n3fit.io package

Submodules

n3fit.io.writer module

Module containing functions dedicated to the write down of the output of n3fit

The goal is to generate the same folder/file structure as the old nnfit code so previously active scripts can still work.

class n3fit.io.writer.SuperEncoder(*, skipkeys=False, ensure_ascii=True, check_circular=True, allow_nan=True, sort_keys=False, indent=None, separators=None, default=None)

Bases: JSONEncoder

Custom json encoder to get around the fact that np.float32 =/= float

default(o)

Implement this method in a subclass such that it returns a serializable object for o, or calls the base implementation (to raise a TypeError).

For example, to support arbitrary iterators, you could implement default like this:

def default(self, o):
    try:
        iterable = iter(o)
    except TypeError:
        pass
    else:
        return list(iterable)
    # Let the base class default method raise the TypeError
    return super().default(o)

class n3fit.io.writer.WriterWrapper(replica_numbers, pdf_objects, stopping_object, all_chi2s, theory, timings)

Bases: object

write_data(save_path, fitname, weights_name)

Save all the data of a fit, for all replicas.

Parameters:

• save_path – path for the fit results, ex: ${PWD}/runcard_name/nnfit

• fitname – name of the fit, ex: Basic_runcard

• weights_name – name of the file to save weights to, if not empty

n3fit.io.writer.evln2lha(evln, nf=6)

n3fit.io.writer.jsonfit(best_epoch, positivity_status, preprocessing, arc_lengths, integrability_numbers, tr_chi2, vl_chi2, true_chi2, stop_epoch, timing)

Generates a dictionary containing all relevant metadata for the fit

Parameters:

• best_epoch (int) – epoch at which the best fit was found

• positivity_status (str) – string describing the positivity status of the fit

• preprocessing (dict) – dictionary of the preprocessing factors

• arc_lengths (list) – list of the arc lengths of the different PDFs

• integrability_numbers (list) – list of the integrability numbers of the different PDFs

• tr_chi2 (float) – chi2 for the training

• vl_chi2 (float) – chi2 for the validation

• true_chi2 (float) – chi2 for the exp (unreplica’d data)

• epoch_stop (int) – epoch at which the stopping stopped (not the one for the best fit!)

• timing (dict) – dictionary of the timing of the different events that happened

n3fit.io.writer.storefit(pdf_object, replica, out_path, theory)

One-trick function which generates all output in the NNPDF format so that all other scripts can still be used.

Parameters:

• pdf_object – N3PDF object constructed from the pdf_model that receives as input a point in x and returns an array of 14 flavours

• replica – the replica index

• out_path – the path where to store the output

• theory – theory information of the fit

n3fit.io.writer.version()

Generates a dictionary with misc version info for this run

Module contents