Skip to content

Latest commit

 

History

History
7 lines (5 loc) · 473 Bytes

File metadata and controls

7 lines (5 loc) · 473 Bytes

UC Berkeley CS188 (Artificial Intelligence) project:

• Built a neural network that classified handwritten digits from 0-9 with 97% accuracy
• Implemented a language identifier using a neural network that categorized individual words as either English, Spanish, Finnish, Dutch, or Polish with accuracy above 80%
• Experimented with different components such as batch size, learning rate, and neural network structure

See models.py for constructed models