{"id":24030691,"url":"https://github.com/rvk007/deepnet","last_synced_at":"2026-05-17T07:32:15.539Z","repository":{"id":40966512,"uuid":"250557756","full_name":"rvk007/DeepNet","owner":"rvk007","description":"A python library for computer vision applications","archived":false,"fork":false,"pushed_at":"2024-07-25T10:58:42.000Z","size":556,"stargazers_count":1,"open_issues_count":3,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-01-08T17:58:46.404Z","etag":null,"topics":["bce-dice","bce-rmse","cifar10","cnn","computer-vision","deep-learning","gradcam","lrfinder","mask-detection","monocular-depth-estimation","object-recognition","python","pytorch"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/rvk007.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2020-03-27T14:36:07.000Z","updated_at":"2020-09-15T14:26:27.000Z","dependencies_parsed_at":"2022-09-03T13:53:05.415Z","dependency_job_id":null,"html_url":"https://github.com/rvk007/DeepNet","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rvk007%2FDeepNet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rvk007%2FDeepNet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rvk007%2FDeepNet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rvk007%2FDeepNet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/rvk007","download_url":"https://codeload.github.com/rvk007/DeepNet/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240790268,"owners_count":19858003,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bce-dice","bce-rmse","cifar10","cnn","computer-vision","deep-learning","gradcam","lrfinder","mask-detection","monocular-depth-estimation","object-recognition","python","pytorch"],"created_at":"2025-01-08T17:56:25.833Z","updated_at":"2026-05-17T07:32:15.476Z","avatar_url":"https://github.com/rvk007.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"images/3.png\" alt=\"tensornet\" /\u003e\n  \u003cbr /\u003e\n\u003c/p\u003e\n\n\n[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1g63kM2rq3pktpTx5neqNlbSVYeT9xvEk)  \n\nDeepnet is an open-source library that can be used for solving problems of Computer vision in Deep Learning.  \n \nNOTE: This documentation applies to the MASTER version of DeepNet only.\n\n\n## Install Dependencies\n\nInstall the required packages  \n`pip install -r requirements.txt`\n\n## Features\n\nDeepNet currently supports the following features:\n\n### Models\n\n| Models | Description |\n| ------- | ----- |\n| [ResNet](./model/models/resnet.py) | ResNet-18 |\n| [ResModNet](./model/models/resmodnet.py) | A modified version of ResNet-18  |\n| [CustomNet](./model/models/customnet.py) | A modified version of ResNet-18   |\n| [MaskNet3](./model/models/masknet.py) | A model to predict the Segmentation mask of the given image. |\n| [DepthMaskNet8](./model/models/depthnet.py) | A model to predict the Monocular Depth Maps of the given image. |\n\n### Training and Validation\n\n| Functionality | Description |\n| ------- | ----- |\n| [Train](/home/rvk/DeepNet/model/train.py) | Training and Validation of the model |\n| [Model](/home/rvk/DeepNet/model/learner.py) | Handles all the function for training a model  |\n| [Dataset](/home/rvk/DeepNet/data/dataset) | Contains classes to handle data for training the model|\n\n### Metrics\n\n- [Mean Absolute Error](https://github.com/rvk007/DeepNet/blob/f67732d2d65798289925ea76d58f1d8636f13273/model/metrics.py#L36)\n- [Root Mean Squared Error](https://github.com/rvk007/DeepNet/blob/f67732d2d65798289925ea76d58f1d8636f13273/model/metrics.py#L50)\n- [Mean Absolute Relative Error](https://github.com/rvk007/DeepNet/blob/f67732d2d65798289925ea76d58f1d8636f13273/model/metrics.py#L67)\n- [Intersection Over Union Error](https://github.com/rvk007/DeepNet/blob/f67732d2d65798289925ea76d58f1d8636f13273/model/metrics.py#L84)\n- [Root Mean Square Error](https://github.com/rvk007/DeepNet/blob/f67732d2d65798289925ea76d58f1d8636f13273/model/metrics.py#L130)\n\n### Losses\n\n| Loss | Description |\n| ------- | ----- |\n| [Dice](./model/losses/dice_loss.py) | ResNet-18 |\n| [SSIM](./model/losses/ssim.py) | A modified version of ResNet-18  |\n| [MSE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss.py#L5) | Mean squared error (squared L2 norm) between each element in the input and target   |\n| [BCE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss.py#L22) | Binary Cross Entropy between the target and the output |\n| [BCEWithLogitsLoss](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss.py#L41) | Combination of Sigmoid layer and the BCE in one single class |\n| [RMSE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss.py#L58) | Root mean squared error (squared L2 norm) between each element in the input and target |\n\nWeighted Combination of loss functions\n- [BCE-RMSE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L10)\n- [SSIM-RMSE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L22)\n- [BCE-SSIM](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L34)\n- [RMSE-SSIM](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L46)\n- [SSIM-DICE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L57)\n- [RMSE-DICE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L68)\n- [BCE-DICE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L79)\n- [RMSE-BCE-DICE](https://github.com/rvk007/DeepNet/blob/44ea35c02df7e719fc5c7f2f0c5da6f0cbfec4e3/model/losses/loss_combination.py#L103)\n\n### Scheduler\n\n- StepLR\n- ReduceLROnPlateau\n- OneCycleLR\n\n### Data Augmentation\n\n  - Resize\n  - Padding\n  - Random Crop\n  - Horizontal Flip\n  - Vertical Flip\n  - Gaussian Blur\n  - Random Rotation\n  - CutOut\n\n### Utilities\n\n| Utility | Description |\n| ------- | ----- |\n| [GRADCAM](./gradcam/gradcam.py) | Calculates GradCAM(Gradient-weighted Class Activation Map) saliency map |\n| [GradCAMpp](./gradcam/gradcam_pp.py) | Calculate GradCAM++ salinecy map using heatmap and image |\n| [LRFinder](./lr_finder/lr_finder.py) | Range test to calculate optimal Learning Rate  |\n| [Checkpoint](./utils/checkpoint.py) | Loading and saving checkpoints  |\n| [ProgressBar](./utils/progress_bar.py) | Display Progress bar |\n| [Tensorboard](./utils/tensorboard.py) | Creates Tensorboard visualization  |\n| [Summary](./utils/summary.py)| Display model summary |\n| [Plot](./utils/plot.py)| Plot the graph of a metric, prediction image and class accuracy |\n\n## Dependencies\n\nDeepNet has the following third-party dependencies\n\n- numpy\n- torch\n- torchvision\n- torchsummary\n- tqdm\n- matplotlib\n- albumentations\n- opencv-python\n\nFor a demo on how to use these modules, refer to the notebooks present in the [examples](./examples) directory.\n\n## Contact/Getting Help\n\nIf you need any help or want to report a bug, raise an issue in the repo.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frvk007%2Fdeepnet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Frvk007%2Fdeepnet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frvk007%2Fdeepnet/lists"}