FlowerImageClassifier

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #typescript #llm #go #golang #android #machine-learning #rust #deep-learning #linux

FlowerImageClassifier

(★ 8)

Training, Testing & Validating a dataset of flower images using Pytorch. Command line application with optional arguments, training on GPU

FlowerImageClassifier 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

Flower Image Classifier
===
This Command Line Image Classifier project was part of my Nanodegree Program with *Udacity*.
This is the first time I've ever created an AI that trains, validates and predicts from scratch.

Authors & Contributors
---
Mariya Shavshyshvili (mariyasha888@gmail.com), with the help and instructions of the *Udacity Nanodegree Program*.

**IMPORTANT CREDITS:**

*Johnny Metz's* video on ```argparse``` helped me out a lot with understanding it's uses:

https://youtu.be/cdblJqEUDNo

*LearnCode.academy's* video on GitHub helped me with uploading this project to GitHub:

https://youtu.be/0fKg7e37bQE

File List
---
cat_to_name.json

functions_and_classes.py

LICENSE

predict.py

README.md

train.py

Terminal Instructions
---
Based on project instructions from *Udacity*, there are a few specific details about the terminal use that are best to note;

**in train.py**

In the terminal, train.py is best called with:
```
cd python train.py ./flowers -g
```
with ```cd``` being the directory in which train.py is in.

with ```-g``` representing the GPU connection so that everything runs faster.

with ```./flowers``` being the directory in which we save our training, validation and testing sets.
The ```./flowers``` directory would have 3 folders: train, valid, test.
These folders would be separated into sub-folders, holding the class number of the flower photos they contain.

**in predict.py**

In the terminal, predict.py is best called with:
```
cd python predict.py ./flowers/test/11/image_03098.jpg ./save_directory/checkpoint.pth -g
```

with ```cd``` being the directory in which predict.py is in.

with ```./flowers/test/11/image_03098.jpg``` being the path to the image we'd like to predict.

with ```./save_directory/checkpoint.pth``` being the path to the checkpoint we saved in train.py.

with ```-g``` representing the GPU connection so that everything runs faster.

Please note
---
As this project was submitted and grated, to avoid direct duplication of the code by future students I won't include the name of the Nanodegree program or the actual name of the project, as stated in the syllabus.

FlowerImageClassifier

FlowerImageClassifier 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

FlowerImageClassifier

FlowerImageClassifier 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README