Spaces:
Running
Running
File size: 2,234 Bytes
7a982d6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 |
---
title: DeepDenoiser
emoji: 🌊
colorFrom: purple
colorTo: blue
sdk: docker
pinned: false
---
# DeepDenoiser: Seismic Signal Denoising and Decomposition Using Deep Neural Networks
[](https://ai4eps.github.io/DeepDenoiser)
## 1. Install [miniconda](https://docs.conda.io/en/latest/miniconda.html) and requirements
- Download DeepDenoiser repository
```bash
git clone https://github.com/wayneweiqiang/DeeoDenoiser.git
cd DeepDenoiser
```
- Install to default environment
```bash
conda env update -f=env.yml -n base
```
- Install to "deepdenoiser" virtual envirionment
```bash
conda env create -f env.yml
conda activate deepdenoiser
```
## 2. Pre-trained model
Located in directory: **model/190614-104802**
## 3. Related papers
- Zhu, Weiqiang, S. Mostafa Mousavi, and Gregory C. Beroza. "Seismic Signal Denoising and Decomposition Using Deep Neural Networks." arXiv preprint arXiv:1811.02695 (2018).
## 4. Interactive example
See details in the [notebook](https://github.com/wayneweiqiang/DeepDenoiser/blob/master/docs/example_interactive.ipynb): [example_interactive.ipynb](example_interactive.ipynb)
## 5. Batch prediction
See details in the [notebook](https://github.com/wayneweiqiang/DeepDenoiser/blob/master/docs/example_batch_prediction.ipynb): [example_batch_prediction.ipynb](example_batch_prediction.ipynb)
## 6. Train
### Data format
Required: two csv files for signal and noise, corresponding directories of the npz files.
The csv file contains four columns: "fname", "itp", "channels"
The npz file contains four variable: "data", "itp", "channels"
The shape of "data" variables has a shape of 9001 x 3
The variables "itp" is the data points of first P arrival times.
Note: In the demo data, for simplicity we use the waveform before itp as noise samples, so the train_noise_list is same as train_signal_list here.
~~~bash
python deepdenoiser/train.py --mode=train --train_signal_dir=./Dataset/train --train_signal_list=./Dataset/train.csv --train_noise_dir=./Dataset/train --train_noise_list=./Dataset/train.csv --batch_size=20
~~~
Please let us know of any bugs found in the code. Suggestions and collaborations are welcomed
|