### Start training PhaseNet model

Source: https://ai4eps.github.io/PhaseNet

Initiates training for the PhaseNet model using a pre-trained model. Specify training data, epochs, and batch size. Plotting is enabled for monitoring.

```python
python phasenet/train.py  --model_dir=model/190703-214543/ --train_dir=test_data/npz --train_list=test_data/npz.csv  --plot_figure --epochs=10 --batch_size=10

```

--------------------------------

### Install Dependencies to Default Environment

Source: https://ai4eps.github.io/PhaseNet

Update the default Conda environment with the dependencies specified in env.yaml.

```bash
conda env update -f=env.yaml -n base
```

--------------------------------

### Get Phase Picks and Prediction Time Series from PhaseNet API

Source: https://ai4eps.github.io/PhaseNet/example_fastapi

Requests both phase picks and prediction probabilities (time series) from the PhaseNet API's /predict_prob endpoint. It then visualizes the Z-component waveform and the predicted P and S phase probabilities.

```python
resp = requests.post(f'{PHASENET_API_URL}/predict_prob', json=req)
print(resp)
picks, preds = resp.json() 
preds = np.array(preds)
print('Picks', picks)

plt.figure()
plt.subplot(211)
plt.plot(data[:,-1], 'k', label="Z")
plt.subplot(212)
plt.plot(preds[0, :, 0, 1], label="P")
plt.plot(preds[0, :, 0, 2], label="S")
plt.legend()
plt.show();

```

--------------------------------

### Predict with mseed format

Source: https://ai4eps.github.io/PhaseNet

Performs batch prediction on data in mseed format. Use --amplitude to get amplitude values. The batch size is set to 1 due to variable data lengths.

```python
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/mseed.csv --data_dir=test_data/mseed --format=mseed --amplitude --response_xml=test_data/stations.xml --batch_size=1 --sampling_rate=100

```

```python
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/mseed2.csv --data_dir=test_data/mseed --format=mseed --amplitude --response_xml=test_data/stations.xml --batch_size=1 --sampling_rate=100

```

--------------------------------

### Read Seismic Stream

Source: https://ai4eps.github.io/PhaseNet/example_gradio

Reads seismic waveform data into an ObsPy stream object and plots it. Ensure ObsPy is installed and data is accessible.

```python
import obspy
stream = obspy.read()
stream.plot();
```

--------------------------------

### Download and Unzip Demo Data

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Navigate to the PhaseNet directory and download the test data zip file, then unzip it to prepare for batch prediction.

```bash
cd PhaseNet
wget https://github.com/wayneweiqiang/PhaseNet/releases/download/test_data/test_data.zip
unzip test_data.zip
```

--------------------------------

### Create and Activate 'phasenet' Environment for Mac ARM

Source: https://ai4eps.github.io/PhaseNet

For Mac ARM chips, create a Conda environment using env_mac.yaml and activate it.

```bash
conda env create -f env_mac.yaml
conda activate phasenet
```

--------------------------------

### Create and Activate 'phasenet' Virtual Environment

Source: https://ai4eps.github.io/PhaseNet

Create a new Conda virtual environment named 'phasenet' and activate it.

```bash
conda env create -f env.yaml
conda activate phasenet
```

--------------------------------

### Predict Phase Picks with PhaseNet

Source: https://ai4eps.github.io/PhaseNet/example_gradio

Initializes the Gradio client with the PhaseNet model and uses it to predict P/S-phase picks from the seismic data. Requires the Gradio client and the PhaseNet model to be accessible.

```python
model = "ai4eps/phasenet"
# model = "http://127.0.0.1:7860"
client = Client(model, serialize=True)
_, picks_csv, picks_json = client.predict(["data.mseed"], json.dumps(data.tolist()), json.dumps([data_id]))
```

--------------------------------

### Import Libraries and Set Project Root

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Imports necessary libraries (pandas, json, os) and defines the project root directory for accessing results.

```python
import pandas as pd
import json
import os
PROJECT_ROOT = os.path.realpath(os.path.join(os.path.abspath(''), ".."))
```

--------------------------------

### Load and Prepare Seismic Data with ObsPy

Source: https://ai4eps.github.io/PhaseNet/example_fastapi

Reads seismic data using ObsPy and prepares it for PhaseNet. It sorts the stream, asserts three components, extracts the waveform data, and formats metadata like station ID and timestamp.

```python
import obspy
stream = obspy.read()
stream.plot();

```

```python
## Extract 3-component data
stream = stream.sort()
assert(len(stream) == 3)
data = []
for trace in stream:
    data.append(trace.data)
data = np.array(data).T
assert(data.shape[-1] == 3)

data_id = stream[0].get_id()[:-1]
timestamp = stream[0].stats.starttime.datetime.strftime("%Y-%m-%dT%H:%M:%S.%f")[:-3]

```

--------------------------------

### Import Libraries

Source: https://ai4eps.github.io/PhaseNet/example_gradio

Imports necessary Python libraries for data handling, client interaction, and visualization.

```python
from gradio_client import Client
import obspy
import numpy as np
import json
import pandas as pd
import matplotlib.pyplot as plt
import json
```

--------------------------------

### Clone PhaseNet Repository

Source: https://ai4eps.github.io/PhaseNet

Clone the PhaseNet repository from GitHub to your local machine.

```bash
git clone https://github.com/wayneweiqiang/PhaseNet.git
cd PhaseNet
```

--------------------------------

### Download Test Data

Source: https://ai4eps.github.io/PhaseNet

Downloads and unzips the test data required for prediction and training.

```bash
wget https://github.com/wayneweiqiang/PhaseNet/releases/download/test_data/test_data.zip
unzip test_data.zip

```

--------------------------------

### Configure PhaseNet API URL

Source: https://ai4eps.github.io/PhaseNet/example_fastapi

Sets the URL for the PhaseNet API endpoint. Several options are commented out, with the active URL pointing to a hosted service.

```python
# PHASENET_API_URL = "http://127.0.0.1:8000"
# PHASENET_API_URL = "http://phasenet.quakeflow.com" ## gcp
# PHASENET_API_URL = "http://test.quakeflow.com:8001" ## local machine
PHASENET_API_URL = "https://ai4eps-eqnet.hf.space"
# PHASENET_API_URL = "http://131.215.74.195:8001" ## local machine

```

--------------------------------

### PhaseNet Predict Script Optional Arguments

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

This displays the available optional arguments for the `predict.py` script, including data input, model, output, and prediction parameters.

```bash
usage: predict.py [-h] [--batch_size BATCH_SIZE]
                  [--model_dir MODEL_DIR]
                  [--data_dir DATA_DIR]
                  [--data_list DATA_LIST]
                  [--hdf5_file HDF5_FILE]
                  [--hdf5_group HDF5_GROUP]
                  [--result_dir RESULT_DIR]
                  [--result_fname RESULT_FNAME]
                  [--min_p_prob MIN_P_PROB]
                  [--min_s_prob MIN_S_PROB]
                  [--mpd MPD]
                  [--amplitude]
                  [--format FORMAT]
                  [--s3_url S3_URL]
                  [--stations STATIONS]
                  [--plot_figure]
                  [--save_prob]

optional arguments:
  -h, --help            show this help message and exit
  --batch_size BATCH_SIZE
                        batch size
  --model_dir MODEL_DIR
                        Checkpoint directory (default: None)
  --data_dir DATA_DIR   Input file directory
  --data_list DATA_LIST
                        Input csv file
  --hdf5_file HDF5_FILE
                        Input hdf5 file
  --hdf5_group HDF5_GROUP
                        data group name in hdf5 file
  --result_dir RESULT_DIR
                        Output directory
  --result_fname RESULT_FNAME
                        Output file
  --min_p_prob MIN_P_PROB
                        Probability threshold for P pick
  --min_s_prob MIN_S_PROB
                        Probability threshold for S pick
  --mpd MPD             Minimum peak distance
  --amplitude           if return amplitude value
  --format FORMAT       input format
  --stations STATIONS   seismic station info
  --plot_figure         If plot figure for test
  --save_prob           If save result for test
```

--------------------------------

### Import Libraries for PhaseNet

Source: https://ai4eps.github.io/PhaseNet/example_fastapi

Imports necessary Python libraries for data manipulation, plotting, and API requests. Ensures the PhaseNet module is accessible.

```python
import os, sys
import numpy as np
import matplotlib.pyplot as plt
import obspy
import requests
sys.path.insert(0, os.path.abspath("../"))

```

--------------------------------

### Visualize Waveforms and Picks

Source: https://ai4eps.github.io/PhaseNet/example_gradio

Generates plots for each seismic component, overlaying the waveform data with predicted P and S phase picks. The alpha value of the vertical lines represents the phase score.

```python
fig, ax = plt.subplots(len(stream), 1, figsize=(10, 10))
for i, tr in enumerate(stream):
    ax[i].plot(tr.times(), tr.data, label=tr.stats.channel, c="k")
    for _, pick in picks.iterrows():
        c = "blue" if pick["phase_type"] == "P" else "red"
        label = pick["phase_type"] if i == 0 else None
        ax[i].axvline((pick["phase_time"]-tr.stats.starttime.datetime).total_seconds(), c=c, label=label, alpha=pick["phase_score"])
    ax[i].legend()
```

--------------------------------

### Predict with seismic array format

Source: https://ai4eps.github.io/PhaseNet

Performs batch prediction on seismic array data, used by QuakeFlow. Requires station information.

```python
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/mseed_array.csv --data_dir=test_data/mseed_array --stations=test_data/stations.json  --format=mseed_array --amplitude

```

--------------------------------

### Run Batch Prediction for mseed format

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Execute the PhaseNet prediction script for data in mseed format. Ensure the model and data paths are correctly specified. The `--plot_figure` argument can be removed for large datasets to improve performance.

```bash
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/mseed.csv --data_dir=test_data/mseed --format=mseed --plot_figure
```

--------------------------------

### Run Batch Prediction for seismic array (mseed_array format)

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Execute the PhaseNet prediction script for seismic array data in mseed_array format, specifying station information. The `--amplitude` flag is used for this format. The `--plot_figure` argument can be removed for large datasets to improve performance.

```bash
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/mseed_array.csv --data_dir=test_data/mseed_array --stations=test_data/stations.json  --format=mseed_array --amplitude
```

```bash
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/mseed2.csv --data_dir=test_data/mseed --stations=test_data/stations.json  --format=mseed_array --amplitude
```

--------------------------------

### Predict P/S-Phase Picks using PhaseNet API

Source: https://ai4eps.github.io/PhaseNet/example_fastapi

Sends prepared seismic waveform data to the PhaseNet API's /predict endpoint to obtain P and S phase picks. It prints the HTTP response and the JSON response containing the picks.

```python
req = {"id": [data_id],
       "timestamp": [timestamp],
       "vec": [data.tolist()]}

resp = requests.post(f'{PHASENET_API_URL}/predict', json=req)
print(resp)
print('Picks', resp.json())

```

--------------------------------

### Run Batch Prediction for hdf5 format

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Execute the PhaseNet prediction script for data in hdf5 format. Ensure the model and data paths are correctly specified. The `--plot_figure` argument can be removed for large datasets to improve performance.

```bash
python phasenet/predict.py --model=model/190703-214543 --hdf5_file=test_data/data.h5 --hdf5_group=data --format=hdf5 --plot_figure
```

--------------------------------

### Predict with hdf5 format

Source: https://ai4eps.github.io/PhaseNet

Performs batch prediction on data in hdf5 format. Specify the hdf5 file and group name.

```python
python phasenet/predict.py --model=model/190703-214543 --hdf5_file=test_data/data.h5 --hdf5_group=data --format=hdf5

```

--------------------------------

### Predict with sac format

Source: https://ai4eps.github.io/PhaseNet

Performs batch prediction on data in sac format.

```python
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/sac.csv --data_dir=test_data/sac --format=sac --batch_size=1

```

--------------------------------

### Read JSON Picks

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Loads seismic picks from a JSON file into a Python list of dictionaries. This format is often more structured for direct use.

```python
with open(os.path.join(PROJECT_ROOT, "results/picks.json")) as fp:
    picks_json = json.load(fp)  
print(picks_json[1])
print(picks_json[0])
```

--------------------------------

### Run Batch Prediction for sac format

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Execute the PhaseNet prediction script for data in sac format. Ensure the model and data paths are correctly specified. The `--plot_figure` argument can be removed for large datasets to improve performance.

```bash
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/sac.csv --data_dir=test_data/sac --format=sac --plot_figure
```

--------------------------------

### Write Stream to MSEED

Source: https://ai4eps.github.io/PhaseNet/example_gradio

Saves the seismic stream data to a MiniSEED file. This is useful for persistent storage or sharing.

```python
stream.write("data.mseed", format="MSEED")
```

--------------------------------

### Run Batch Prediction for numpy format

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Execute the PhaseNet prediction script for data in numpy format. Ensure the model and data paths are correctly specified. The `--plot_figure` argument can be removed for large datasets to improve performance.

```bash
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/npz.csv --data_dir=test_data/npz --format=numpy --plot_figure
```

--------------------------------

### Read and Process CSV Picks

Source: https://ai4eps.github.io/PhaseNet/example_batch_prediction

Reads picks from a tab-separated CSV file and processes the index and probability columns, which are stored as strings representing lists.

```python
picks_csv = pd.read_csv(os.path.join(PROJECT_ROOT, "results/picks.csv"), sep="\t")
picks_csv.loc[:, 'p_idx'] = picks_csv["p_idx"].apply(lambda x: x.strip("[]").split(","))
picks_csv.loc[:, 'p_prob'] = picks_csv["p_prob"].apply(lambda x: x.strip("[]").split(","))
picks_csv.loc[:, 's_idx'] = picks_csv["s_idx"].apply(lambda x: x.strip("[]").split(","))
picks_csv.loc[:, 's_prob'] = picks_csv["s_prob"].apply(lambda x: x.strip("[]").split(","))
print(picks_csv.iloc[1])
print(picks_csv.iloc[0])
```

--------------------------------

### Predict with numpy format

Source: https://ai4eps.github.io/PhaseNet

Performs batch prediction on data in numpy format.

```python
python phasenet/predict.py --model=model/190703-214543 --data_list=test_data/npz.csv --data_dir=test_data/npz --format=numpy

```

--------------------------------

### Extract 3-Component Data

Source: https://ai4eps.github.io/PhaseNet/example_gradio

Sorts the seismic stream, asserts it contains exactly three components, and extracts the waveform data. It also prepares a data ID and timestamp.

```python
stream = stream.sort()
assert(len(stream) == 3)
data = []
for trace in stream:
    data.append(trace.data)
data = np.array(data).T
assert(data.shape[-1] == 3)

data_id = stream[0].get_id()[:-1]
timestamp = stream[0].stats.starttime.datetime.strftime("%Y-%m-%dT%H:%M:%S.%f")[:-3]
```

--------------------------------

### Load Prediction Results

Source: https://ai4eps.github.io/PhaseNet/example_gradio

Parses the JSON output from the PhaseNet prediction into a pandas DataFrame for easier manipulation and analysis.

```python
picks = pd.read_json(picks_json)
# picks = pd.read_csv(csv, parse_dates=["phase_time"])
# print(picks)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.