File "main.py", line 66, in _eval_epoch 'inferred_ids': inferred_ids, NameError: name 'inferred_ids' is not defined about abstractive-summarization-with-transfer-learning

[Vibha111094]

i think you need to add this part:
start_tokens = tf.fill([tx.utils.get_batch_size(src_input_ids)],
bos_token_id)
predictions = decoder(
memory=encoder_output,
memory_sequence_length=src_input_length,
decoding_strategy='infer_greedy',
beam_width=beam_width,
alpha=alpha,
start_tokens=start_tokens,
end_token=eos_token_id,
max_decoding_length=400,
mode=tf.estimator.ModeKeys.PREDICT
)
if beam_width <= 1:
inferred_ids = predictions[0].sample_id
else:
# Uses the best sample by beam search
inferred_ids = predictions['sample_id'][:, :, 0]

I would suggest you tryout the colab version to first ,it would help you understand the flow.

[santhoshkolloju]

I have updated it plz check