resources's Introduction

📚 Datature Resources 📚

⚡ A repository of resources used in our tutorials and guides ⚡

👋 Hey there! This library is a collection of useful scripts and tutorials that can be used for integrating with our platform tools on Datature Nexus, or for general Computer Vision application purposes. We also showcase demos of some of the state-of-the-art (SOTA) techniques in areas like Generative AI, Multiple Object Tracking (MOT), and many others.

Made with ❤️ by Datature

📌 Script Categories 📌

📑 Example Scripts

This section contains example scripts that can be used for integrating with our platform tools, or for general CV application purposes.

Topic	Description	Links
	Action Recognition For identifying and classifying actions with keypoint detection models.
	Active Learning For performing active learning on your dataset.
	Data Preprocessing Useful tools for preprocessing your data.
	Explainable AI For interpreting your deep learning model.
	Generative AI Synthetic data generation to boost your dataset.
	Inference Dashboard For easy visualizations of inference results.
	Industry Use Cases For industry-specific use cases.
	K-Shot Learning Sample scripts for one-shot and few-shot learning.
	Tracking For single and multi-object tracking in videos.

🔨 SDK Guides

This section contains guides and code snippets on how to use our Datature Python SDK for automating tasks without having to interact with our Nexus platform. The SDK is available on PyPI. It can be installed by running the following command:

pip install -U datature

The SDK can either be invoked in Python, or through the command line interface (CLI). Installing the pip package will install both the SDK and CLI together. For more information or advanced features on the SDK, please refer to the SDK documentation.

🏭 Deployment

This section contains scripts on how to deploy your models trained on Nexus for inference. We currently support the following deployment methods:

Topic	Description	Links
	Edge Deployment For deploying models on edge devices such as Raspberry Pi and NVIDIA Jetson Orin.
	Inference API Where models are hosted on our servers and inference can be performed through API calls.
	Local Inference For running simple inference scripts on your local machine.

🚀 Getting Started 🚀

⚙️ Prerequisites

Firstly, users should clone this repository and change to the resource folder directory.

git clone https://github.com/datature/resources.git
cd resources

In each folder, there will be a requirements.txt file that contains the dependencies required for Python scripts to run. Users can install the dependencies by running the following command:

pip install -r requirements.txt

For running Jupyter notebooks locally, users should install Jupyter by running the following command:

pip install notebook

We recommend users to create a virtual environment before installing any dependencies. For more information on virtual environments, please refer to:

🔑 Usage

Each folder contains a README.md file that contains the instructions for running the scripts. Please refer to the README.md file for more information.

✏️ Contributing

Do let us know via an issue if there are any bugs or something is not working, and we will rectify it as soon as we can! Alternatively, you can also submit a pull request with your changes and our team will review it.

We also welcome new contributions to this repository if there are any interesting cutting-edge techniques that we might not have spotted. Please refer to CONTRIBUTING.md for more information on what areas you can contribute in and coding best practice guidelines.

resources's People

Contributors

Stargazers

Watchers

resources's Issues

Issue with DeepSORT notebook

DeeoSORT notebook at this path
https://github.com/datature/resources/tree/main/example-scripts/tracking/DeepSORT
is generating this error.

Run DeepSORT for Video

video_path = "./pexels-rodnae-productions-5699922.mp4" # this video must be there in pwd
model_path = "./model" # dir model must be in pwd. It must contain some infor about the model as weights, configurations, or other resources used by a machine learning model.
output_vid_trk_path = "./output_" + video_path.split("./")[1].split(".")[0] + ".mp4"
size = "320,320"
threshold = "0.7"
output_format = "mp4v"
COLORS = np.random.randint(0, 255, size=(200, 3))

start = time.time()
output_vid = DeepSORT(
video_path, model_path, size, threshold, output_format, output_vid_trk_path, COLORS
)
end = time.time()
print(
"Tracking by DeepSORT for " + video_path + " takes " + str(int(end - start)) + "s."
)

Error
OSError Traceback (most recent call last)
in <cell line: 10>()
8
9 start = time.time()
---> 10 output_vid = DeepSORT(
11 video_path, model_path, size, threshold, output_format, output_vid_trk_path, COLORS
12 )

4 frames
/usr/local/lib/python3.10/dist-packages/tensorflow/python/saved_model/loader_impl.py in parse_saved_model(export_dir)
117 raise IOError(f"Cannot parse file {path_to_pbtxt}: {str(e)}.") from e
118 else:
--> 119 raise IOError(
120 f"SavedModel file does not exist at: {export_dir}{os.path.sep}"
121 f"{{{constants.SAVED_MODEL_FILENAME_PBTXT}|"

OSError: SavedModel file does not exist at: ./model/saved_model/{saved_model.pbtxt|saved_model.pb}

Note:

In original repo. No such file having saved model is present.
video_path = "./pexels-rodnae-productions-5699922.mp4" #

This video must be there in present working dir, but ?I cannot findany

model_path = "./model"
Directory model must be in present working directory, but nothing is there.

Recommend Projects

datature / resources Goto Github PK

resources's Introduction

📚 Datature Resources 📚

📌 Script Categories 📌

📑 Example Scripts

Action Recognition

Active Learning

Data Preprocessing

Explainable AI

Generative AI

Inference Dashboard

Industry Use Cases

K-Shot Learning

Tracking