microsoft/AI-For-Beginners
Publicmirrored fromhttps://github.com/microsoft/AI-For-BeginnersAvailable
lessons/3-NeuralNetworks/04-OwnFramework/lab/README.md
20lines · modecode
| 1 | # MNIST Classification with Our Own Framework |
| 2 | |
| 3 | Lab Assignment from [AI for Beginners Curriculum](https://github.com/microsoft/ai-for-beginners). |
| 4 | |
| 5 | ## Task |
| 6 | |
| 7 | Solve the MNIST handwritten digit classification problem using 1-, 2- and 3-layered perceptron. Use the neural network framework we have developed in the lesson. |
| 8 | |
| 9 | ## Stating Notebook |
| 10 | |
| 11 | Start the lab by opening [MyFW_MNIST.ipynb](MyFW_MNIST.ipynb) |
| 12 | |
| 13 | ## Questions |
| 14 | |
| 15 | As a result of this lab, try to answer the following questions: |
| 16 | |
| 17 | - Does the inter-layer activation function affect network performance? |
| 18 | - Do we need 2- or 3-layered network for this task? |
| 19 | - Did you experience any problems training the network? Especially as the number of layers increased. |
| 20 | - How do weights of the network behave during training? You may plot max abs value of weights vs. epoch to understand the relation. |
| 21 | |