fimm 0.0.3

Finetune PyTorch Image Models with TIMM

FIMM

Finetune PyTorch Image Models with TIMM

This project provides a simple way to finetune PyTorch Image Models with TIMM.

Installation

To install FIMM (fimm), you can simply use pip:

pip install fimm

Install from source

To install from source, you can clone this repo and install with pip:

git clone https://github.com/rapanti/fimm
pip install -e fimm  # -e for editable mode

Usage

To use FIMM, you can simply run the follwing command to train or finetune a model:

train --data-dir /path/to/dataset --model resnet50 --experiment resnet50 # this trains a resnet50 model from scratch
train --data-dir /path/to/dataset --model resnet50 --experiment resnet50 --pretrained  # this finetunes a resnet50 model

To validate the performance of a model, you can simply run the following command:

validate --data-dir /path/to/eval/dataset --model resnet50 --checkpoint output/train/resnet50/model_best.pth.tar # this tests the resnet50 model