaml new file mode 100644 index 0000000..0f2ddb5 --- /dev/null +++ b/runs/detect/train15/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train15 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train15 diff --git a/runs/detect/train16/args.yaml b/runs/detect/train16/args.yaml new file mode 100644 index 0000000..de81f9f --- /dev/null +++ b/runs/detect/train16/args.yaml @@ -0,0 +1,107 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset\data.yaml +epochs: 25 +time: null +patience: 10 +batch: 8 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: cpu +workers: 8 +project: null +name: train16 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: true +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +copy_paste_mode: flip +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: C:\Users\Brandon Shen\Documents\SearchVision\runs\detect\train16 diff --git a/runs/detect/train17/F1_curve.png b/runs/detect/train17/F1_curve.png new file mode 100644 index 0000000..5158e5e Binary files /dev/null and b/runs/detect/train17/F1_curve.png differ diff --git a/runs/detect/train17/PR_curve.png b/runs/detect/train17/PR_curve.png new file mode 100644 index 0000000..33f9935 Binary files /dev/null and b/runs/detect/train17/PR_curve.png differ diff --git a/runs/detect/train17/P_curve.png b/runs/detect/train17/P_curve.png new file mode 100644 index 0000000..525f554 Binary files /dev/null and b/runs/detect/train17/P_curve.png differ diff --git a/runs/detect/train17/R_curve.png b/runs/detect/train17/R_curve.png new file mode 100644 index 0000000..359d844 Binary files /dev/null and b/runs/detect/train17/R_curve.png differ diff --git a/runs/detect/train17/args.yaml b/runs/detect/train17/args.yaml new file mode 100644 index 0000000..55ff50a --- /dev/null +++ b/runs/detect/train17/args.yaml @@ -0,0 +1,107 @@ +task: detect +mode: train +model: yolov8n.pt +data: C:\Users\Brandon Shen\Documents\SearchVision\dataset\data.yaml +epochs: 25 +time: null +patience: 10 +batch: 8 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: cpu +workers: 8 +project: null +name: train17 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: true +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +copy_paste_mode: flip +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: C:\Users\Brandon Shen\Documents\SearchVision\runs\detect\train17 diff --git a/runs/detect/train17/confusion_matrix.png b/runs/detect/train17/confusion_matrix.png new file mode 100644 index 0000000..6ff56da Binary files /dev/null and b/runs/detect/train17/confusion_matrix.png differ diff --git a/runs/detect/train17/confusion_matrix_normalized.png b/runs/detect/train17/confusion_matrix_normalized.png new file mode 100644 index 0000000..04e394f Binary files /dev/null and b/runs/detect/train17/confusion_matrix_normalized.png differ diff --git a/runs/detect/train17/events.out.tfevents.1740701719.PW4638167770234.55244.0 b/runs/detect/train17/events.out.tfevents.1740701719.PW4638167770234.55244.0 new file mode 100644 index 0000000..9a5bbf6 Binary files /dev/null and b/runs/detect/train17/events.out.tfevents.1740701719.PW4638167770234.55244.0 differ diff --git a/runs/detect/train17/labels.jpg b/runs/detect/train17/labels.jpg new file mode 100644 index 0000000..5baaba5 Binary files /dev/null and b/runs/detect/train17/labels.jpg differ diff --git a/runs/detect/train17/labels_correlogram.jpg b/runs/detect/train17/labels_correlogram.jpg new file mode 100644 index 0000000..7351913 Binary files /dev/null and b/runs/detect/train17/labels_correlogram.jpg differ diff --git a/runs/detect/train17/results.csv b/runs/detect/train17/results.csv new file mode 100644 index 0000000..e4b1c96 --- /dev/null +++ b/runs/detect/train17/results.csv @@ -0,0 +1,26 @@ +epoch,time,train/box_loss,train/cls_loss,train/dfl_loss,metrics/precision(B),metrics/recall(B),metrics/mAP50(B),metrics/mAP50-95(B),val/box_loss,val/cls_loss,val/dfl_loss,lr/pg0,lr/pg1,lr/pg2 +1,21.0361,0.91108,2.80163,1.17384,0.00856,0.9625,0.34137,0.24818,0.39739,2.98326,0.90614,6e-05,6e-05,6e-05 +2,33.0195,0.98657,2.64668,1.13301,0.00867,0.975,0.4323,0.35792,0.46403,2.96352,0.91283,0.000134456,0.000134456,0.000134456 +3,43.9278,0.84994,2.10175,1.12605,0.00867,0.975,0.4622,0.37828,0.53374,2.91079,0.92489,0.000202576,0.000202576,0.000202576 +4,55.7088,0.81153,1.78668,1.07674,0.00878,0.9875,0.51515,0.40305,0.61849,2.79785,0.95287,0.00026436,0.00026436,0.00026436 +5,67.7567,0.90498,1.49133,1.07601,0.94146,0.2011,0.5381,0.396,0.70973,2.6416,0.97848,0.000319808,0.000319808,0.000319808 +6,80.1714,0.936,1.94005,1.25179,0.95326,0.25506,0.55077,0.43206,0.69786,2.55737,0.96863,0.00036892,0.00036892,0.00036892 +7,91.0975,0.86754,1.52,1.12672,0.96576,0.3526,0.58323,0.48808,0.66286,2.46892,0.97026,0.000411696,0.000411696,0.000411696 +8,101.657,0.80064,1.6376,1.12076,0.96167,0.31364,0.57858,0.47984,0.66931,2.4826,0.97193,0.000448136,0.000448136,0.000448136 +9,112.643,0.63558,1.39311,1.05431,1,0.25529,0.57879,0.47119,0.68391,2.58967,0.97364,0.00047824,0.00047824,0.00047824 +10,123.155,0.78626,1.54893,1.13025,1,0.21903,0.55938,0.45304,0.69666,2.63499,0.97718,0.000502008,0.000502008,0.000502008 +11,133.726,0.7854,1.34338,1.08913,1,0.36744,0.56225,0.47962,0.70092,2.51946,0.9834,0.00051944,0.00051944,0.00051944 +12,144.197,0.79021,1.4295,1.09774,0.92005,0.4316,0.54507,0.46676,0.67422,2.34246,0.99195,0.000530536,0.000530536,0.000530536 +13,154.671,0.72066,1.33911,1.04883,0.79628,0.43977,0.56778,0.47914,0.65884,2.23407,0.98003,0.000535296,0.000535296,0.000535296 +14,165.229,0.69716,1.22407,1.04598,0.86489,0.56014,0.68678,0.6031,0.64374,2.10599,0.97019,0.00053372,0.00053372,0.00053372 +15,176.004,0.84211,1.1759,1.03366,0.84858,0.63049,0.79059,0.69692,0.62518,1.94564,0.99782,0.000525808,0.000525808,0.000525808 +16,186.86,0.5805,1.83191,0.97853,0.84858,0.63049,0.79059,0.69692,0.62518,1.94564,0.99782,0.00051156,0.00051156,0.00051156 +17,197.522,0.69049,1.67263,0.98821,0.89012,0.675,0.8282,0.71474,0.63029,1.88828,1.00195,0.000490976,0.000490976,0.000490976 +18,210.531,0.62329,1.87898,0.98679,0.88448,0.675,0.83214,0.7142,0.65848,1.86514,1.02756,0.000464056,0.000464056,0.000464056 +19,221.325,0.65359,1.47793,0.95454,0.88448,0.675,0.83214,0.7142,0.65848,1.86514,1.02756,0.0004308,0.0004308,0.0004308 +20,231.667,0.65471,1.68673,1.0228,0.96476,0.68446,0.82951,0.70684,0.66719,1.86883,1.04404,0.000391208,0.000391208,0.000391208 +21,242.31,0.644,1.74126,1.06733,0.9188,0.70729,0.84282,0.72777,0.65072,1.83517,1.00727,0.00034528,0.00034528,0.00034528 +22,252.655,0.60214,1.57435,0.9613,0.9188,0.70729,0.84282,0.72777,0.65072,1.83517,1.00727,0.000293016,0.000293016,0.000293016 +23,263.138,0.69024,1.60948,1.02949,0.908,0.74026,0.87572,0.75413,0.62707,1.7348,0.97154,0.000234416,0.000234416,0.000234416 +24,273.372,0.5577,1.43103,0.922,0.908,0.74026,0.87572,0.75413,0.62707,1.7348,0.97154,0.00016948,0.00016948,0.00016948 +25,284.072,0.79471,1.6108,1.10699,0.91309,0.78797,0.88191,0.75918,0.61545,1.61585,0.94993,9.8208e-05,9.8208e-05,9.8208e-05 diff --git a/runs/detect/train17/results.png b/runs/detect/train17/results.png new file mode 100644 index 0000000..3e885f3 Binary files /dev/null and b/runs/detect/train17/results.png differ diff --git a/runs/detect/train17/train_batch0.jpg b/runs/detect/train17/train_batch0.jpg new file mode 100644 index 0000000..48009b5 Binary files /dev/null and b/runs/detect/train17/train_batch0.jpg differ diff --git a/runs/detect/train17/train_batch1.jpg b/runs/detect/train17/train_batch1.jpg new file mode 100644 index 0000000..990b246 Binary files /dev/null and b/runs/detect/train17/train_batch1.jpg differ diff --git a/runs/detect/train17/train_batch2.jpg b/runs/detect/train17/train_batch2.jpg new file mode 100644 index 0000000..b354a9d Binary files /dev/null and b/runs/detect/train17/train_batch2.jpg differ diff --git a/runs/detect/train17/train_batch60.jpg b/runs/detect/train17/train_batch60.jpg new file mode 100644 index 0000000..1dea173 Binary files /dev/null and b/runs/detect/train17/train_batch60.jpg differ diff --git a/runs/detect/train17/train_batch61.jpg b/runs/detect/train17/train_batch61.jpg new file mode 100644 index 0000000..608396a Binary files /dev/null and b/runs/detect/train17/train_batch61.jpg differ diff --git a/runs/detect/train17/train_batch62.jpg b/runs/detect/train17/train_batch62.jpg new file mode 100644 index 0000000..8b35e40 Binary files /dev/null and b/runs/detect/train17/train_batch62.jpg differ diff --git a/runs/detect/train17/val_batch0_labels.jpg b/runs/detect/train17/val_batch0_labels.jpg new file mode 100644 index 0000000..524c0b5 Binary files /dev/null and b/runs/detect/train17/val_batch0_labels.jpg differ diff --git a/runs/detect/train17/val_batch0_pred.jpg b/runs/detect/train17/val_batch0_pred.jpg new file mode 100644 index 0000000..f870c41 Binary files /dev/null and b/runs/detect/train17/val_batch0_pred.jpg differ diff --git a/runs/detect/train17/val_batch1_labels.jpg b/runs/detect/train17/val_batch1_labels.jpg new file mode 100644 index 0000000..d6ccc49 Binary files /dev/null and b/runs/detect/train17/val_batch1_labels.jpg differ diff --git a/runs/detect/train17/val_batch1_pred.jpg b/runs/detect/train17/val_batch1_pred.jpg new file mode 100644 index 0000000..19da4b7 Binary files /dev/null and b/runs/detect/train17/val_batch1_pred.jpg differ diff --git a/runs/detect/train17/weights/best.pt b/runs/detect/train17/weights/best.pt new file mode 100644 index 0000000..db68787 Binary files /dev/null and b/runs/detect/train17/weights/best.pt differ diff --git a/runs/detect/train17/weights/last.pt b/runs/detect/train17/weights/last.pt new file mode 100644 index 0000000..b8938b7 Binary files /dev/null and b/runs/detect/train17/weights/last.pt differ diff --git a/runs/detect/train2/args.yaml b/runs/detect/train2/args.yaml new file mode 100644 index 0000000..9716e45 --- /dev/null +++ b/runs/detect/train2/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train2 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train2 diff --git a/runs/detect/train3/args.yaml b/runs/detect/train3/args.yaml new file mode 100644 index 0000000..fd7e27e --- /dev/null +++ b/runs/detect/train3/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train3 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train3 diff --git a/runs/detect/train4/args.yaml b/runs/detect/train4/args.yaml new file mode 100644 index 0000000..1871847 --- /dev/null +++ b/runs/detect/train4/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train4 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train4 diff --git a/runs/detect/train5/args.yaml b/runs/detect/train5/args.yaml new file mode 100644 index 0000000..6278269 --- /dev/null +++ b/runs/detect/train5/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train5 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train5 diff --git a/runs/detect/train6/args.yaml b/runs/detect/train6/args.yaml new file mode 100644 index 0000000..e19c16b --- /dev/null +++ b/runs/detect/train6/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train6 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train6 diff --git a/runs/detect/train7/args.yaml b/runs/detect/train7/args.yaml new file mode 100644 index 0000000..bec1959 --- /dev/null +++ b/runs/detect/train7/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train7 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train7 diff --git a/runs/detect/train8/args.yaml b/runs/detect/train8/args.yaml new file mode 100644 index 0000000..032cb2b --- /dev/null +++ b/runs/detect/train8/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train8 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train8 diff --git a/runs/detect/train9/args.yaml b/runs/detect/train9/args.yaml new file mode 100644 index 0000000..00bb6a7 --- /dev/null +++ b/runs/detect/train9/args.yaml @@ -0,0 +1,106 @@ +task: detect +mode: train +model: yolov8n.pt +data: dataset/data.yaml +epochs: 50 +time: null +patience: 100 +batch: 16 +imgsz: 640 +save: true +save_period: -1 +cache: false +device: null +workers: 8 +project: null +name: train9 +exist_ok: false +pretrained: true +optimizer: auto +verbose: true +seed: 0 +deterministic: true +single_cls: false +rect: false +cos_lr: false +close_mosaic: 10 +resume: false +amp: true +fraction: 1.0 +profile: false +freeze: null +multi_scale: false +overlap_mask: true +mask_ratio: 4 +dropout: 0.0 +val: true +split: val +save_json: false +save_hybrid: false +conf: null +iou: 0.7 +max_det: 300 +half: false +dnn: false +plots: true +source: null +vid_stride: 1 +stream_buffer: false +visualize: false +augment: false +agnostic_nms: false +classes: null +retina_masks: false +embed: null +show: false +save_frames: false +save_txt: false +save_conf: false +save_crop: false +show_labels: true +show_conf: true +show_boxes: true +line_width: null +format: torchscript +keras: false +optimize: false +int8: false +dynamic: false +simplify: false +opset: null +workspace: 4 +nms: false +lr0: 0.01 +lrf: 0.01 +momentum: 0.937 +weight_decay: 0.0005 +warmup_epochs: 3.0 +warmup_momentum: 0.8 +warmup_bias_lr: 0.1 +box: 7.5 +cls: 0.5 +dfl: 1.5 +pose: 12.0 +kobj: 1.0 +label_smoothing: 0.0 +nbs: 64 +hsv_h: 0.015 +hsv_s: 0.7 +hsv_v: 0.4 +degrees: 0.0 +translate: 0.1 +scale: 0.5 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +bgr: 0.0 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 +auto_augment: randaugment +erasing: 0.4 +crop_fraction: 1.0 +cfg: null +tracker: botsort.yaml +save_dir: runs\detect\train9 diff --git a/src/auto_annotate_images.py b/src/auto_annotate_images.py new file mode 100644 index 0000000..204cbb3 --- /dev/null +++ b/src/auto_annotate_images.py @@ -0,0 +1,78 @@ +from ultralytics import YOLO +import os +from PIL import Image +import logging + +logger = logging.getLogger(__name__) + +def auto_annotate_images(image_folder, labels_folder): + """ + Auto-annotate images using YOLOv8 and save annotations in YOLO format. + + Args: + image_folder: Directory containing images + labels_folder: Directory where annotation files will be saved + """ + # Load the pre-trained YOLOv8 model + model = YOLO('yolov8n.pt') + + os.makedirs(labels_folder, exist_ok=True) + + # Track statistics + processed_count = 0 + annotated_count = 0 + error_count = 0 + + # Loop through images in the specified folder + for image_file in os.listdir(image_folder): + if image_file.endswith(('.jpg', '.jpeg', '.png')): + image_path = os.path.join(image_folder, image_file) + + # Check if the image can be opened and processed + try: + # Get image dimensions + with Image.open(image_path) as img: + img_width, img_height = img.size + + # Run detection + results = model(image_path) + processed_count += 1 + + # Create annotation filename + label_filename = os.path.splitext(image_file)[0] + ".txt" + label_path = os.path.join(labels_folder, label_filename) + + # Process the results + for result in results: + if result.boxes is not None and len(result.boxes) > 0: + # Extract bounding boxes + boxes = result.boxes.xyxy.cpu().numpy() + + with open(label_path, 'w') as f: + for box in boxes: + if len(box) >= 4: + x_min, y_min, x_max, y_max = box[:4] + + # Calculate YOLO format (normalized) + x_center = ((x_min + x_max) / 2) / img_width + y_center = ((y_min + y_max) / 2) / img_height + width = (x_max - x_min) / img_width + height = (y_max - y_min) / img_height + + # Class 0 for all objects (single class) + f.write(f"0 {x_center:.6f} {y_center:.6f} {width:.6f} {height:.6f}\n") + annotated_count += 1 + + logger.info(f"Saved YOLO annotations for {image_file}") + else: + logger.warning(f"No objects detected in {image_file}") + + except Exception as e: + logger.error(f"Error processing image {image_path}: {e}") + error_count += 1 + continue + + logger.info(f"Auto-annotation complete: processed {processed_count} images, " + f"created {annotated_count} annotations, {error_count} errors") + + return processed_count, annotated_count, error_count \ No newline at end of file diff --git a/src/create_data_yaml.py b/src/create_data_yaml.py new file mode 100644 index 0000000..b1ead0f --- /dev/null +++ b/src/create_data_yaml.py @@ -0,0 +1,36 @@ +import os +import yaml + +def create_data_yaml(annotations_path, object_name="object"): + """ + Create a YAML configuration file for YOLOv8 training. + + Args: + annotations_path: Path to annotations (used for determining dataset path) + object_name: Name of the object class + + Returns: + Path to the created YAML file + """ + # Use absolute paths + base_dir = os.path.dirname(os.path.dirname(os.path.abspath(annotations_path))) + train_images_path = os.path.join(base_dir, "train", "images") + train_labels_path = os.path.join(base_dir, "train", "labels") + + # Dataset structure expected by YOLOv8 with absolute paths + data = { + 'path': base_dir, # Absolute base path + 'train': train_images_path, # Absolute train images path + 'val': train_images_path, # Using same images for validation + 'names': { + 0: object_name # Single class detection + } + } + + # Create the YAML file + yaml_path = os.path.join(base_dir, 'data.yaml') + with open(yaml_path, 'w') as f: + yaml.dump(data, f, sort_keys=False, default_flow_style=False) + + print(f"Created data.yaml at {yaml_path}") + return yaml_path \ No newline at end of file diff --git a/src/download_images.py b/src/download_images.py new file mode 100644 index 0000000..398dc20 --- /dev/null +++ b/src/download_images.py @@ -0,0 +1,44 @@ +# src/download_images.py + +import requests +import os + + +def download_images(image_urls, download_path="dataset/train/images"): + """ + Downloads images from a list of URLs and saves them to the specified directory. + + :param image_urls: List of image URLs to download. + :param download_path: Directory to save downloaded images. + :return: List of file paths for successfully downloaded images. + """ + print("Starting image download...") # Debugging statement + + # Ensure the download directory exists + if not os.path.exists(download_path): + os.makedirs(download_path) + + # List to hold paths of successfully downloaded images + downloaded_paths = [] + + # Iterate over the image URLs and download each image + for idx, url in enumerate(image_urls): + print(f"Attempting to download: {url}") # Debugging statement + try: + response = requests.get(url) + if response.status_code == 200: + file_path = os.path.join(download_path, f"image_{idx}.jpg") + with open(file_path, "wb") as f: + f.write(response.content) + print(f"Downloaded: {file_path}") + downloaded_paths.append(file_path) # Add path to list + else: + print(f"Failed to download {url}") + except Exception as e: + print(f"Error downloading {url}: {e}") + + # Check if any images were downloaded + if not downloaded_paths: + print("No images were downloaded.") + + return downloaded_paths diff --git a/src/main.py b/src/main.py index 381d431..c9abc99 100644 --- a/src/main.py +++ b/src/main.py @@ -1,31 +1,61 @@ import os -from fastapi import FastAPI, Form, HTTPException +from fastapi import FastAPI, Form, HTTPException, Query from fastapi.responses import HTMLResponse from fastapi.staticfiles import StaticFiles import logging +from dotenv import load_dotenv +from src.auto_annotate_images import auto_annotate_images +from src.download_images import download_images from src.search_images import search_images +from src.search_most_dissimilar_images import select_most_dissimilar_images from src.train_model import train_model from src.scrape_similar import scrape_similar_images +from PIL import Image +from src.create_data_yaml import create_data_yaml +import shutil +from src.utils.annotation_converter import convert_to_yolo_format, ensure_directory + +# Paths to the directories +images_path = "dataset/train/images" +labels_path = "dataset/train/labels" + +# Clear the directories on startup + + +def clear_directory(path): + if os.path.exists(path): + shutil.rmtree(path) # Remove the directory and its contents + os.makedirs(path) # Recreate the directory + + +# Clear the images and labels directories +clear_directory(images_path) +clear_directory(labels_path) + +load_dotenv() # Initialize the FastAPI app app = FastAPI() # Set up logging -logging.basicConfig(level=logging.INFO) +log_file_path = os.path.join(os.getcwd(), 'app_logs.txt') +logging.basicConfig( + level=logging.INFO, + format='%(asctime)s - %(levelname)s - %(message)s', + handlers=[ + logging.FileHandler(log_file_path), + logging.StreamHandler() + ] +) logger = logging.getLogger(__name__) -# Serve static files (like images) from the "static" directory -try: - static_path = os.path.join(os.path.dirname(__file__), "static") - if not os.path.exists(static_path): - raise FileNotFoundError( - f"Static directory '{static_path}' does not exist.") - app.mount("/static", StaticFiles(directory=static_path), name="static") -except Exception as e: - logger.error(f"Error setting up static files: {e}") +# Path for saving and serving static images +download_path = "dataset/train/images" +os.makedirs(download_path, exist_ok=True) -# Route to handle the main form +# Serve static files (like images) from the "dataset/train/images" directory +app.mount("/images", StaticFiles(directory=download_path), name="images") @app.get("/", response_class=HTMLResponse) @@ -46,50 +76,268 @@ async def index(): logger.error(f"Error generating the index page: {e}") raise HTTPException(status_code=500, detail="Internal Server Error") -# Route to handle the search and display images - @app.post("/search", response_class=HTMLResponse) async def search(query: str = Form(...)): try: - api_key = "YOUR_API_KEY" - search_engine_id = "YOUR_SEARCH_ENGINE_ID" + api_key = os.getenv("GOOGLE_API_KEY") + search_engine_id = os.getenv("SEARCH_ENGINE_ID") + original_query = query images = search_images(query, api_key, search_engine_id) - if not images: - return HTMLResponse("

No images found for the query.

", status_code=404) + # Filter the most dissimilar images + selected_images = select_most_dissimilar_images(images, 9) - html_content = f"

Results for: {query}

" - for image_url in images: - html_content += f"
" - html_content += "
" + # Create a more detailed selection interface + html_content = f""" + + + + + + +

Select the images that best represent: {query}

+

Please choose images that clearly show the object you want to detect.

+
+ + {generate_image_previews(selected_images)} +
+ +
+ + + """ return html_content except Exception as e: logger.error(f"Error during image search: {e}") raise HTTPException(status_code=500, detail="Internal Server Error") -# Route to handle selected images and scrape for similar images +def generate_image_previews(images): + return "\n".join([ + f""" +
+
+ +
Select this image +
+ """ for url in images + ]) @app.post("/select", response_class=HTMLResponse) -async def select(selected_images: list[str] = Form(...)): +async def select( + selected_images: list[str] = Form(...), + original_query: str = Form(...), + model_type: str = Query('yolov8', enum=['yolov5', 'yolov8'])): try: if not selected_images: raise HTTPException(status_code=400, detail="No images selected.") - similar_images = scrape_similar_images(selected_images) + # Step 1: Download the selected images and store their local paths + local_image_paths = download_images(selected_images, download_path) + + # Step 2: Display the locally downloaded images for annotation using + # pure HTML5 Canvas + html_content = f""" + + + + + +

Annotate the selected images

+
+ + """ + for idx, local_image_path in enumerate(local_image_paths): + image_filename = os.path.basename(local_image_path) + served_image_path = f"/images/{image_filename}" + + # Ensure each canvas and variable is uniquely named to avoid + # conflict + html_content += f""" +
+

Image {idx + 1}

+ + + +

+
+ + """ + + html_content += """ + +
+ + + """ + return html_content - return "

Model Training Complete

" - except HTTPException as he: - logger.error(f"HTTP error during image selection: {he}") - raise he except Exception as e: - logger.error(f"Error during image selection or model training: {e}") + logger.error(f"Error during image selection: {e}") raise HTTPException(status_code=500, detail="Internal Server Error") + + +@app.post("/save_annotations", response_class=HTMLResponse) +async def save_annotations( + image_urls: list[str] = Form(...), + annotations: list[str] = Form(...), + original_query: str = Form(...), + model_type: str = Query('yolov8', enum=['yolov5', 'yolov8'])): + try: + # Create directories + dataset_dir = "dataset" + images_dir = os.path.join(dataset_dir, "train", "images") + labels_dir = os.path.join(dataset_dir, "train", "labels") + ensure_directory(labels_dir) + + # Process user annotations (convert from canvas JSON to YOLO format) + logger.info(f"Processing {len(annotations)} user annotations") + for image_url, annotation_json in zip(image_urls, annotations): + try: + # Get image path + image_name = os.path.basename(image_url) + image_path = os.path.join(images_dir, image_name) + + # Get image dimensions + with Image.open(image_path) as img: + img_width, img_height = img.size + + # Convert annotation to YOLO format + yolo_annotation = convert_to_yolo_format(annotation_json, img_width, img_height) + + # Save YOLO annotation + label_filename = os.path.splitext(image_name)[0] + ".txt" + label_path = os.path.join(labels_dir, label_filename) + + with open(label_path, 'w') as f: + f.write(yolo_annotation) + + logger.info(f"Converted annotation for {image_name} to YOLO format") + + except Exception as e: + logger.error(f"Error processing annotation for {image_url}: {e}") + + # Scrape and download similar images + api_key = os.getenv("GOOGLE_API_KEY") + search_engine_id = os.getenv("SEARCH_ENGINE_ID") + + logger.info(f"Scraping similar images for query: {original_query}") + similar_images = scrape_similar_images( + image_urls, + original_query, + api_key, + search_engine_id, + num_results_per_image=10, + total_images_to_download=30 # Reduced for faster processing + ) + + # Download images + logger.info(f"Downloading {len(similar_images)} similar images") + download_images(similar_images, images_dir) + + # Auto-annotate similar images + logger.info("Auto-annotating similar images") + auto_annotate_images(images_dir, labels_dir) + + # Create data.yaml config file + logger.info("Creating YAML configuration") + data_yaml_path = create_data_yaml(labels_dir, original_query) + + # Train model + logger.info(f"Training {model_type} model") + model_path = train_model(data_yaml_path, model_type) + + if model_path and os.path.exists(model_path): + return f""" + + +

Training Complete!

+

Your model has been trained to detect: {original_query}

+

Model saved at: {model_path}

+ Train Another Model + + + """ + else: + return f""" + + +

Training Error

+

There was a problem during model training. Check the logs for details.

+ Try Again + + + """ + + except Exception as e: + logger.error(f"Error during model training process: {e}") + raise HTTPException( + status_code=500, + detail=f"Error in model training process: {str(e)}" + ) \ No newline at end of file diff --git a/src/scrape_similar.py b/src/scrape_similar.py index 5000ee6..c90a430 100644 --- a/src/scrape_similar.py +++ b/src/scrape_similar.py @@ -1,14 +1,42 @@ -# Function to scrape similar images based on selected images -from src import search_images +from src.search_images import search_images -def scrape_similar_images(selected_image_urls, api_key, search_engine_id): +def scrape_similar_images( + selected_image_urls, + original_query, + api_key, + search_engine_id, + num_results_per_image=10, + total_images_to_download=50): similar_images = [] - for url in selected_image_urls: - # Use the selected image URL or related terms to search for similar images - # Here, you can use the original search term or variations of it - query = f"related:{url}" # Use the URL to find related images - images = search_images(query, api_key, search_engine_id, num_results=5) - similar_images.extend(images) - - return similar_images + + # Enhanced fallback queries for better results + fallback_queries = [ + f"{original_query} clear photo", + f"{original_query} high resolution", + f"{original_query} isolated", + f"{original_query} product photo", + f"{original_query} professional photo" + ] + + # Add specific filters to the query + base_query = f"{original_query} filetype:jpg OR filetype:png" + images = search_images(base_query, api_key, search_engine_id, num_results=num_results_per_image) + similar_images.extend(images) + + # Use fallback queries only if needed + for fallback_query in fallback_queries: + if len(similar_images) >= total_images_to_download: + break + + fallback_images = search_images( + f"{fallback_query} filetype:jpg OR filetype:png", + api_key, + search_engine_id, + num_results=num_results_per_image) + similar_images.extend(fallback_images) + + # Remove duplicates while preserving order + similar_images = list(dict.fromkeys(similar_images)) + + return similar_images[:total_images_to_download] diff --git a/src/search_images.py b/src/search_images.py index ada2251..a9476b6 100644 --- a/src/search_images.py +++ b/src/search_images.py @@ -1,18 +1,35 @@ import requests -# Function to search for images using Google Custom Search API - - -def search_images(query, api_key, search_engine_id, num_results=5): - - search_url = "https://www.googleapis.com/customsearch/v1" - params = { - 'q': query, - 'cx': search_engine_id, - 'key': api_key, - 'searchType': 'image', - 'num': num_results - } - response = requests.get(search_url, params=params) - results = response.json() - image_urls = [item['link'] for item in results.get('items', [])] - return image_urls + + +def search_images(query, api_key, search_engine_id, num_results=10): + images = [] + # Google Custom Search allows a maximum of 10 results per page + results_per_page = 10 + start_index = 1 + + while len(images) < num_results: + # Adjust the start index for pagination + search_url = ( + f"https://www.googleapis.com/customsearch/v1?" + f"q={query}&searchType=image&key={api_key}&cx={search_engine_id}" + f"&start={start_index}&num={min(results_per_page, num_results - len(images))}") + + response = requests.get(search_url) + if response.status_code != 200: + raise Exception( + f"Failed to fetch images: Status code {response.status_code}, Response: {response.text}") + + data = response.json() + if 'items' not in data: + break # No more results + + for item in data['items']: + images.append(item['link']) # Get the image URL + + # Increment the start index for the next batch of results + start_index += results_per_page + + if len(data['items']) < results_per_page: + break # No more results available + + return images diff --git a/src/search_most_dissimilar_images.py b/src/search_most_dissimilar_images.py new file mode 100644 index 0000000..e21ff98 --- /dev/null +++ b/src/search_most_dissimilar_images.py @@ -0,0 +1,89 @@ +import numpy as np +from sklearn.metrics.pairwise import cosine_distances +from PIL import Image +from torchvision import models, transforms +import torch +from src.download_images import download_images + +# Load a pre-trained model (e.g., ResNet) for feature extraction +# Updated to use the correct 'weights' parameter +model = models.resnet50(weights='IMAGENET1K_V1') +model = model.eval() # Set the model to evaluation mode + +# Transformation for input images (resize, normalize, etc.) +transform = transforms.Compose([ + transforms.Resize((224, 224)), + transforms.ToTensor(), + transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +]) + + +def extract_features(image_path): + """ + Extracts features from an image using a pre-trained model. + + :param image_path: Path to the image. + :return: Feature vector of the image. + """ + try: + image = Image.open(image_path).convert( + 'RGB') # Ensure the image is in RGB format + image_tensor = transform(image).unsqueeze( + 0) # Transform and add batch dimension + with torch.no_grad(): + # Use the model to extract features + # Convert tensor to numpy array and flatten + features = model(image_tensor).flatten().numpy() + return features + except Exception as e: + print(f"Error extracting features from {image_path}: {e}") + return None + + +def select_most_dissimilar_images(image_urls, num_images): + """ + Selects the most dissimilar images from a list. + + :param image_urls: List of image URLs. + :param num_images: Number of most dissimilar images to select. + :return: List of the most dissimilar image URLs. + """ + # Download images to a local directory + # Temporary directory to save downloaded images + download_path = "dataset/train/images" + image_paths = download_images(image_urls, download_path=download_path) + + # Validate that images were downloaded + if not image_paths: + print("No images were downloaded.") + return [] + + # Extract features from each downloaded image + features = [] + valid_image_paths = [] + for path in image_paths: + feature = extract_features(path) + if feature is not None: + features.append(feature) + valid_image_paths.append(path) + + # Ensure there are enough features to proceed + if len(features) < num_images: + print("Not enough images to select the most dissimilar ones.") + return image_urls[:len(features)] # Return as many images as available + + # Convert features list to a numpy array + features = np.array(features) + + # Compute the cosine distance matrix between image features + distance_matrix = cosine_distances(features) + + # Sum the distances for each image and sort them in descending order + dissimilarity_scores = np.sum(distance_matrix, axis=1) + most_dissimilar_indices = np.argsort(dissimilarity_scores)[-num_images:] + + # Select the most dissimilar image URLs + most_dissimilar_images = [image_urls[idx] + for idx in most_dissimilar_indices] + + return most_dissimilar_images diff --git a/src/tests/conftest.py b/src/tests/conftest.py index b7e473d..f211786 100644 --- a/src/tests/conftest.py +++ b/src/tests/conftest.py @@ -3,4 +3,9 @@ import os # Add the src directory to the Python path -sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '../'))) +sys.path.insert( + 0, + os.path.abspath( + os.path.join( + os.path.dirname(__file__), + '../'))) diff --git a/src/tests/test_train_model.py b/src/tests/test_train_model.py index d7cece8..22719ee 100644 --- a/src/tests/test_train_model.py +++ b/src/tests/test_train_model.py @@ -1,16 +1,24 @@ -# src/tests/test_train_model.py from src.train_model import train_model - +import os +import pytest def test_train_model(): - """Test the train_model function with a sample input.""" - image_paths = ["http://example.com/image1.jpg", - "http://example.com/image2.jpg"] - annotations = [] + """Test the train_model function with sample data.""" + # Create test data.yaml + test_yaml = "test_data.yaml" + with open(test_yaml, "w") as f: + f.write(""" +path: dataset +train: train/images +val: train/images +names: + 0: object + """) try: - train_model(image_paths, annotations) + train_model(test_yaml) except Exception as e: assert False, f"train_model raised an exception: {e}" - - # Check for specific output or side effects, if any + finally: + if os.path.exists(test_yaml): + os.remove(test_yaml) \ No newline at end of file diff --git a/src/train_model.py b/src/train_model.py index 3066158..d8134a5 100644 --- a/src/train_model.py +++ b/src/train_model.py @@ -1,10 +1,48 @@ -# Function to train YOLOv8 model from ultralytics import YOLO +import os +import logging +logger = logging.getLogger(__name__) -def train_model(image_paths, annotations): - model = YOLO('yolov8n.pt') # Load a pre-trained YOLOv8 model - # Train the model - model.train(data={'images': image_paths, - 'annotations': annotations}, epochs=50) - model.save('yolov8_trained_model.pt') +def train_model(data_yaml_path, model_type='yolov8'): + """ + Trains the YOLO model using the annotated dataset. + + Args: + data_yaml_path: Path to the data.yaml file containing dataset configuration + model_type: Type of YOLO model to train ('yolov8' recommended) + + Returns: + Path to the trained model + """ + try: + # Initialize YOLO model + model = YOLO('yolov8n.pt') # Start with pre-trained model + + # Train with specific parameters + results = model.train( + data=data_yaml_path, + epochs=25, # Reduced epochs for faster training + imgsz=640, # Image size + batch=8, # Batch size (reduce if memory issues) + patience=10, # Early stopping patience + save=True, # Save model + device='cpu' # Change to 'cuda' if GPU available + ) + + # Get the best model path + model_dir = "runs/detect/train" + if os.path.exists(model_dir): + latest_train_dir = max([os.path.join(model_dir, d) for d in os.listdir(model_dir) + if os.path.isdir(os.path.join(model_dir, d))], + key=os.path.getmtime) + model_path = os.path.join(latest_train_dir, "weights", "best.pt") + logger.info(f"Model trained and saved at {model_path}") + return model_path + else: + logger.error("Training directory not found. Training may have failed.") + return None + + except Exception as e: + logger.error(f"Error during model training: {e}") + return None \ No newline at end of file diff --git a/src/utils/__init__.py b/src/utils/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/src/utils/annotation_converter.py b/src/utils/annotation_converter.py new file mode 100644 index 0000000..bccf3a1 --- /dev/null +++ b/src/utils/annotation_converter.py @@ -0,0 +1,38 @@ +import os +import json +from PIL import Image + +def convert_to_yolo_format(json_annotation, img_width, img_height): + """ + Convert JSON bbox annotation to YOLO format (normalized coordinates). + + Args: + json_annotation: JSON string with x, y, width, height + img_width: Width of the image + img_height: Height of the image + + Returns: + YOLO format string: " " + """ + # Parse JSON string to dict + bbox = json.loads(json_annotation) + + # YOLO uses center coordinates and normalized dimensions + x_center = (bbox['x'] + bbox['width'] / 2) / img_width + y_center = (bbox['y'] + bbox['height'] / 2) / img_height + width = abs(bbox['width'] / img_width) # Use abs to handle negative width from right-to-left drawing + height = abs(bbox['height'] / img_height) # Use abs to handle negative height from bottom-to-top drawing + + # Clamp values between 0 and 1 + x_center = max(0, min(1, x_center)) + y_center = max(0, min(1, y_center)) + width = max(0, min(1, width)) + height = max(0, min(1, height)) + + # Return YOLO format string (class 0) + return f"0 {x_center:.6f} {y_center:.6f} {width:.6f} {height:.6f}" + +def ensure_directory(directory): + """Create directory if it doesn't exist.""" + if not os.path.exists(directory): + os.makedirs(directory) \ No newline at end of file