facebookresearch / GDT
We present a framework for training multi-modal deep learning models on unlabelled video data by forcing the network to learn invariances to transformations applied to both the audio and video streams.
Summary
  |
|||
| Main Code: 5,688 LOC (21 files) = PY (96%) + YML (3%) Secondary code: Test: 0 LOC (0); Generated: 0 LOC (0); Build & Deploy: 487 LOC (5); Other: 232 LOC (3); |
|||
| Duplication: 17% | |||
| File Size: 0% long (>1000 LOC), 16% short (<= 200 LOC) | |||
| Unit Size: 25% long (>100 LOC), 45% short (<= 10 LOC) | |||
| Conditional Complexity: 10% complex (McCabe index > 50), 56% simple (McCabe index <= 5) | |||
|
|
Logical Component Decomposition: primary (3 components) | ||
|
|
6 months old
|
||
|
|
0% of code updated more than 50 times Also see temporal dependencies for files frequently changed in same commits. |
||
| Goals: Keep the system simple and easy to change (4) | |||
|
Features of interest:
TODOs
1 file |
|||
Latest commit date: 2021-08-29
|
0
commits
(30 days)
0
contributors
(30 days) |
|