文件列表:

.devcontainer/ (0, 2023-10-11)
.devcontainer/devcontainer.json (326, 2023-10-11)
.env.example (55, 2023-10-11)
.vscode/ (0, 2023-10-11)
.vscode/settings.json (137, 2023-10-11)
assets/ (0, 2023-10-11)
assets/prediction.png (208100, 2023-10-11)
assets/trend.png (481380, 2023-10-11)
data/ (0, 2023-10-11)
data/interim/ (0, 2023-10-11)
data/processed/ (0, 2023-10-11)
data/raw/ (0, 2023-10-11)
data/raw/main_cpu-test.csv (1517, 2023-10-11)
data/raw/main_cpu-train.csv (15017, 2023-10-11)
models/ (0, 2023-10-11)
models/arima_default.sav (488372, 2023-10-11)
requirements.txt (250, 2023-10-11)
src/ (0, 2023-10-11)
src/app.py (69, 2023-10-11)
src/explore.ipynb (612841, 2023-10-11)
src/utils.py (261, 2023-10-11)

# Time Series - Analysis This project goal is to understand how to analysis a Time Series prediction, observing the trends, seasonality and the variability of the data. Time series is a predictive model that can be quite complex, so for this project I used a Server CPU Time uptime. ## Key Takes Time Series are great to understand data that is set upon a timeline, and using ARIMA model we can try to get the future outcomes of the behaviour of our data. Still, analyzing the time series will also help understand your data in depth and immediately take observations to better know what decisions you can make. But not always they'll be good, as it was in my dataset. If we look at the trend it looks that we will be able to predict properly: But in reality after ARIMA, the data was so poor, that we couldn't predict upcoming uptimes of the CPU: Time Series are very complex models, but understanding how it works and with good data work, they can help predict and solve a lot of problems.

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：