Takes results dictionary and saves the f1 vs epoch plot in stats_graph_folder. from_json indicates if the results dictionary was loaded from

Fit the model.

Ensure data types are correct in the parameter dictionary. Args: param (dict): dictionary of parameter settings.

Initialize stats_graph_folder. Args: parameters (type): description.

Load data or models from the package folder. Args: name (str): name of the resource content_type (str): either "data" or "tr

Get the parameters from the config file.

Get paths for the datasets. Args: parameters (type): description. dataset_types (type): description.

Helper function for bio_to_bioes

Helper function for convert_conll_from_bio_to_bioes

y_pred: list of predicted labels y_true: list of gold labels evaluation_mode: 'bio', 'token', or 'binary' Both y_pred and y_true mus

Check parameters are compatible. Args: parameters (type): description. dataset_filepaths (type): description

Get the default parameters.

Check parameters are compatible.

Specify figure size in centimeter in matplotlib Source: http://stackoverflow.com/a/22787457/395857 By gns-ank

convert conll file in conll-filepath to brat annotations and output to brat_output_folder, with reference to the existing text files in brat

Define training procedure

generates reference text files and adds the corresponding filename and token offsets to conll file. conll_input_filepath: path to a conl

http://stackoverflow.com/questions/37517587/how-can-i-change-the-intensity-of-a-colormap-in-matplotlib

http://stackoverflow.com/questions/415511/how-to-get-current-time-in-python

Inspired by: - http://stackoverflow.com/a/16124677/395857 - http://stackoverflow.com/a/25074150/395857

dataset_filepaths : dictionary with keys 'train', 'valid', 'test', 'deploy'

Load parameters from the ini file if specified, take into account any command line argument, and ensure that each parameter is cast to the

NeuroNER main method Args: parameters_filepath the path to the parameters file output_folder the path to the output folder

Parse the NeuroNER arguments arguments: arguments the arguments, optionally given as argument

Predict.

Copy the dataset.pickle, parameters.ini, and model checkpoint files after removing the data used for training. The dataset and labe

Save results

Heatmap with text in each cell with matplotlib's pyplot Source: http://stackoverflow.com/a/25074150/395857 By HYRY

Remove the dataset and labels from dataset.pickle. If delete_token_mappings = True, then also remove token_to_index and index_to_token excep

Remove all token embeddings except UNK.

dataset_filepaths : dictionary with keys 'train', 'valid', 'test', 'deploy' Overwrites the data of type specified in dataset_types us

Assumes '.txt' and '.ann' files are in the input_folder. Checks for the compatibility between .txt and .ann at the same time.

Check if brat annotation and text files are compatible.

check if token offsets match between conll and brat .txt files. conll_filepath: path to conll file brat_folder: folder that contains th

Check the contents of dataset.pickle and model_xxx.ckpt. model_folder: folder containing dataset.pickle and model_xxx.ckpt to be checked.

http://stackoverflow.com/questions/1773793/convert-configparser-items-to-dictionary

http://stackoverflow.com/questions/1868714/how-do-i-copy-an-entire-directory-of-files-into-an-existing-directory-using-pyth

Create the folder if it doesn't exist already.

Fetch a dataset. If name is provided, fetch from the package folder. If url is provided, fetch from a remote location. Args: nam

Fetch a pre-trained model and copy to a local "trained_models" folder If name is provided, fetch from the package folder. Args:

Getting the basename of the filepath without the extension E.g. 'data/formatted/movie_reviews.pickle' -> 'movie_reviews'

http://stackoverflow.com/questions/5998245/get-current-time-in-milliseconds-in-python

http://stackoverflow.com/questions/5998245/get-current-time-in-milliseconds-in-python

Get valid filepaths for the datasets.

http://stackoverflow.com/questions/38987/how-can-i-merge-two-python-dictionaries-in-a-single-expression Given any number of dicts, shallow co

Order a dictionary by 'key' or 'value'. mode should be either 'key' or 'value' http://stackoverflow.com/questions/613183/sort-a-python-di

http://stackoverflow.com/questions/3438756/some-built-in-to-pad-a-list-in-python Example: pad_list([6,2,3], 5, 0) returns [6,2,3,0,0]

Plot scikit-learn classification report. Extension based on http://stackoverflow.com/a/31689645/395857

Predict Args: text (str): Description.

Predict labels using trained model

Loads results.json file in the ../stats_graphs/folder_name, and plot f1 vs epoch. Use for debugging purposes, or in case the program stopped

http://stackoverflow.com/questions/483666/python-reverse-inverse-a-mapping http://stackoverflow.com/questions/25480089/right-way-to-initializ

Sanity test to check if all proper model output files are created in the output folder

Train.

Attach a lot of summaries to a Tensor (for TensorBoard visualization). From https://www.tensorflow.org/get_started/summaries_and_tensorboard