bird_species_guide

KO | EN

gitlite — search

#python #java #python3 #arduino #golang #machine-learning #rust #html #flask #javascript #seismology #nodejs

bird_species_guide

★ 16 Open GitHub ↗

Telegran bot for classification bird species by image

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Bird species telegram bot

## Dataset preparation 
[code for collecting dataset ](https://github.com/LadaChernenko/bird_species_guide/tree/main/collecting_dataset)

A dataset was collected for neural network training. It was based on the latin names of birds from european part of Russia and included 350 species in total.

At least 100 images per class.

[Bird dataset from Google Drive](https://drive.google.com/file/d/1cUw8hBoF0PEYHbLMfWB2x-k2lV1YMPb5/view?usp=sharing)

- the latin names of the birds were taken from [here](https://www.ebirds.ru/russia/index.html)
- images were collected from google images by latin names using libraries  selenium; urllib; user_agent; logging 
- dataset was manually cleared

___
## Image classification

[image_classification](https://github.com/LadaChernenko/bird_species_guide/blob/main/bird_classification_colab/bird_classification.ipynb)

### Architecture

As a model pretrained  [EfficientNet-b0](https://arxiv.org/pdf/1905.11946.pdf) was used

350 classes on the last layer

![Architecture](https://github.com/LadaChernenko/bird_species_guide/blob/main/img/Architecture-of-EfficientNet-B0-with-MBConv-as-Basic-building-blocks.png?raw=true )
*EfficientNet-b0*

### Results
![img_classification](https://github.com/LadaChernenko/bird_species_guide/blob/main/img/classification_pred.png?raw=true)

Loss function **CrossEntropyLoss**

dataset part |  CrossEntropyLoss  | Accuracy  
--- | --- | ---
train | 0.665 | 82.559685%
validation | 0.693 | 81.350510%

___
## Telegram bot
@bird_species_bot  [link](https://t.me/bird_species_bot)

[Telegram bot code](https://github.com/LadaChernenko/bird_species_guide/tree/main/telegram_bot)
### Libraries:
- aiogram==2.17.1
- torch==1.10.0+cpu
- torchvision==0.11.1+cpu
- efficientnet-pytorch==0.7.1
- wikipedia==1.4.0

[requirements.txt](https://github.com/LadaChernenko/bird_species_guide/blob/main/telegram_bot/requirements.txt)
### How it works:
- 1 Telegram bot **bot.py** gets the image 
- 2 Saves to folder 
- 3 **app.py** gets img
  - 3.1 Transform
  - 3.2 Pretrained efficientnet-b0 predicts labels and probability
- 4 Bot gives to the user  top-3 predicted classes and probability
- 5 Deletes the image
- 6 **wiki_parser.py** searches wikipedia page for latin name of species and give it to user

bird_species_guide

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README