Skip to content

Commit a6c6405

Browse files
Costaki33Costaki33
committed
updated README
1 parent c706234 commit a6c6405

File tree

1 file changed

+52
-49
lines changed

1 file changed

+52
-49
lines changed

‎eqcctpro/README.md‎

Lines changed: 52 additions & 49 deletions
Original file line numberDiff line numberDiff line change
@@ -8,6 +8,7 @@ EQCCTPro is a high-performace seismic event detection and processing framework t
88
- Includes tools for evaluating system performance for optimal usecase configurations
99
- Automatic selection of best-usecase configurations
1010
- Efficient handling of large-scale seismic data
11+
- Dataset creation with necessary data structure using custom FDSNWS connection
1112

1213
# **Installation Guide**
1314
There are **two installation methods** for EQCCTPro:
@@ -17,7 +18,6 @@ There are **two installation methods** for EQCCTPro:
1718

1819
It is **highly recommended** that first-time users pull the `EQCCTPro` folder, which includes sample waveform data and code to help get acquainted with **EQCCTPro**.
1920

20-
---
2121

2222
## **Method 1: Install EQCCTPro (No Sample Data)**
2323
This method installs only the EQCCTPro package **without** the sample waveform data.
@@ -78,13 +78,14 @@ conda activate eqcctpro
7878
For additional package updates, continue to check either this repository or visit the **EQCCTPro PyPI page**:
7979
🔗 [EQCCTPro on PyPI](https://pypi.org/project/eqcctpro/)
8080

81-
---
8281

83-
### **Using Sample Waveform Data**
82+
83+
# **Using Sample Waveform Data**
8484
To understand how **EQCCTPro** works, it is **highly recommended** to use provided sample seismic waveform data as the data source when testing the package.
8585

8686
1-minute long sample seismic waveforms from 229 TexNet stations have been provided in the repository under the `230_stations_1_min_dt.zip` file.
8787

88+
## **Downloading and Understanding the Provided Waveform Data**
8889
### **Step 1: Unzip the Sample Wavefrom Data**
8990
After downloading the `.zip` file through the GitHub methods above, run:
9091
```sh
@@ -111,10 +112,54 @@ TX.PB35.00.HH2__20241215T115800Z__20241215T120100Z.mseed
111112
```
112113
EQCCT (i.e., the ML model) requires at least one pose per station for detection, but using multiple poses enhances P and S wave directionality.
113114

114-
You have successfully installed EQCCTPro and set up the required sample waveform dataset for testing.
115+
## **Dataset creation using a FDSNWS connection**
116+
It is now possible to create the necesary dataset structure with your own data using the provided script: 'create_dataset.py'.
117+
The script:
118+
1. Retrieves waveform data from a user defined FDSNWS webservice.
119+
2. Selects data according to network, station, channel and location codes.
120+
3. Has the option for defining time chunks according to the users requirements.
121+
4. Automatically downloads and creates the required folder structure for eqcctpro.
122+
5. Optionally denoises the data using seisbench as backend.
115123

116-
---
117-
### **Using EQCCTPro**
124+
An example is provided below:
125+
```sh
126+
python create_dataset.py -h
127+
```
128+
129+
output:
130+
```sh
131+
usage: create_dataset.py [-h] [--start START] [--end END] [--networks NETWORKS] [--stations STATIONS] [--locations LOCATIONS]
132+
[--channels CHANNELS] [--host HOST] [--output OUTPUT] [--chunk CHUNK] [--denoise]
133+
134+
Download FDSN waveforms in equal-time chunks.
135+
136+
options:
137+
-h, --help show this help message and exit
138+
--start START Start time, e.g. 2024-12-03T00:00:00Z
139+
--end END End time, e.g. 2024-12-03T02:00:00Z
140+
--networks NETWORKS Comma-separated network codes or *
141+
--stations STATIONS Comma-separated station codes or *
142+
--locations LOCATIONS
143+
Comma-separated location codes or *
144+
--channels CHANNELS Comma-separated channel codes or *
145+
--host HOST FDSNWS base URL
146+
--output OUTPUT Base output directory
147+
--chunk CHUNK Chunk size in minutes. Splits start■end into N windows.
148+
--denoise If set, apply seisbench.DeepDenoiser to each chunk.
149+
```
150+
151+
An example to download waveforms from a local fdsnws server is given below:
152+
```sh
153+
python create_dataset.py --start 2025-10-31T00:00 --end 2025-10-31T04:00 --networks TX --stations "*" --locations "*" --channels HH?,HN? --host http://localhost:8080 --output waveforms_directory --chunk 60
154+
```
155+
156+
The resulting output folder contains the data to be processed by EQCCTPro.
157+
Note: Please make sure that you set a consistant chunk size in the download script, as well as in EQCCTPro itself to avoid issues.
158+
E.G.: If you set a time chunk of 20 minutes in the download script, then also use 20 minutes as chunk size when calling EQCCTPro.
159+
This is so that data won't be processed eroniusly.
160+
161+
162+
# **Using EQCCTPro**
118163
There are three main capabilities of EQCCTPro:
119164
1. **Process mSEED data from singular or multiple seismic stations using either CPUs or GPUs**
120165
2. **Evaluate your system to identify the optimal parallelization configurations needed to get the minimum runtime performance out of your system**
@@ -278,7 +323,7 @@ eqcct_runner.run_eqcctpro()
278323

279324
---
280325

281-
### **Evaluating Your Systems Runtime Performance Capabilites**
326+
### **Evaluating Your System's Runtime Performance Capabilites**
282327
To evaluate your system’s runtime performance capabilites for both your CPU(s) and GPU(s), the **EvaluateSystem** class allows you to autonomously evaluate your system:
283328

284329

@@ -492,48 +537,6 @@ For **OptimalGPUConfigurationFinder.find_optimal_for()**, the function requires
492537
## **Configuration**
493538
The `environment.yml` file specifies the dependencies required to run EQCCTPro. Ensure you have the correct versions installed by using the provided conda environment setup.
494539

495-
##Dataset creation
496-
It is now possible to create the necesary dataset structure with your own data using the provided script 'create_dataset.py'.
497-
The script:
498-
1. Retrieves waveform data from a user defined FDSNWS webservice.
499-
2. Selects data according to network, station, channel and location codes.
500-
3. Has the option for defining time chunks according to the users requirements.
501-
4. Automatically downloads and creates the required folder structure for eqcctpro.
502-
5. Optionally denoises the data using seisbench as backend.
503-
An example is provided below
504-
```sh
505-
python create_dataset.py -h
506-
```
507-
output:
508-
````
509-
usage: create_dataset.py [-h] [--start START] [--end END] [--networks NETWORKS] [--stations STATIONS] [--locations LOCATIONS]
510-
[--channels CHANNELS] [--host HOST] [--output OUTPUT] [--chunk CHUNK] [--denoise]
511-
512-
Download FDSN waveforms in equal-time chunks.
513-
514-
options:
515-
-h, --help show this help message and exit
516-
--start START Start time, e.g. 2024-12-03T00:00:00Z
517-
--end END End time, e.g. 2024-12-03T02:00:00Z
518-
--networks NETWORKS Comma-separated network codes or *
519-
--stations STATIONS Comma-separated station codes or *
520-
--locations LOCATIONS
521-
Comma-separated location codes or *
522-
--channels CHANNELS Comma-separated channel codes or *
523-
--host HOST FDSNWS base URL
524-
--output OUTPUT Base output directory
525-
--chunk CHUNK Chunk size in minutes. Splits start■end into N windows.
526-
--denoise If set, apply seisbench.DeepDenoiser to each chunk.
527-
```
528-
An example to download waveforms from a local fdsnws server is given below:
529-
```sh
530-
python create_dataset.py --start 2025-10-31T00:00 --end 2025-10-31T04:00 --networks TX --stations "*" --locations "*" --channels HH?,HN? --host http://localhost:8080 --output waveforms_directory --chunk 60
531-
```
532-
The resulting output folder contains the data to be processed by Eqcctpro.
533-
Note: Please make sure that you set a consistant chunk size in the download script, as well as in eqcctpro itself to avoid issues.
534-
E.G.: If you set a time chunk of 20 minutes in the download script, then also use 20 minutes as chunk size when calling eqcctpro.
535-
This is so that data won't be processed eroniusly.
536-
537540

538541
## **License**
539542
EQCCTPro is provided under an open-source license. See LICENSE for details.

0 commit comments

Comments
 (0)