-
# Deep learning is EXPENSIVE
e.g.
train ResNet50 with ImageNet dataset for 80 epochs
80 * 1.3M images * 7.7B ops per img
# Solution?
- **Data Parallelism (large batch training)**
![image](http…
-
## Bug report
**Required Info:**
- Operating System:
- Ubuntu 20.04 Focal
- ROS2 Version:
- Galactic
- Version or commit hash:
- ros-galactic-navigation2 1.0.12-1focal.20220730.081413…
-
Replicate QGIS steps into python commands for easy "replicate-ability"
- ~Define intersections (this was previously done using LION line intersection to determine point)~
- ~Obtain signalized inte…
-
```python
import torch
import pandas as pd
import random
from pathlib import Path
from torch.utils.data import DataLoader
from typing import Union, Optional
from pprint import pprint
from om…
-
Some our lessons use the framing "let's... " others use "I can..." Standardize to let's.
- [x] bar and pie
- [x] boxplot
- [x] checking your work
- [x] choosing your dataset
- [x] codap bar
-…
-
**This should not be completed until after**
- [x] #2002
**_When this issue is complete, read and close #1679_ **
RT
- [x] advanced-displays
- [x] bar-and-pie-charts
- [x] checking-your…
-
There must be some strong correlation in read depth along the sequence, which drops off after `n` bases, where `n` is some function of the mean read length. E.g. two sites that are adjacent along the …
-
Right now each event has to be explicitly added to the event specification (e.g. via for loop). It would be helpful to have high-level generators such that you'd only have to specify something along t…
-
We plan to write a short summary for every implemented method in our wiki space.
Please follow the template below and write the draft here.
**Paper Title**
[![paper](https://img.shields.io/badg…
-