avalanche

avalanche copied to clipboard

Memory Replay increased GPU memory consumption after the first experience

2

Hi, I am not sure if this is expected behavior when I run experiment with Memory Replay plugin only, but after the first experience the GPU memory usage is increased...

matkowski-voy

Add Expert Gate to Avalanche

23

Hi, this is my initial draft of the ExpertGate Plugin. It deals with issue #1023 A few points: - The ExpertGate model is implemented as a wrapper which holds two...

niniack

New mixin-based template structure

7

I'm creating this draft PR for the new mixin-based template structure. For now, the main goal is to implement working version of the templates that we want and get feedback...

HamedHemati

🐛 **Describe the bug** As the title says, the property `classes_in_experience` of the class `GenericCLScenario` doesn't return an set of unique tensors containing the labels of the experience's patterns. Instead,...

DWarez

ImportError: 'make_lvis_metrics'

6

🐛 **Describe the bug** A clear and concise description of what the bug is. I got this error message: `ImportError: cannot import name 'make_lvis_metrics' from 'avalanche.evaluation.metrics' ` while running one...

alaa-shubbak

Distributed support (rework)

16

This draft contains the implementation for supporting distributed (multi-GPU) training. The examples folder contains a simple example (to run naive, replay, replay+scheduler) and a bash script to launch it. This...

lrzpellegrini

JointTraining strategy does not shuffle the whole dataset when picking minibatch

1

So I noticed that when calling JointTraining on a `dataset_benchmark` the mini-batches are not iid w.r.t. the whole datasets. Any clue on how to solve the issue?

ema-marconato

periodic eval is broken in some subtle cases. Instead of calling `eval` on the entire stream, it calls it multiple times for each experience, breaking stream-level metrics. Reproduce: ``` from...

AntonioCarta

Core50 benchmark with ICARL strategy

6

ICarl strategy returns an error with Core50 benchmark: `scenario = CORe50(scenario="nc", run=9)` The error is: `AttributeError: 'GenericCLScenario' object has no attribute 'n_classes_per_exp'` Other nc benchmarks belong to "NCScenario" class and...

gab709

Issue with class-incremental SI and LwF

15

Hi! We are currently trying to use Avalanche in our research, as it looks like an amazing tool providing a lot of ready-to-use tools. However, we encountered some issues that...

Nyderx