Machine Learning and Molecules (MLmol)
Students write their own neural network code from scratch using Python and use it to predict chemical properties of molecules. The performance is compared to standard machine learning packages such as scikit-learn, Keras, and DeepChem.
Basic principles behind Python programming, machine learning, and cheminformatics. Classification and regression using neural networks. Activation functions, back propagation using gradient descent, Overfitting, regularisation, hyperparameter optimisation, and training/validation/test sets. SMILES strings, molecular fingerprints, and graph convolution as applied to molecules.
Data manipulation and visualisation using Pandas, numpy, and Matplotlib/Seaborn. Manipulation of chemical data using RDKit. Use of scikit-learn, Keras, and DeepChem.
Prediction of chemical properties using machine learning. Critical evaluation of machine learning models.
Videolectures and classroom discussion
See Absalon for a list of course literature
First year organic chemistry and mathematics
- 7,5 ECTS
- Type of assessment
Oral examination, 30 minutes
- Type of assessment details
- A final 30-minute individual oral examination is without preparation and based on the project report. The exam begins with a 5 minute presentation from the examinee, after which the internal assessors ask questions in the project area.
- All aids allowed
- Marking scale
- passed/not passed
- Censorship form
- No external censorship
Several internal examiners
Criteria for exam assessment
See Learning Outcome
- Class Instruction
- Project work
- Course number
- 7,5 ECTS
- Programme level
- Block 2, Block 3 And Block 4
Much of the instruction comes from video lectures supplemented by one weekly meeting for Q&A.
- The number of seats may be reduced in the late registration period
- Study Board of Physics, Chemistry and Nanoscience
- Department of Chemistry
- Faculty of Science
- Jan Halborg Jensen (8-6f6d6f6a73786a7345686d6a7233707a336970)
Are you BA- or KA-student?
Courseinformation of students