C3W1 Assignment 1: Deep N-grams

Course Q&A Natural Language Processing NLP with Sequence Models
1

When creating a post, please add:

  • Week # must be added in the tags option of the post.
  • Link to the classroom item you are referring to:
  • Description (include relevant info but please do not post solution code or your entire notebook)
    Deep N-grams | Coursera

Getting value error while executing

# UNGRADED CLASS: GenerativeModel
class GenerativeModel(tf.keras.Model):


Suspecting issue with passing null value , could you please help to resolve.

print(gen.generate_n_chars(32, " "), '\n\n' + '_'*80)

# UNIT TEST
# Fix the seed to get replicable results for testing
tf.random.set_seed(272)
gen = GenerativeModel(model, vocab, temperature=0.5)

print(gen.generate_n_chars(32, " "), '\n\n' + '_'*80)
print(gen.generate_n_chars(32, "Dear"), '\n\n' + '_'*80)
print(gen.generate_n_chars(32, "KING"), '\n\n' + '_'*80)


Error:
###################

ValueError                                Traceback (most recent call last)
Cell In[108], line 6
      3 tf.random.set_seed(272)
      4 gen = GenerativeModel(model, vocab, temperature=0.5)
----> 6 print(gen.generate_n_chars(32, " "), '\n\n' + '_'*80)
      7 print(gen.generate_n_chars(32, "Dear"), '\n\n' + '_'*80)
      8 print(gen.generate_n_chars(32, "KING"), '\n\n' + '_'*80)

Cell In[107], line 64, in GenerativeModel.generate_n_chars(self, num_chars, prefix)
     62 result = [next_char]
     63 for n in range(num_chars):
---> 64     next_char, states = self.generate_one_step(next_char, states=states)
     65     result.append(next_char)
     67 return tf.strings.join(result)[0].numpy().decode('utf-8')

File /usr/local/lib/python3.8/dist-packages/tensorflow/python/util/traceback_utils.py:153, in filter_traceback.<locals>.error_handler(*args, **kwargs)
    151 except Exception as e:
    152   filtered_tb = _process_traceback_frames(e.__traceback__)
--> 153   raise e.with_traceback(filtered_tb) from None
    154 finally:
    155   del filtered_tb

File /tmp/__autograph_generated_file87tl4msj.py:11, in outer_factory.<locals>.inner_factory.<locals>.tf__generate_one_step(self, inputs, states)
      9 do_return = False
     10 retval_ = ag__.UndefinedReturnValue()
---> 11 input_ids = ag__.converted_call(ag__.ld(line_to_tensor), (ag__.ld(self).inputs, ag__.ld(self).vocab), None, fscope)
     12 (predicted_logits, states) = ag__.converted_call(ag__.ld(self).model, (ag__.converted_call(ag__.ld(tf).expand_dims, (ag__.ld(input_ids), 0), None, fscope),), dict(states=ag__.ld(states), return_state=True), fscope)
     13 predicted_logits = ag__.ld(predicted_logits)[0, (- 1), :]

File /tmp/__autograph_generated_filen5auk7bg.py:11, in outer_factory.<locals>.inner_factory.<locals>.tf__line_to_tensor(line, vocab)
      9 do_return = False
     10 retval_ = ag__.UndefinedReturnValue()
---> 11 chars = ag__.converted_call(ag__.ld(tf).strings.unicode_split, (ag__.ld(line),), dict(input_encoding='UTF-8'), fscope)
     12 ids = ag__.converted_call(ag__.converted_call(ag__.ld(tf).keras.layers.StringLookup, (), dict(vocabulary=ag__.converted_call(ag__.ld(list), (ag__.ld(vocab),), None, fscope), mask_token=None), fscope), (ag__.ld(chars),), None, fscope)
     13 try:

ValueError: in user code:

    File "/tmp/ipykernel_14/4281713250.py", line 34, in generate_one_step  *
        input_ids = line_to_tensor(self.inputs, self.vocab)
    File "/tmp/ipykernel_14/2955566712.py", line 16, in line_to_tensor  *
        chars = tf.strings.unicode_split(line, input_encoding='UTF-8')

    ValueError: None values not supported.
1 Like
2

Hi!
It seems like you are using the object’s inputs (self) (which do not exist in the constructor function) but you have to use the one passed in the function generate_one_step.

1 Like