Let's dive into the world of data together!
Projects
After training and evaluation, the model achieved an accuracy of 92% on the testing set. The precision, recall, and F1 score were calculated to be 0.91, 0.93, and 0.92, respectively. These results indicate that the model performs well in differentiating between forest and non-forest images.
We compared the performances of the different models using a bar chart visualization. The bar chart displays the accuracy scores of each model, allowing for an easy comparison of their performance.
Step 2: Load the Datasets
Step 3: Explore the Datasets
Step 4: Handle Missing Data
Step 5: Data Cleaning
Assign unique column names to Penguins Iter dataset
Projects
- Covid-19 Positive: Precision
- Covid-19 Negative: Precision
- Virus: Precision
- Contributing:
- Contributions to this project are welcome! If you find any issues or want to add new features, feel free to open a pull request.
Results
After training and evaluation, the model achieved an accuracy of 90% on the testing set. The precision, recall, and F1 score were calculated to be 0.89, 0.92, and 0.90, respectively. These results indicate that the model performs well in distinguishing between forest and building images.
Project Description
In this project, we analyze a dataset containing information about old cars, including their make, model, year, mileage, price, and other relevant features. Through exploratory data analysis, we aim to uncover patterns, correlations, and key insights related to car prices.
Power- Bi Projects
- Created an interactive dashboard to visualize online sales data trends.
- Implemented complex parameters to enable detailed drill-down and customization using filters and slicers.
- Utilized data connections and joined new tables, alongside calculations
- Visualizations for sales trends over time.
- Product performance analysis by category and region.
- Customer segmentation based on purchasing behavior.
- Interactive filters to drill down into specific data subsets.
- Cross-page interactions for a seamless user experience.
- Earthquake frequency analysis by city and country.
- Magnitude distribution histograms for selected regions.
- Geospatial visualization of earthquake epicenters on a map.
- Interactive filters to refine the data view.
- Data tables with detailed earthquake information.
- 📊 Data Analysis
- 🧹 Data Cleaning and Preprocessing
- 📈 Statistical Analysis
- 📊 Data Visualization
- 🔮 Predictive Modeling
- 🤖 Machine Learning
- 📊 SQL Querying
- 📖 Data Storytelling
- 🚀 Problem-Solving
- 💬 Effective Communication
I'm continuously expanding my skill set to stay ahead in the rapidly evolving world of data analysis. Let's collaborate and leverage the power of data together!
Skills and Expertise
The Art of Data Analysis
- Data Exploration & Visualization
- Statistical Modeling & Hypothesis Testing
- Predictive Analytics & Machine Learning
- Data Cleaning & Preprocessing
- SQL Querying & Database Management
- Business Intelligence & Dashboards
- Data Storytelling & Communication
- Problem-Solving & Critical Thinking
I'm driven by the potential of data to unveil hidden truths and spark transformative change. Let's collaborate and weave data's magic into your endeavors!
"My Creative Portfolio: Exploring the World of Cats and Dogs"
Data Odyssey - Projects in Focus
Project Report: Cat vs Dog Image Classification
Project Steps :
1. Import Libraries and Mount Google Drive
The necessary libraries and modules are imported, and Google Drive is mounted to access the dataset.
2. Load and Preprocess Images
The training image paths are loaded, and the paths are shuffled randomly. The first five image paths are displayed to verify the loading process. The OpenCV library is used to read and resize the images, and the images are converted to arrays. The class labels are extracted from the image paths
3. Split the Dataset
The dataset is split into training and validation sets using the train_test_split function from scikit-learn. The sizes of the training and validation sets are displayed.
4. Data Augmentation
An ImageDataGenerator object is created to apply data augmentation techniques, such as rotation, shifting, shearing, zooming, and flipping, to the training images. This helps to increase the robustness and variability of the training data.
5. Create the CNN Model
A CNN model based on the LeNet architecture is defined using the Keras Sequential API. The model consists of convolutional, activation, and pooling layers, followed by fully connected layers. The model is compiled with the Adam optimizer and categorical cross-entropy loss.
6. Train the Model
The model is trained using the fit function with the training data and validation data. The number of epochs and batch size are defined. The training progress is displayed during the training process.
7. Plot Training and Validation Accuracy
The training and validation accuracy values are plotted against the number of epochs to visualize the model's learning progress.
8. Save the Trained Model
The trained model is saved to "/content/drive/MyDrive/best_model/cat_dog_new.model" for future use.
9. Test the Model on Test Images
The saved model is loaded, and the test images are processed to make predictions. The predictions are displayed along with the corresponding images. The image numbers and predicted labels are stored in a DataFrame for further analysis.
10. Convert Predicted Labels to Numeric Values
The predicted labels are converted from categorical values ("Cat" and "Dog") to numeric values (0 and 1) for analysis purposes.
11. Create DataFrame with Image Numbers and Predicted Labels
A DataFrame is created with columns for image numbers and predicted labels.
Data Dashboard Showcase
Data Dashboard Showcase
Discover the power of data visualization through our video showcase. Watch as we dive into interactive data dashboards, offering insights and applications. Stay tuned for new additions, and contact us with any questions or project inquiries.
Discover the power of data visualization through our video showcase. Watch as we dive into interactive data dashboards, offering insights and applications. Stay tuned for new additions, and contact us with any questions or project inquiries.
Let's Get in Touch!
Excited to collaborate or have questions about my work? Drop me a line or two! I'm just an email away at aliasadcivil007@gmail.com.
You can also reach out via LinkedIn: https://www.linkedin.com/in/asad-ali-mulla-044240262/
Let's embark on an extraordinary data journey together!
