Applying machine learning and deep learning
techniques to real life datasets:
improve your skills by applying ML!
AML Basic: connect to the Teams channel
This course is part of the Bioinformatics LM curriculum and of the Data Science and Computation PhD programme at University of Bologna. It will introduce the learner to applied machine learning and deep learning, focusing more on intuition on the theory and practical techniques and methods than on the theoretical statistical foundations behind these methods. The course will proceed through phases, starting from intuition-based understanding of the key aspects of machine learning, moving to hands-on exercises with various tools needed in a "data scientist virtual backpack", then proceeding through applications of ML/DL methods to problems of increasing complexity. Applications to real life datasets - as well as some peculiar scientific datasets - will be discussed throughout the course.
For the AY 2020-21, the course is offered as a set of lectures and hands-on (either fully online or hybrid) corresponding to 4+6 CFU (roughly 80 hours) in total. It is divided into two parts. The first part is "Applied ML - Basic" (4 CFU). The second part is "Applied ML - Advanced" (6 CFU). Attendance is open, anyone interested is always welcome! This is specially true for PhD students from programmes other than DSC (e.g. in the past we had PhD students in Physics, Mathematics, Economics, and more). The final evaluation will require to pass a written exam and to finalize a coding project (details at the lectures).
Pre-requisites
This course is intended for learners with basic Python and/or programming background. Minimal statistics background is expected. The introductory lessons will offer a refresh of basic concepts as well as tutorial-like crash courses on key tools needed. Read more by clicking the link below.
Tools
The primary language used in the course will be Python via Jupyter notebooks. Mini crash courses on e.g. numpy, pandas, matplotlib, seaborn and more will be given. The primary ML/DL libraries and frameworks used will be Scikit-learn, Google Tensorflow, Keras and (time permitting) Pytorch.
Material
Selected material from the classes - both slides and code - will be available to students attending the course. All code base will be in Python and students will be able to play with it via Jupyter notebooks.
Exam
The final evaluation will consist in a written exam plus a self-contained coding project, proposed by the student or selected in a list proposed by the teacher. Deadlines for project submission will be inserted in Almaesami.
Calendar
The
lectures
period
for
the AY
2020-21
will
be:
9
March
- 9
April
2021
for
"Applied
ML -
Basic",
6
May -
11
June
2021
for
"Applied
ML -
Advanced".
Lecture slots are constrained by the Bioinformatics (LM) timetable, and will be on Thursday mornings and Friday afternoons.
AML Basic: connect
to the Teams
channel
Check
out a
detailed
calendar
in the
link
below.
Room
The
lectures
will
be in
hybrid
mode:
Thursday
mornings
online-only and
Friday
afternoons
in
presence
(in
the
"Aula Farbiomot"
(access
in Via
Selmi 3). Please
note
that
this
may
(probably)
change
and
ALL
lectures
might
be
online-only,
especially
in the
"Basic"
first
part
of the
course.
Check
out more details in the link below.