replace interpolate with resample

src/noisepy/seis/noise_module.py

@@ -154,7 +154,8 @@ def preprocess_raw(
     # make downsampling if needed
     if abs(samp_freq - sps) > 1e-4:
         # downsampling here
-        st.interpolate(samp_freq, method="weighted_average_slopes")
+        # st.interpolate(samp_freq, method="weighted_average_slopes")
+        st.resample(samp_freq)
         delta = st[0].stats.delta
 
         # when starttimes are between sampling points

tests/test_monitoring_utils.py

@@ -5,10 +5,8 @@
     dtw_dvv,
     mwcs_dvv,
     stretching,
-    stretching_vect,
-    wtdtw_allfreq,
+    wcc_dvv,
     wtdtw_dvv,
-    wts_allfreq,
     wts_dvv,
     wxs_dvv,
 )
@@ -17,15 +15,15 @@
     (
         np.sin(np.arange(100) / 10),
         np.sin(np.arange(100) / 10.1),
-        {"twin": [0, 100], "freq": [0.01, 0.49], "dt": 1},
+        {"twin": [0, 100], "t": np.arange(101), "freq": [0.01, 0.49], "dt": 1},
     )
 ]
 
 data2 = [
     (
         np.sin(np.arange(100) / 10),
         np.sin(np.arange(100) / 10.1),
-        {"twin": [0, 99], "t": np.array([0, 100]), "freq": [0.01, 0.49], "dt": 1},
+        {"twin": [0, 99], "t": np.array(np.arange(100)), "freq": [0.01, 0.49], "dt": 1},
     )
 ]
 
@@ -41,13 +39,11 @@ def test_stretching(d1: np.ndarray, d2: np.ndarray, param: dict):
 
 
 @pytest.mark.parametrize("d1,d2,param", data)
-def test_stretching_vect(d1: np.ndarray, d2: np.ndarray, param: dict):
-    dv_range = 0.05
-    nbtrial = 50
-    dv, error, cc, _ = stretching_vect(d1, d2, dv_range, nbtrial, param)
-    assert np.isclose(dv, -1, rtol=0.2)
-    assert np.isclose(error, 0.0, atol=1e-2)
-    assert np.isclose(cc, 1.0, atol=1e-1)
+def test_wcc_dvv(d1: np.ndarray, d2: np.ndarray, param: dict):
+    moving_window_length = 10.0
+    window_slide_step = 1.0
+    dv, error = wcc_dvv(d1, d2, moving_window_length, window_slide_step, param)
+    assert dv < 0
 
 
 @pytest.mark.parametrize("d1,d2,param", data)
@@ -67,7 +63,7 @@ def test_mwcs_dvv(d1: np.ndarray, d2: np.ndarray, param: dict):
     assert dv < 0
 
 
-@pytest.mark.parametrize("d1,d2,param", data)
+@pytest.mark.parametrize("d1,d2,param", data2)
 @pytest.mark.parametrize("allfreq", [True, False])
 def test_wxs_dvv(d1: np.ndarray, d2: np.ndarray, param: dict, allfreq: bool):
     if allfreq:
@@ -78,7 +74,7 @@ def test_wxs_dvv(d1: np.ndarray, d2: np.ndarray, param: dict, allfreq: bool):
     assert np.all(dvv < 0)
 
 
-@pytest.mark.parametrize("d1,d2,param", data)
+@pytest.mark.parametrize("d1,d2,param", data2)
 @pytest.mark.parametrize("allfreq", [True, False])
 @pytest.mark.parametrize("normalize", [True, False])
 def test_wts_dvv(d1: np.ndarray, d2: np.ndarray, param: dict, allfreq: bool, normalize: bool):
@@ -92,35 +88,6 @@ def test_wts_dvv(d1: np.ndarray, d2: np.ndarray, param: dict, allfreq: bool, nor
     assert np.all(dvv < 0)
 
 
-@pytest.mark.parametrize("d1,d2,param", data)
-@pytest.mark.parametrize("allfreq", [True, False])
-@pytest.mark.parametrize("normalize", [True, False])
-def test_wts_allfreq(d1: np.ndarray, d2: np.ndarray, param: dict, allfreq: bool, normalize: bool):
-    dv_range = 0.05
-    nbtrial = 50
-    if allfreq:
-        _, dvv, err = wts_allfreq(d1, d2, allfreq, param, dv_range, nbtrial, normalize=normalize)
-    else:
-        dvv, err = wts_allfreq(d1, d2, allfreq, param, dv_range, nbtrial, normalize=normalize)
-    assert np.all(np.isclose(err, 0, atol=1e-1))
-    assert np.all(dvv < 0)
-
-
-@pytest.mark.parametrize("d1,d2,param", data)
-@pytest.mark.parametrize("allfreq", [True, False])
-@pytest.mark.parametrize("normalize", [True, False])
-def test_wtdtw_allfreq(d1: np.ndarray, d2: np.ndarray, param: dict, allfreq: bool, normalize: bool):
-    maxlag = 5
-    b = 10
-    direction = 1
-    if allfreq:
-        _, dvv, err = wtdtw_allfreq(d1, d2, allfreq, param, maxlag, b, direction, normalize=normalize)
-    else:
-        dvv, err = wtdtw_allfreq(d1, d2, allfreq, param, maxlag, b, direction, normalize=normalize)
-    assert np.all(np.isclose(err, 0, atol=2e-1))
-    assert np.all(np.isclose(dvv, 0, atol=2e0))
-
-
 @pytest.mark.parametrize("d1,d2,param", data2)
 @pytest.mark.parametrize("allfreq", [True, False])
 @pytest.mark.parametrize("normalize", [True, False])

tests/test_stretching.py

@@ -4,12 +4,7 @@
 import pytest
 from obspy.signal.invsim import cosine_taper
 
-from noisepy.monitoring.monitoring_methods import (
-    mwcs_dvv,
-    stretching,
-    stretching_vect,
-    wcc_dvv,
-)
+from noisepy.monitoring.monitoring_methods import mwcs_dvv, stretching, wcc_dvv
 
 # This short script is intended as a test for the stretching routine
 # it takes a generic sine curve with known stretching factor and ensures
@@ -26,7 +21,9 @@ def test_stretching():
     t_stretch = np.linspace(0.0, 10.0, 2500)
     stretched_signal = np.interp(t, t_stretch, original_signal)
 
+    # {"twin": [0, 100], "t": np.arange(101), "freq": [0.01, 0.49], "dt": 1},
     para = {}
+    para["t"] = np.arange(2501)
     para["dt"] = 1.0 / 250.0
     para["twin"] = [0.0, 10.0]
     para["freq"] = [9.9, 10.1]
@@ -37,26 +34,6 @@ def test_stretching():
     assert dvv + 0.5 < para["dt"]  # assert result is -0.5%
 
 
-def test_stretching_vect():
-    t = np.linspace(0.0, 9.95, 2500)  # 0.5 % perturbation
-    original_signal = np.sin(t * 10.0) * cosine_taper(2500, p=0.75)
-
-    t_stretch = np.linspace(0.0, 10.0, 2500)
-    stretched_signal = np.interp(t, t_stretch, original_signal)
-
-    para = {}
-    para["dt"] = 1.0 / 250.0
-    para["twin"] = [0.0, 10.0]
-    para["freq"] = [9.9, 10.1]
-
-    dvv, error, cc, cdp = stretching_vect(
-        ref=original_signal, cur=stretched_signal, dv_range=0.05, nbtrial=100, para=para
-    )
-
-    assert pytest.approx(cc) == 1.0
-    assert dvv + 0.5 < para["dt"]  # assert result is -0.5%
-
-
 def test_wcc_dvv():
     t = np.linspace(0.0, 9.95, 2500)  # 0.5 % perturbation
     original_signal = np.sin(t * 10.0) * cosine_taper(2500, p=0.75)
@@ -121,12 +98,6 @@ def test_mwcs_dvv():
         test_stretching()
     print("Done stretching, no errors, %4.2fs." % (time.time() - t))
 
-    print("Running stretching using numpy...")
-    t = time.time()
-    for i in range(100):
-        test_stretching_vect()
-    print("Done stretching, no errors, %4.2fs." % (time.time() - t))
-
     t = time.time()
     for i in range(100):
         test_mwcs_dvv()

tutorials/_config.yml

@@ -9,6 +9,8 @@ logo: ../docs_old/figures/logo.png
 # See https://jupyterbook.org/content/execute.html
 execute:
   execute_notebooks: force
+  exclude_patterns:
+    - noisepy_aws_batch.ipynb
   timeout: 360
 
 only_build_toc_files: true

tutorials/_toc.yml

@@ -2,14 +2,12 @@
 # Learn more at https://jupyterbook.org/customize/toc.html
 
 format: jb-book
-root: noise_configuration.md
+root: noisepy_configuration.md
 chapters:
 - file: get_started.ipynb
 - file: noisepy_datastore.ipynb
 - file: noisepy_scedc_tutorial.ipynb
 - file: noisepy_ncedc_tutorial.ipynb
 - file: noisepy_compositestore_tutorial.ipynb
 - file: CLI.md
-- file: cloud/checklist.md
-- file: cloud/aws-ec2.md
-- file: cloud/aws-batch.md
+- file: cloud/noisepy_aws_batch.ipynb

tutorials/cloud/README.md

@@ -0,0 +1,28 @@
+# Running NoisePy with AWS
+
+## EC2 and Jupyter Lab
+Please refer to [SCOPED HPS Book](https://seisscoped.org/HPS-book/chapters/cloud/AWS_101.html) for full detailed instruction on launching an AWS EC2 instance and/or running the notebooks within a containerized environment.
+
+## Submit Batch Job
+For large job load, please refer to the [notebook tutorial](./noisepy_aws_batch.ipynb) for more instruction.
+
+## Command Line Interface
+You may create or edit the [config.yml](../config.yml) file with appropriate parameters. The cross-correlation function is written to the `ccf_path`.
+
+```bash
+noisepy cross_correlate --format numpy --raw_data_path s3://scedc-pds/continuous_waveforms/ \
+--xml_path s3://scedc-pds/FDSNstationXML/CI/ \
+--ccf_path s3://<S3_BUCKET>/<CC_PATH> \
+--stations=SBC,RIO,DEV \
+--start=2022-02-02 \
+--end=2022-02-03
+```
+
+This toy problem gathers the all the cross-correlations calculated and stack them into the NumPy format on the S3 bucket, specificed by the `stack_path`.
+
+```bash
+noisepy stack \
+--format numpy \
+--ccf_path s3://<S3_BUCKET>/<CC_PATH> \
+--stack_path s3://<S3_BUCKET>/<STACK_PATH> \
+```