Downloads · 30 days
0
karemaa/A4_4310154_karimah
A4_4310154_karimah is a machine learning model from karemaa. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Course: AI417 - Deep Learning | Organization: AI417UPM
Downloads · 30 days
0
Access
Public
Updated Feb 20, 2026
Repo size
5.5 MB
Likes
0
Public
Click a slice to open those files.
.bin5.5 MB · 100%
From the Hugging Face model README
Course: AI417 - Deep Learning | Organization: AI417UPM
A deep feedforward network with LeakyReLU activations, trained on FashionMNIST.
| Detail | Value |
|---|---|
| Architecture | 784 → 512 → 384 → 256 → 192 → 128 → 96 → 64 → 48 → 32 → 10 |
| Activation | LeakyReLU (α=0.01) |
| Weight Init | Kaiming Uniform + Xavier (last layer) |
| Optimizer | AdamW (lr=0.001, weight_decay=0.01) |
| Epochs | 300 |
| Final Loss | 0.70707 |
| Eval Accuracy | 89.72% |
| Training Time | ~294 seconds |
A simple feedforward network without activation functions, trained on FashionMNIST.
| Detail | Value |
|---|---|
| Architecture | 784 → 512 → 256 → 128 → 64 → 32 → 10 |
| Activation | None |
| Weight Init | Xavier Uniform |
| Optimizer | Adagrad (lr=0.001) |
| Epochs | 300 |
| Final Loss | 0.44286 |
| Eval Accuracy | 84.43% |
| Training Time | ~172 seconds |
from huggingface_hub import snapshot_download
import torch as tr
repo = snapshot_download('ai417upm/A4_YOURID_YOURNAME', token="YOUR-TOKEN")
# Load VanillaNet
model = tr.hub.load(repo, 'VanillaNet', source='local')
model.eval()
predictions = model(tr.randn(10000, 784))
# Load DummyNet
model = tr.hub.load(repo, 'DummyNet', source='local')
model.eval()
predictions = model(tr.randn(10000, 784))