Converting .csv file to SEG2 format .dat files)

[jaisinghverma]

Greetings of the day!
How can I convert my .csv file to SEG2 format .dat files) before using the file for processing in swprocess?

Thank you.

[jpvantassel]

Hi @jaisinghverma,
I do not know of an open-source SEG2 writer, but swprocess can read a variety of formats (SEG2, SEGY, SU). What I would recommend is to use obspy to write your data to SU format. Some information on that is provided here.

[jaisinghverma]

Respected Sir, I am receiving the following error after executing the code which says, " OverflowError: cannot convert float infinity to integer". I tried different options but received the same error again and again. Here is the complete error message: C:\Users\Dell\miniconda3\lib\site-packages\swprocess\sensor1c.py:182: RuntimeWarning: divide by zero encountered in log10 x = round(x, -np.sign(scaleco) * int(np.log10(abs(scaleco))))

…

---------------------------------------------------------------------------OverflowError Traceback (most recent call last) Input In [65], in <cell line: 17>() 16 wavefieldtransforms = [] 17 for fnames in fnames_set: ---> 18 wavefieldtransforms.append(swprocess.Masw.run(fnames=fnames, settings=settings)) 19 end = time.perf_counter() 20 print(f"Elapsed Time (s): {round(end-start,2)}") File ~\miniconda3\lib\site-packages\swprocess\masw.py:87, in Masw.run(fnames, settings, map_x, map_y) 83 workflow = Workflow(fnames=fnames, settings=settings, 84 map_x=map_x, map_y=map_y) 86 # Run and return.---> 87 return workflow.run() File ~\miniconda3\lib\site-packages\swprocess\maswworkflows.py:136, in TimeDomainWorkflow.run(self) 135 def run(self):--> 136 self.array = Array1D.from_files(self.fnames, map_x=self.map_x, 137 map_y=self.map_y) 138 self.check() 139 self.trim_offsets() File ~\miniconda3\lib\site-packages\swprocess\array1d.py:657, in Array1D.from_files(cls, fnames, map_x, map_y) 655 sensors = [] 656 for trace in stream.traces:--> 657 sensor = Sensor1C.from_trace(trace, map_x=map_x, map_y=map_y) 658 sensors.append(sensor) 660 # Define source File ~\miniconda3\lib\site-packages\swprocess\sensor1c.py:120, in Sensor1C.from_trace(cls, trace, read_header, map_x, map_y, nstacks, delay, x, y, z) 118 return cls._from_trace_seg2(trace, map_x=map_x, map_y=map_y) 119 elif _format == "SU":--> 120 return cls._from_trace_su(trace, map_x=map_x, map_y=map_y) 121 elif _format == "SEGY": 122 return cls._from_trace_segy(trace, map_x=map_x, map_y=map_y) File ~\miniconda3\lib\site-packages\swprocess\sensor1c.py:182, in Sensor1C._from_trace_su(cls, trace, map_x, map_y) 180 int_x = int(header["group_coordinate_x"]) 181 x = int_x / abs(scaleco) if scaleco < 0 else int_x * scaleco--> 182 x = round(x, -np.sign(scaleco) * int(np.log10(abs(scaleco)))) 184 int_y = int(header["group_coordinate_y"]) 185 y = int_y / abs(scaleco) if scaleco < 0 else int_x * scaleco OverflowError: cannot convert float infinity to integer

On Thu, Aug 15, 2024 at 7:50 PM Joseph Vantassel ***@***.***> wrote: Hi @jaisinghverma <https://github.com/jaisinghverma>, I do not know of an open-source SEG2 writer, but *swprocess* can read a variety of formats (SEG2, SEGY, SU). What I would recommend is to use *obspy* to write your data to SU format. Some information on that is provided here <https://docs.obspy.org/packages/autogen/obspy.core.stream.Stream.write.html> . — Reply to this email directly, view it on GitHub <#26 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZY6RAMUR5KMP4Y4EC45CX3ZRS2JBAVCNFSM6AAAAABMMFOEFGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEOJRGM2TMNZSGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Thanks & Regards,

[jpvantassel]

Hi @jaisinghverma,

This issue is caused by a bad value in your file's header. What version of swprocess are you running? I believe I implemented a work around to this problem in v0.3.0, following issue #16.

[jaisinghverma]

Respected Sir, I was using the older version earlier. Now, I am using the latest version. The earlier issue regarding value error has been resolved. However, I am getting another error which reads "*ValueError: Sensors must have unique x positions*." The complete error is as follows: C:\Users\Dell\miniconda3\lib\site-packages\swprocess\sensor1c.py:199: UserWarning: Resetting nstacks from zero to one. warnings.warn(msg) C:\Users\Dell\miniconda3\lib\site-packages\swprocess\sensor1c.py:205: UserWarning: Resetting scale to be applied to all coordinates from zero to one. warnings.warn(msg) C:\Users\Dell\miniconda3\lib\site-packages\swprocess\array1d.py:676: UserWarning: Coordinate units is unset, assuming length in meters and not degrees minutes seconds. warnings.warn(msg) C:\Users\Dell\miniconda3\lib\site-packages\swprocess\array1d.py:684: UserWarning: Resetting scale to be applied to all coordinates from zero to one. warnings.warn(msg) --------------------------------------------------------------------------- ValueError Traceback (most recent call last) Input In [8], in <cell line: 17>() 16 wavefieldtransforms = [] 17 for fnames in fnames_set: ---> 18 wavefieldtransforms.append(swprocess.Masw.run(fnames=fnames, settings=settings)) 19 end = time.perf_counter() 20 print(f"Elapsed Time (s): {round(end-start,2)}") File ~\miniconda3\lib\site-packages\swprocess\masw.py:87, in Masw.run(fnames, settings, map_x, map_y) 83 workflow = Workflow(fnames=fnames, settings=settings, 84 map_x=map_x, map_y=map_y) 86 # Run and return. ---> 87 return workflow.run() File ~\miniconda3\lib\site-packages\swprocess\maswworkflows.py:136, in TimeDomainWorkflow.run(self) 135 def run(self): --> 136 self.array = Array1D.from_files(self.fnames, map_x=self.map_x, 137 map_y=self.map_y) 138 self.check() 139 self.trim_offsets() File ~\miniconda3\lib\site-packages\swprocess\array1d.py:720, in Array1D.from_files(cls, fnames, map_x, map_y, obspy_read_kwargs) 717 return Source(x=map_x(x), y=map_y(y), z=0) 719 source = parse_source(trace.stats) --> 720 obj = cls(sensors, source) 722 # Stack additional traces, if necessary 723 if len(fnames) > 0: File ~\miniconda3\lib\site-packages\swprocess\array1d.py:84, in Array1D.__init__(self, sensors, source) 68 """Initialize from an iterable of `Sensor1C`s and a `Source`. 69 70 Parameters (...) 81 82 """ 83 logger.info("Initializing Array1D") ---> 84 self.sensors, self.source = self._check_array(sensors, source) File ~\miniconda3\lib\site-packages\swprocess\array1d.py:58, in Array1D._check_array(sensors, source) 56 raise ValueError("All sensors must be similar.") 57 if sensor.x in positions: ---> 58 raise ValueError("Sensors must have unique x positions.") 59 positions.append(sensor.x) 61 # Sort sensors in terms of x from small to large ValueError: Sensors must have unique x positions.

…

On Thu, Sep 19, 2024 at 3:17 AM Joseph Vantassel ***@***.***> wrote: Hi @jaisinghverma <https://github.com/jaisinghverma>, This issue is caused by a bad value in your file's header. What version of swprocess are you running? I believe I implemented a work around to this problem in v0.3.0, following issue #16 <#16>. — Reply to this email directly, view it on GitHub <#26 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZY6RAIQ5D3W7YKNBY5PCHDZXHYHJAVCNFSM6AAAAABMMFOEFGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNJZGQ3DEMJWGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Thanks & Regards,

[jpvantassel]

Hi @jaisinghverma,

The error is because your sensors (e.g., geophones) "must have unique x positions". It is most-likely that the header of your file is not setup correctly so the code cannot understand which sensors go at which position. You may be able to fix this by editing the text-based header and/or properly configuring your data acquisition system.

[alangi]

I think that another option would be to run the files through Geopsy and assign coordinates there. Although I have only done this for passive processing, not MASW, so I'm not sure how to get the files out again.

[alangi]

Maybe this could be an addition. A utility to import a settings file containing coordinates.

[jaisinghverma]

Hi @jaisinghverma,

The error is because your sensors (e.g., geophones) "must have unique x positions". It is most-likely that the header of your file is not setup correctly so the code cannot understand which sensors go at which position. You may be able to fix this by editing the text-based header and/or properly configuring your data acquisition system.

[jaisinghverma]

Respected Sir, Could you please elaborate? Currently I am following the given steps to get the data in KGS (.dat file) format: 1. Collecting the vel-time history (.txt format without any headers) from MASW simulations in ABAQUS. 2. Importing the .txt data file in SurfSeis software and converting it into KGS (.dat) format 3. Using the KGS (.dat) file obtained above in the swprocess package.

…

On Wed, Sep 25, 2024 at 1:29 AM Joseph Vantassel ***@***.***> wrote: Hi @jaisinghverma <https://github.com/jaisinghverma>, The error is because your sensors (e.g., geophones) "must have unique x positions". It is most-likely that the header of your file is not setup correctly so the code cannot understand which sensors go at which position. You may be able to fix this by editing the text-based header and/or properly configuring your data acquisition system. — Reply to this email directly, view it on GitHub <#26 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZY6RANOSECF36FA5OBSQJ3ZYHABDAVCNFSM6AAAAABMMFOEFGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNZSGI2TIOJZGM> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Thanks & Regards,

[jpvantassel]

Hi @jaisinghverma,

In SurfSeis there should be an option to assign coordinates (or positions) to your sensors. It would appear from the error above that all of your sensors are shown as being at the same location in the current .dat file. You will need to correct this.

An alternative to your current approach that may be easier for you is to use the components swprocess provides to create an seismic unix (SU) based data file. You do this by creating Sensor1C and Source objects to then create an Array1D object. The Array1D object has a method to write the data to_file. This can write the file to SU and then you can process that SU formatted data using the masw.ipynb.

Joe