### TensorBoard Setup

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Instructions to start TensorBoard for visualizing training logs. View the logs at the provided URL.

```text
TensorBoard:  Start with 'tensorboard --logdir runs/detect/train', view at http://localhost:6006/
```

--------------------------------

### Install Soccer AI Project

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/README.md

Installs the Soccer AI project from source and sets up the necessary requirements for the soccer example. Ensure you have Python 3.8 or higher.

```bash
pip install git+https://github.com/roboflow/sports.git
cd examples/soccer
pip install -r requirements.txt
./setup.sh
```

--------------------------------

### TensorBoard Logging Setup

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Provides the command to start TensorBoard for visualizing training logs. This helps in monitoring training progress and performance.

```text
TensorBoard:  Start with 'tensorboard --logdir runs/pose/train', view at http://localhost:6006/
```

--------------------------------

### Install Ultralytics and Roboflow Libraries

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Installs the necessary libraries for training YOLOv8 models and interacting with Roboflow. The '-q' flag suppresses verbose output during installation.

```python
!pip install -q ultralytics roboflow
```

--------------------------------

### Training Initialization

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Logs the model graph to TensorBoard and sets up the dataloader workers. It then starts the training process for a specified number of epochs.

```text
TensorBoard: model graph visualization added ✅
Image sizes 640 train, 640 val
Using 2 dataloader workers
Logging results to  runs/pose/train 
Starting training for 100 epochs...
```

--------------------------------

### Training Initialization

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Sets the image sizes for training and validation, specifies the number of dataloader workers, and logs results to the specified directory. Starts training for 50 epochs.

```text
Image sizes 1280 train, 1280 val
Using 8 dataloader workers
Logging results to  runs/detect/train 
Starting training for 50 epochs...
```

--------------------------------

### Install Roboflow Sports Package

Source: https://github.com/roboflow/sports/blob/main/README.md

Install the Roboflow sports package from source. Ensure you have a Python environment with version 3.8 or higher.

```bash
pip install git+https://github.com/roboflow/sports.git
```

--------------------------------

### Training Initialization

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Indicates the start of the training process, including logging results to a specified directory and the total number of epochs. This marks the beginning of the model training phase.

```text
Logging results to  runs/detect/train 
Starting training for 50 epochs...
```

--------------------------------

### Display Validation Prediction Example

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Changes the current directory to HOME and displays an example image of a prediction made during validation. This helps in visually assessing model performance on unseen data.

```python
%cd {HOME}
Image(filename=f'{HOME}/runs/detect/train/val_batch0_pred.jpg', width=600)
```

--------------------------------

### Get Current Working Directory

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Prints the current working directory. This is useful for understanding where files will be saved.

```python
import os
HOME = os.getcwd()
print(HOME)
```

--------------------------------

### List Training Run Files

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Lists the files generated in the training run directory. This helps in inspecting the output artifacts of the training process.

```python
!ls {HOME}/runs/pose/train/
```

--------------------------------

### Display Training Results Summary

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Changes the current directory to HOME and displays the overall results image from the training. This image typically summarizes key metrics.

```python
%cd {HOME}
Image(filename=f'{HOME}/runs/detect/train/results.png', width=600)
```

--------------------------------

### Downloading Pretrained Weights

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Downloads the YOLOv8n pretrained weights from the specified URL.

```text
Downloading https://github.com/ultralytics/assets/releases/download/v8.2.0/yolov8n.pt to 'yolov8n.pt'...
100% 6.25M/6.25M [00:00<00:00, 104MB/s]
```

--------------------------------

### Import Libraries and Initialize Roboflow

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Imports required libraries including Roboflow for dataset management and Google Colab's userdata for API key retrieval. It then initializes the Roboflow client.

```python
from roboflow import Roboflow
from google.colab import userdata
from IPython.display import Image
```

--------------------------------

### Data Scanning and Cache Creation

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Illustrates the process of scanning training and validation datasets, checking for corrupt images, and creating cache files for faster loading. This is crucial for efficient data handling during training.

```text
Scanning /content/datasets/football-ball-detection-2/train/labels... 1966 images, 8 backgrounds, 0 corrupt: 100% 1966/1966 [00:01<00:00, 1339.13it/s]
New cache created: /content/datasets/football-ball-detection-2/train/labels.cache
Scanning /content/datasets/football-ball-detection-2/valid/labels... 121 images, 0 backgrounds, 0 corrupt: 100% 121/121 [00:00<00:00, 1170.74it/s]
New cache created: /content/datasets/football-ball-detection-2/valid/labels.cache
```

--------------------------------

### List Training Artifacts

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Lists the files generated in the training directory. This helps in identifying available results and logs.

```python
!ls {HOME}/runs/detect/train/
```

--------------------------------

### YOLOv8x-pose Model Summary

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Provides a summary of the YOLOv8x-pose model architecture, including layer details, parameter counts, and computational complexity.

```text
                   from  n    params  module                                       arguments                     
  0                  -1  1      2320  ultralytics.nn.modules.conv.Conv             [3, 80, 3, 2]                 
  1                  -1  1    115520  ultralytics.nn.modules.conv.Conv             [80, 160, 3, 2]               
  2                  -1  3    436800  ultralytics.nn.modules.block.C2f             [160, 160, 3, True]           
  3                  -1  1    461440  ultralytics.nn.modules.conv.Conv             [160, 320, 3, 2]              
  4                  -1  6   3281920  ultralytics.nn.modules.block.C2f             [320, 320, 6, True]           
  5                  -1  1   1844480  ultralytics.nn.modules.conv.Conv             [320, 640, 3, 2]              
  6                  -1  6  13117440  ultralytics.nn.modules.block.C2f             [640, 640, 6, True]           
  7                  -1  1   3687680  ultralytics.nn.modules.conv.Conv             [640, 640, 3, 2]              
  8                  -1  3   6969600  ultralytics.nn.modules.block.C2f             [640, 640, 3, True]           
  9                  -1  1   1025920  ultralytics.nn.modules.block.SPPF            [640, 640, 5]                 
 10                  -1  1         0  torch.nn.modules.upsampling.Upsample         [None, 2, 'nearest']          
 11             [-1, 6]  1         0  ultralytics.nn.modules.conv.Concat           [1]                           
 12                  -1  3   7379200  ultralytics.nn.modules.block.C2f             [1280, 640, 3]                
 13                  -1  1         0  torch.nn.modules.upsampling.Upsample         [None, 2, 'nearest']          
 14             [-1, 4]  1         0  ultralytics.nn.modules.conv.Concat           [1]                           
 15                  -1  3   1948800  ultralytics.nn.modules.block.C2f             [960, 320, 3]                 
 16                  -1  1    922240  ultralytics.nn.modules.conv.Conv             [320, 320, 3, 2]              
 17            [-1, 12]  1         0  ultralytics.nn.modules.conv.Concat           [1]                           
 18                  -1  3   7174400  ultralytics.nn.modules.block.C2f             [960, 640, 3]                 
 19                  -1  1   3687680  ultralytics.nn.modules.conv.Conv             [640, 640, 3, 2]              
 20             [-1, 9]  1         0  ultralytics.nn.modules.conv.Concat           [1]                           
 21                  -1  3   7379200  ultralytics.nn.modules.block.C2f             [1280, 640, 3]                
 22        [15, 18, 21]  1  10379251  ultralytics.nn.modules.head.Pose             [1, [32, 3], [320, 640, 640]] 
YOLOv8x-pose summary: 390 layers, 69,813,891 parameters, 69,813,875 gradients, 265.4 GFLOPs
```

--------------------------------

### Download Custom Dataset from Roboflow

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Initializes the Roboflow client with an API key, selects a project and version, and downloads the dataset in YOLOv8 format. Ensure your ROBOFLOW_API_KEY is set in Google Colab secrets.

```python
from roboflow import Roboflow
from google.colab import userdata
from IPython.display import Image

!mkdir {HOME}/datasets
%cd {HOME}/datasets

ROBOFLOW_API_KEY = userdata.get('ROBOFLOW_API_KEY')
rf = Roboflow(api_key=ROBOFLOW_API_KEY)

project = rf.workspace("roboflow-jvuqo").project("football-ball-detection-rejhg")
version = project.version(2)
dataset = version.download("yolov8")
```

--------------------------------

### Display Confusion Matrix

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Changes the current directory to HOME and displays the confusion matrix image from the training results. Requires the 'Image' display utility.

```python
%cd {HOME}
Image(filename=f'{HOME}/runs/detect/train/confusion_matrix.png', width=600)
```

--------------------------------

### Run Player Detection Mode

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/README.md

Execute the script for detecting players, goalkeepers, referees, and the ball. Ensure correct video paths and device are set.

```bash
python main.py --source_video_path data/2e57b9_0.mp4 \
  --target_video_path data/2e57b9_0-player-detection.mp4 \
  --device mps --mode PLAYER_DETECTION
```

--------------------------------

### Run Pitch Detection Mode

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/README.md

Execute the script to detect soccer field boundaries and key points. Specify source and target video paths, and the device to use (e.g., 'mps').

```bash
python main.py --source_video_path data/2e57b9_0.mp4 \
  --target_video_path data/2e57b9_0-pitch-detection.mp4 \
  --device mps --mode PITCH_DETECTION
```

--------------------------------

### Configure Dataset Paths

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Updates the dataset configuration file to specify the locations of training and validation image directories.

```bash
!sed -i 's|\(train: \).*|\1../train/images|' {dataset.location}/data.yaml
!sed -i 's|\(val: \).*|\1../valid/images|' {dataset.location}/data.yaml
```

--------------------------------

### Model Summary

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Displays the layers, parameters, gradients, and GFLOPs of the YOLOv8n model.

```text
Model summary: 365 layers, 68,156,460 parameters, 68,156,444 gradients, 258.1 GFLOPs
```

--------------------------------

### Run Player Tracking Mode

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/README.md

Execute the script to track players across video frames, maintaining their identification. Specify video paths and the device.

```bash
python main.py --source_video_path data/2e57b9_0.mp4 \
  --target_video_path data/2e57b9_0-player-tracking.mp4 \
  --device mps --mode PLAYER_TRACKING
```

--------------------------------

### Albumentations Configuration

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Lists the data augmentation transformations applied during training, including blur, grayscale, and CLAHE.

```text
albumentations:  Blur(p=0.01, blur_limit=(3, 7)), MedianBlur(p=0.01, blur_limit=(3, 7)), ToGray(p=0.01), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))
```

--------------------------------

### Check GPU Availability

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Verifies if a GPU is available and displays its specifications. Navigate to 'Edit' -> 'Notebook settings' -> 'Hardware accelerator' and set to 'GPU' if needed.

```python
!nvidia-smi
```

--------------------------------

### Optimizer Configuration

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Automatically determines the best optimizer, learning rate (lr0), and momentum based on the 'optimizer=auto' setting.

```text
optimizer: 'optimizer=auto' found, ignoring 'lr0=0.01' and 'momentum=0.937' and determining best 'optimizer', 'lr0' and 'momentum' automatically...
optimizer: AdamW(lr=0.00125, momentum=0.9) with parameter groups 97 weight(decay=0.0), 104 weight(decay=0.000515625), 103 bias(decay=0.0)
```

--------------------------------

### Train YOLOv8 Detector

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Initiates the YOLOv8 object detection model training process. Ensure you are in the correct directory and have the dataset configured.

```bash
%cd {HOME}

!yolo task=detect mode=train model=yolov8x.pt data={dataset.location}/data.yaml batch=12 epochs=50 imgsz=1280 plots=True
```

--------------------------------

### Model Summary

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Displays a summary of the model architecture, including the number of layers, parameters, gradients, and GFLOPs. This is useful for understanding model complexity.

```text
Model summary: 365 layers, 68,153,571 parameters, 68,153,555 gradients, 258.1 GFLOPs
```

--------------------------------

### Optimizer Configuration

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Shows the automatic determination of the optimizer, learning rate (lr0), and momentum. It details the chosen optimizer (AdamW) and its specific parameters.

```text
optimizer: 'optimizer=auto' found, ignoring 'lr0=0.01' and 'momentum=0.937' and determining best 'optimizer', 'lr0' and 'momentum' automatically...
optimizer: AdamW(lr=0.002, momentum=0.9) with parameter groups 97 weight(decay=0.0), 104 weight(decay=0.00046875), 103 bias(decay=0.0)
```

--------------------------------

### Run RADAR Mode

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/README.md

Execute the script in RADAR mode to generate a comprehensive visualization of player positions and movements on the soccer field. Ensure all paths and device are correctly specified.

```bash
python main.py --source_video_path data/2e57b9_0.mp4 \
  --target_video_path data/2e57b9_0-radar.mp4 \
  --device mps --mode RADAR
```

--------------------------------

### Run Team Classification Mode

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/README.md

Run the script to classify detected players into teams based on visual features. Set the source and target video paths and the device.

```bash
python main.py --source_video_path data/2e57b9_0.mp4 \
  --target_video_path data/2e57b9_0-team-classification.mp4 \
  --device mps --mode TEAM_CLASSIFICATION
```

--------------------------------

### Display Training Results Image

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Displays a specific image file from the training results. This is useful for visualizing metrics or sample outputs from the training run.

```python
%cd {HOME}
Image(filename=f'{HOME}/runs/pose/train/results.png', width=600)
```

--------------------------------

### Train YOLOv8 Pose Model

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

This command initiates the training of a YOLOv8 pose detection model. Ensure the dataset is correctly configured in `data.yaml` and adjust batch size, epochs, and image size as needed. Mosaic augmentation is disabled for this specific training run.

```python
%cd {HOME}

!yolo task=pose mode=train model=yolov8x-pose.pt data={dataset.location}/data.yaml batch=16 epochs=100 imgsz=640 mosaic=0.0 plots=True
```

--------------------------------

### Run Ball Detection Mode

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/README.md

Use this command to detect and track the ball's position in the video. Provide the source and target video paths and the processing device.

```bash
python main.py --source_video_path data/2e57b9_0.mp4 \
  --target_video_path data/2e57b9_0-ball-detection.mp4 \
  --device mps --mode BALL_DETECTION
```

--------------------------------

### Dataset Scanning and Caching

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Scans and caches the training and validation dataset labels. Caching speeds up data loading during training.

```text
train: Scanning /content/datasets/football-field-detection-12/train/labels... 222 images, 0 backgrounds, 0 corrupt: 100% 222/222 [00:00<00:00, 1070.36it/s]
train: New cache created: /content/datasets/football-field-detection-12/train/labels.cache
val: Scanning /content/datasets/football-field-detection-12/valid/labels... 30 images, 0 backgrounds, 0 corrupt: 100% 30/30 [00:00<00:00, 271.58it/s]
val: New cache created: /content/datasets/football-field-detection-12/valid/labels.cache
```

--------------------------------

### Optimizer Configuration

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Automatically determines the best optimizer, learning rate (lr0), and momentum based on the model and dataset. Here, AdamW is selected.

```text
optimizer: 'optimizer=auto' found, ignoring 'lr0=0.01' and 'momentum=0.937' and determining best 'optimizer', 'lr0' and 'momentum' automatically...
optimizer: AdamW(lr=0.002, momentum=0.9) with parameter groups 103 weight(decay=0.0), 113 weight(decay=0.0005), 112 bias(decay=0.0)
```

--------------------------------

### Scanning Training Data

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Scans the training dataset labels to create a cache file for faster loading. Reports the number of images and backgrounds found.

```text
train:  Scanning /content/datasets/football-players-detection-10/train/labels... 250 images, 0 backgrounds, 0 corrupt: 100% 250/250 [00:00<00:00, 1106.41it/s]
train:  New cache created: /content/datasets/football-players-detection-10/train/labels.cache
```

--------------------------------

### Multiprocessing Warning

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

A RuntimeWarning related to os.fork() being incompatible with multithreaded code, particularly affecting JAX.

```text
/usr/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.
  self.pid = os.fork()
```

--------------------------------

### Deploy YOLOv8 Pose Model

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Use this snippet to deploy a trained YOLOv8 pose model to Roboflow. Ensure the model path points to your trained model artifacts.

```python
project.version(dataset.version).deploy(model_type="yolov8-pose", model_path=f"{HOME}/runs/pose/train/")
```

--------------------------------

### TensorBoard Model Graph

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Adds the model graph visualization to TensorBoard for analysis.

```text
TensorBoard:  model graph visualization added ✅
```

--------------------------------

### Deploy YOLOv8 Model

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Use this snippet to deploy a trained YOLOv8 model to Roboflow. Ensure the `model_path` points to your trained model's directory.

```python
project.version(dataset.version).deploy(model_type="yolov8", model_path=f"{HOME}/runs/detect/train/")
```

--------------------------------

### Download Roboflow Dataset

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Downloads a specific version of a Roboflow dataset in YOLOv8 format. Ensure your ROBOFLOW_API_KEY is set in Colab secrets.

```python
!mkdir {HOME}/datasets
%cd {HOME}/datasets

ROBOFLOW_API_KEY = userdata.get('ROBOFLOW_API_KEY')
rf = Roboflow(api_key=ROBOFLOW_API_KEY)

project = rf.workspace("roboflow-jvuqo").project("football-field-detection-f07vi")
version = project.version(12)
dataset = version.download("yolov8")
```

--------------------------------

### Transferring Pretrained Weights

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Loads pretrained weights for the model. This step is crucial for transfer learning and can significantly speed up training.

```text
Transferred 601/637 items from pretrained weights
```

--------------------------------

### Train YOLOv8 Player Detector Model

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

This code snippet initiates the training process for a YOLOv8 detection model. It specifies the model architecture, dataset configuration, training epochs, image size, and batch size. Ensure the dataset is correctly located and configured in `data.yaml`.

```python
%cd {HOME}

!yolo task=detect mode=train model=yolov8x.pt data={dataset.location}/data.yaml batch=6 epochs=50 imgsz=1280 plots=True
```

--------------------------------

### Plotting Labels

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Generates and saves a plot of the dataset labels to a file for visualization.

```text
Plotting labels to runs/detect/train/labels.jpg...
```

--------------------------------

### Freezing Layer

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Indicates that a specific layer ('model.22.dfl.conv.weight') is being frozen during training.

```text
Freezing layer 'model.22.dfl.conv.weight'
```

--------------------------------

### Validate Pose Model

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Runs validation on the trained pose detection model using a specified dataset and configuration. This command evaluates the model's performance metrics.

```python
%cd {HOME}

!yolo task=pose mode=val model={HOME}/runs/pose/train/weights/best.pt data={dataset.location}/data.yaml
```

--------------------------------

### Display Validation Prediction Image

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_pitch_keypoint_detector.ipynb

Displays a sample prediction image from the validation set. This allows for a visual inspection of the model's performance on unseen data.

```python
%cd {HOME}
Image(filename=f'{HOME}/runs/pose/train/val_batch0_pred.jpg', width=600)
```

--------------------------------

### Validate Trained Model

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_ball_detector.ipynb

Performs validation on the trained YOLOv8 model using specified parameters. This command evaluates the model's performance on the validation dataset.

```bash
%cd {HOME}

!yolo task=detect mode=val model={HOME}/runs/detect/train/weights/best.pt data={dataset.location}/data.yaml imgsz=1280
```

--------------------------------

### Scanning Validation Data

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Scans the validation dataset labels to create a cache file. Reports the number of images and backgrounds found.

```text
val:  Scanning /content/datasets/football-players-detection-10/valid/labels... 43 images, 0 backgrounds, 0 corrupt: 100% 43/43 [00:00<00:00, 964.05it/s]
val:  New cache created: /content/datasets/football-players-detection-10/valid/labels.cache
```

--------------------------------

### AMP Checks Passed

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Confirms that the Automatic Mixed Precision (AMP) checks have passed successfully.

```text
AMP:  checks passed ✅
```

--------------------------------

### AMP Checks

Source: https://github.com/roboflow/sports/blob/main/examples/soccer/notebooks/train_player_detector.ipynb

Automatic Mixed Precision (AMP) checks are performed with YOLOv8n to ensure compatibility and performance.

```text
AMP:  running Automatic Mixed Precision (AMP) checks with YOLOv8n...
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.