{"id":26575665,"url":"https://github.com/miladbadeleh/weld-defect-detection-using-resnet50","last_synced_at":"2026-03-07T03:31:40.066Z","repository":{"id":282615575,"uuid":"949141305","full_name":"miladbadeleh/Weld-Defect-Detection-Using-ResNet50","owner":"miladbadeleh","description":"This repository contains a PyTorch implementation of a ResNet50 model for classifying weld defects. The model is trained on a custom dataset of weld images to detect various types of aesthetic defects.","archived":false,"fork":false,"pushed_at":"2025-04-08T11:01:14.000Z","size":12,"stargazers_count":7,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-10-04T11:37:45.278Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/miladbadeleh.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-03-15T19:15:14.000Z","updated_at":"2025-09-12T03:05:42.000Z","dependencies_parsed_at":"2025-08-16T04:01:55.202Z","dependency_job_id":null,"html_url":"https://github.com/miladbadeleh/Weld-Defect-Detection-Using-ResNet50","commit_stats":null,"previous_names":["miladbadeleh/weld-defect-detection-using-resnet50"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/miladbadeleh/Weld-Defect-Detection-Using-ResNet50","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/miladbadeleh%2FWeld-Defect-Detection-Using-ResNet50","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/miladbadeleh%2FWeld-Defect-Detection-Using-ResNet50/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/miladbadeleh%2FWeld-Defect-Detection-Using-ResNet50/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/miladbadeleh%2FWeld-Defect-Detection-Using-ResNet50/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/miladbadeleh","download_url":"https://codeload.github.com/miladbadeleh/Weld-Defect-Detection-Using-ResNet50/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/miladbadeleh%2FWeld-Defect-Detection-Using-ResNet50/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":30206567,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-07T03:24:23.086Z","status":"ssl_error","status_checked_at":"2026-03-07T03:23:11.444Z","response_time":53,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"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":[],"created_at":"2025-03-23T02:29:50.752Z","updated_at":"2026-03-07T03:31:39.807Z","avatar_url":"https://github.com/miladbadeleh.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# ResNet50 Weld Defects Classification\n\nThis repository contains a PyTorch implementation of a ResNet50 model for classifying weld defects. The model is trained on a custom dataset of weld images to detect various types of aesthetic defects.\n\n## Overview\n\nThe project uses a pre-trained ResNet50 model, which is a deep convolutional neural network architecture that has been widely used for image classification tasks. The model is fine-tuned on a custom dataset of weld images to classify different types of weld defects.\n\n## Dataset\n\nThe dataset consists of images of welds, categorized into different classes based on the type of defect. The dataset is divided into three sets:\n- **Training set**: Used to train the model.\n- **Validation set**: Used to tune the model and prevent overfitting.\n- **Test set**: Used to evaluate the final performance of the model.\n\nThe images are resized to 300x30 pixels and normalized using the mean and standard deviation of the ImageNet dataset.\n\n## Model Architecture\n\nThe model used in this project is based on the ResNet50 architecture, which consists of 50 layers. The key feature of ResNet is the use of residual blocks, which help in training deeper networks by addressing the vanishing gradient problem.\n\n### Modifications to ResNet50\n\nThe final fully connected layer of the ResNet50 model is modified to match the number of classes in the custom dataset. This allows the model to output predictions for the specific number of defect classes in the dataset.\n\n## Training\n\nThe model is trained using the following configuration:\n- **Loss Function**: CrossEntropyLoss\n- **Optimizer**: Adam with a learning rate of 0.001\n- **Number of Epochs**: 10\n\nDuring training, the model's performance is monitored on both the training and validation sets. The training process includes:\n- Forward pass to compute the output and loss.\n- Backward pass to compute gradients.\n- Optimization step to update the model's weights.\n\n## Evaluation\n\nAfter training, the model is evaluated on the test set to measure its performance. The evaluation metrics include:\n- **Test Loss**: The average loss over the test set.\n- **Test Accuracy**: The percentage of correctly classified images in the test set.\n\n## Results\n\nThe model achieves the following performance on the test set:\n- **Test Loss**: 0.2215\n- **Test Accuracy**: 89.97%\n\n## Usage\n\nTo use this model, follow these steps:\n\n1. **Clone the repository**:\n   ```bash\n   git clone https://github.com/yourusername/resnet50_weld_defects.git\n   cd resnet50_weld_defects\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmiladbadeleh%2Fweld-defect-detection-using-resnet50","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmiladbadeleh%2Fweld-defect-detection-using-resnet50","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmiladbadeleh%2Fweld-defect-detection-using-resnet50/lists"}