ValueError: Number of processes must be at least 1

[WATCH0620]

When I attempted to use RoseTTAFold, I encountered the following error at the end.

ValueError                                Traceback (most recent call last)
[<ipython-input-6-db7245cb6df4>](https://localhost:8080/#) in <cell line: 1>()
----> 1 mn.build_model(user_json='/content/drive/MyDrive/synthoseis-master/config/example.json', run_id=1,test_mode=50)

5 frames
[/content/drive/MyDrive/synthoseis-master/main.py](https://localhost:8080/#) in build_model(user_json, run_id, test_mode, rpm_factors)
     53     seismic = SeismicVolume(p, f, closures)
     54     seismic.build_elastic_properties("inv_vel")
---> 55     seismic.build_seismic_volumes()
     56 
     57     closures.write_closure_info_to_log(seismic.rfc_raw[1:4, ...])

[/content/drive/MyDrive/synthoseis-master/datagenerator/Seismic.py](https://localhost:8080/#) in build_seismic_volumes(self)
    249         if hasattr(self.cfg, "wavelets"):
    250             self.bandlimit_volumes_wavelets(n_wavelets=1)
--> 251         normalised_cumsum = self.postprocess_rfc_cubes(
    252             self.rfc_noise_added[:], "", stack=True
    253         )

[/content/drive/MyDrive/synthoseis-master/datagenerator/Seismic.py](https://localhost:8080/#) in postprocess_rfc_cubes(self, rfc_data, filename_suffix, bb, stack)
    164             bandlimited = self.apply_bandlimits(rfc_data, (4.0, 90.0))
    165         else:
--> 166             bandlimited = self.apply_bandlimits(rfc_data)
    167         if self.cfg.lateral_filter_size > 1:
    168             bandlimited = self.apply_lateral_filter(bandlimited)

[/content/drive/MyDrive/synthoseis-master/datagenerator/Seismic.py](https://localhost:8080/#) in apply_bandlimits(self, data, frequencies)
    509         if self.cfg.multiprocess_bp:
    510             # Much faster to use multiprocessing and apply band-limitation to entire cube at once
--> 511             with Pool(processes=min(num, os.cpu_count() - 2)) as p:
    512                 iterator = zip(
    513                     [data[x, ...] for x in range(num)],

[/usr/lib/python3.10/multiprocessing/context.py](https://localhost:8080/#) in Pool(self, processes, initializer, initargs, maxtasksperchild)
    117         '''Returns a process pool object'''
    118         from .pool import Pool
--> 119         return Pool(processes, initializer, initargs, maxtasksperchild,
    120                     context=self.get_context())
    121 

[/usr/lib/python3.10/multiprocessing/pool.py](https://localhost:8080/#) in __init__(self, processes, initializer, initargs, maxtasksperchild, context)
    203             processes = os.cpu_count() or 1
    204         if processes < 1:
--> 205             raise ValueError("Number of processes must be at least 1")
    206         if maxtasksperchild is not None:
    207             if not isinstance(maxtasksperchild, int) or maxtasksperchild <= 0:

ValueError: Number of processes must be at least 1

[tpmerrifield]

Hi, it looks like os.cpu_count() is returning a processes value that is < 1, which is a bit strange. What machine are you using? You should be able to turn off multiprocessing by setting the "multiprocess_bp" to false in the config file.

[WATCH0620]

Hi, it looks like os.cpu_count() is returning a processes value that is < 1, which is a bit strange. What machine are you using? You should be able to turn off multiprocessing by setting the "multiprocess_bp" to false in the config file.

This problem has been solved, thank you. The machine OS i use is Windows7. I have other problem, how can i synthesize a data without noise added?