Skip to content
GitHub Explorer

RaoulLuque/ImageRecognitionFromScratch

View on GitHub ↗Related repositories →

Deep neural network implementation from scratch using python and numpy to classify MNIST dataset

active 2024-11-25 → 2025-01-17 (UTC)

Complete coverage26,512 / 26,512 hourly files (100%) · 2 absent upstream2023-08-15 → 2026-08-23 (UTC)
Events
109
Pushes
55
Pull requests
0
Issues
40
Stars
1
Forks
0

Activity over time

Daily event counts in the loaded window

Line chart, 54 days from 2024-11-25 to 2025-01-17. Pushes: 55 total, peak 13 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 40 total, peak 11 in a day. Comments: 11 total, peak 5 in a day. Stars: 1 total, peak 1 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
RaoulLuque10655011

Recent activity

Latest issues, pull requests and releases

  • Issue comment#19RaoulLuque2024-12-11 10:40
    [Optional] Try out different learning rate schedulers
  • Issue#19RaoulLuque2024-12-11 10:40
    [Optional] Try out different learning rate schedulers
  • Issue comment#25RaoulLuque2024-12-11 10:38
    Implement automatic creation of necessary proof for hand-in upon each run of model
  • Issue#25RaoulLuque2024-12-11 10:38
    Implement automatic creation of necessary proof for hand-in upon each run of model
  • Issue comment#28RaoulLuque2024-12-11 10:36
    [Optimization] Test out CuPy to use GPU instead of CPU for computations
  • Issue#28RaoulLuque2024-12-11 10:36
    [Optimization] Test out CuPy to use GPU instead of CPU for computations
  • Issue comment#13RaoulLuque2024-12-11 10:34
    Achieve <0.5% error rate
  • Issue#13RaoulLuque2024-12-11 10:34
    Achieve <0.5% error rate
  • Issue comment#12RaoulLuque2024-12-11 10:34
    Achieve <1% error rate
  • Issue#12RaoulLuque2024-12-11 10:34
    Achieve <1% error rate
  • Issue#42RaoulLuque2024-12-11 10:33
    Add to README that models can also be loaded
  • Issue#42RaoulLuque2024-12-02 12:08
    Add to README that models can also be loaded
  • Issue#41RaoulLuque2024-12-02 09:41
    Rename D_batch_size occurrences to N_batch_size
  • Issue#39RaoulLuque2024-12-01 19:33
    Add loading bar showing epoch progress
  • Issue#40RaoulLuque2024-12-01 12:59
    Move activation_functions.py to /add_ons
  • Issue#1RaoulLuque2024-12-01 12:16
    Add tests
  • Issue#16RaoulLuque2024-11-30 10:00
    [Feature] Implement batch normalization layers
  • Issue comment#30RaoulLuque2024-11-29 13:21
    [Optimization] Make forward propagation actually use the batches and propagate all the batch at once
  • Issue#27RaoulLuque2024-11-29 13:19
    Revise weight initialization
  • Issue#39RaoulLuque2024-11-29 01:18
    Add loading bar showing epoch progress
  • Issue#38RaoulLuque2024-11-28 19:02
    Implement (max) pooling 2D layer
  • Issue#38RaoulLuque2024-11-28 19:02
    Implement (max) pooling 2D layer
  • Issue#4RaoulLuque2024-11-28 16:18
    Implement 2D convolution layers
  • Issue comment#31RaoulLuque2024-11-28 13:32
    Achieve <2% error rate
  • Issue#31RaoulLuque2024-11-28 13:32
    Achieve <2% error rate

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 1 stars here means stars gained during the window, not the repo's star count.