3778 / icd-prediction-mimic Goto Github PK

self.embedding_matrix = np.concatenate((self.embedding_matrix, np.zeros((2,W2V_SIZE))), axis=0)
File "<array_function internals>", line 6, in concatenate
ValueError: all the input array dimensions for the concatenation axis must match exactly, but along dimension 1, the array at index 0 has size 100 and the array at index 1 has size 300

@dsevero @arthurreys
I get TypeError: 'Series' objects are mutable, thus they cannot be hashed when running MIMIC_train_w2v.py

Attaching the traceback

Traceback (most recent call last):
  File "MIMIC_train_w2v.py", line 56, in <module>
    main(args)
  File "MIMIC_train_w2v.py", line 26, in main
    mimic.split()
  File "C:\Users\kiran\Music\icd-prediction-mimic-master\datasets.py", line 83, in split
    self.all_icds_train = utils.make_icds_histogram(self.df.query("HADM_ID.isin(@hadm_ids[0])")).index.tolist()
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\frame.py", line 3184, in query
    res = self.eval(expr, **kwargs)
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\frame.py", line 3300, in eval
    return _eval(expr, inplace=inplace, **kwargs)
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\computation\eval.py", line 327, in eval
    ret = eng_inst.evaluate()
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\computation\engines.py", line 70, in evaluate
    res = self._evaluate()
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\computation\engines.py", line 118, in _evaluate
    _check_ne_builtin_clash(self.expr)
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\computation\engines.py", line 27, in _check_ne_builtin_clash
    names = expr.names
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\computation\expr.py", line 850, in names
    return frozenset([self.terms.name])
  File "C:\Users\kiran\Anaconda3\lib\site-packages\pandas\core\generic.py", line 1886, in __hash__
    " hashed".format(self.__class__.__name__)
TypeError: 'Series' objects are mutable, thus they cannot be hashed

When trying to run the MIMIC_train_w2v.py script, I am receiving the RuntimeError listed at the top. One change I made into the code was changing "size" to "vector_size" on line 75 in the feature_extraction.py do to an error of it not recognizing the "size" argument. Unsure if those are related. Full traceback of the error is below:

Edit: Looks like size has been updated to vector_size. So, that is just a simple fix. However, I have not found a solution for the vocab not being built.

Traceback (most recent call last): File "C:\Users\paige\Desktop\690 Project\icd-prediction-mimic-master\icd-prediction-mimic-master\MIMIC_train_w2v.py", line 56, in <module> main(args) File "C:\Users\paige\Desktop\690 Project\icd-prediction-mimic-master\icd-prediction-mimic-master\MIMIC_train_w2v.py", line 32, in main w2v.fit(mimic) File "C:\Users\paige\Desktop\690 Project\icd-prediction-mimic-master\icd-prediction-mimic-master\feature_extraction.py", line 111, in fit epochs=self.model_w2v.epochs) File "C:\Python37\lib\site-packages\gensim\models\word2vec.py", line 1034, in train self._check_training_sanity(epochs=epochs, total_examples=total_examples, total_words=total_words) File "C:\Python37\lib\site-packages\gensim\models\word2vec.py", line 1524, in _check_training_sanity raise RuntimeError("you must first build vocabulary before training the model") RuntimeError: you must first build vocabulary before training the model