Need help with Image Classification using Pytorch.

I need help with Image classification using Pytorch. I have started it and am almost done with section 1 coding but its throwing some errors which I am struggling to fix. Ideally, I would want someone to sit with me and spend a few hours teaching me rather than solving it for me.
The tasks are;
1. Function implementation
1.1 PyTorch Dataset and DataLoader classes
1.2 PyTorch Model class for a simple MLP model
1.3 PyTorch Model class for a simple CNN model
2. Model training

2.1 Train on TinyImageNet30 dataset
2.2 Generate confusion matrices and ROC curves
2.3 Strategies for tackling overfitting
2.3.1 Data augmentation
2.3.2 Dropout
2.3.3 Hyperparameter tuning (e.g. changing learning rate)
3. Model Fine-tuning on CIFAR10 dataset

3.1 Fine-tune your model (initialise your model with pretrained weights from (2))
3.2 Fine-tune model with frozen base convolution layers
3.3 Compare complete model retraining with pretrained weights and with frozen layers. Comment on what you observe?
4. Model testing

4.1 Test your final model in (2) on test set - code to do this
4.2 Upload your result to Kaggle
5. Model comparison

5.1 Load pretrained AlexNet and finetune on TinyImageNet30 until model convergence
5.2 Compare the results of your CNN model with pretrained AlexNet on the same validation set. Provide performance values (loss graph, confusion matrix, top-1 accuracy, execution time)
6. Interpretation of results

6.1 Use grad-CAM on your model and on AlexNet
6.2 Visualise and compare the results from your model and from AlexNet


Please only bid if you think you can help me learn this.

Thanks