def testBatchDevIterator(self):
batch_size = 3
self.assertEqual(len(self.dev_data), 20)
# make data iterator
dev_iter = make_data_iter(self.dev_data, train=False, shuffle=False,
batch_size=batch_size)
self.assertEqual(dev_iter.batch_size, batch_size)
self.assertFalse(dev_iter.shuffle)
self.assertFalse(dev_iter.train)
self.assertEqual(dev_iter.epoch, 0)
self.assertEqual(dev_iter.iterations, 0)
expected_src0 = torch.Tensor(
[[29, 8, 5, 22, 5, 8, 16, 7, 19, 5, 22, 5, 24, 8, 7, 5, 7, 19,
16, 16, 5, 31, 10, 19, 11, 8, 17, 15, 10, 6, 18, 5, 7, 4, 10, 6,
5, 25, 3],
[10, 17, 11, 5, 28, 12, 4, 23, 4, 5, 0, 10, 17, 11, 5, 22, 5, 14,
8, 7, 7, 5, 10, 17, 11, 5, 14, 8, 5, 31, 10, 6, 5, 9, 3, 1,
1, 1, 1],
[29, 8, 5, 22, 5, 18, 23, 13, 4, 6, 5, 13, 8, 18, 5, 9, 3, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1]]).long()
expected_src0_len = torch.Tensor([39, 35, 17]).long()
expected_trg0 = torch.Tensor(
[[13, 11, 12, 4, 22, 4, 12, 5, 4, 22, 4, 25, 7, 6, 8, 4, 14, 12,
4, 24, 14, 5, 7, 6, 26, 17, 14, 10, 20, 4, 23, 3],
[14, 0, 28, 4, 7, 6, 18, 18, 13, 4, 8, 5, 4, 24, 11, 4, 7, 11,
16, 11, 4, 9, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[13, 11, 12, 4, 22, 4, 7, 11, 27, 27, 5, 4, 9, 3, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]]).long()
expected_trg0_len = torch.Tensor([33, 24, 15]).long()
total_samples = 0
for b in iter(dev_iter):
self.assertEqual(type(b), TorchTBatch)
b = Batch(b, pad_index=self.pad_index)
# test the sorting by src length
self.assertEqual(type(b), Batch)
before_sort = b.src_lengths
b.sort_by_src_lengths()
after_sort = b.src_lengths
self.assertTensorEqual(torch.sort(before_sort, descending=True)[0],
after_sort)
self.assertEqual(type(b), Batch)
if total_samples == 0:
self.assertTensorEqual(b.src, expected_src0)
self.assertTensorEqual(b.src_lengths, expected_src0_len)
self.assertTensorEqual(b.trg, expected_trg0)
self.assertTensorEqual(b.trg_lengths, expected_trg0_len)
total_samples += b.nseqs
self.assertLessEqual(b.nseqs, batch_size)
self.assertEqual(total_samples, len(self.dev_data))
def validate_on_data(model: Model, data: Dataset,
logger: Logger,
batch_size: int,
use_cuda: bool, max_output_length: int,
level: str, eval_metric: Optional[str],
loss_function: torch.nn.Module = None,
beam_size: int = 1, beam_alpha: int = -1,
batch_type: str = "sentence",
postprocess: bool = True
) \
-> (float, float, float, List[str], List[List[str]], List[str],
List[str], List[List[str]], List[np.array]):
"""
Generate translations for the given data.
If `loss_function` is not None and references are given,
also compute the loss.
:param model: model module
:param logger: logger
:param data: dataset for validation
:param batch_size: validation batch size
:param use_cuda: if True, use CUDA
:param max_output_length: maximum length for generated hypotheses
:param level: segmentation level, one of "char", "bpe", "word"
:param eval_metric: evaluation metric, e.g. "bleu"
:param loss_function: loss function that computes a scalar loss
for given inputs and targets
:param beam_size: beam size for validation.
If <2 then greedy decoding (default).
:param beam_alpha: beam search alpha for length penalty,
disabled if set to -1 (default).
:param batch_type: validation batch type (sentence or token)
:param postprocess: if True, remove BPE segmentation from translations
:return:
- current_valid_score: current validation score [eval_metric],
- valid_loss: validation loss,
- valid_ppl:, validation perplexity,
- valid_sources: validation sources,
- valid_sources_raw: raw validation sources (before post-processing),
- valid_references: validation references,
- valid_hypotheses: validation_hypotheses,
- decoded_valid: raw validation hypotheses (before post-processing),
- valid_attention_scores: attention scores for validation hypotheses
"""
if batch_size > 1000 and batch_type == "sentence":
logger.warning(
"WARNING: Are you sure you meant to work on huge batches like "
"this? 'batch_size' is > 1000 for sentence-batching. "
"Consider decreasing it or switching to"
" 'eval_batch_type: token'.")
valid_iter = make_data_iter(dataset=data,
batch_size=batch_size,
batch_type=batch_type,
shuffle=False,
train=False)
valid_sources_raw = data.src
pad_index = model.src_vocab.stoi[PAD_TOKEN]
# disable dropout
model.eval()
# don't track gradients during validation
with torch.no_grad():
all_outputs = []
valid_attention_scores = []
total_loss = 0
total_ntokens = 0
total_nseqs = 0
for valid_batch in iter(valid_iter):
# run as during training to get validation loss (e.g. xent)
batch = Batch(valid_batch, pad_index, use_cuda=use_cuda)
# sort batch now by src length and keep track of order
sort_reverse_index = batch.sort_by_src_lengths()
# run as during training with teacher forcing
if loss_function is not None and batch.trg is not None:
batch_loss = model.get_loss_for_batch(
batch, loss_function=loss_function)
total_loss += batch_loss
total_ntokens += batch.ntokens
total_nseqs += batch.nseqs
# run as during inference to produce translations
output, attention_scores = model.run_batch(
batch=batch,
beam_size=beam_size,
beam_alpha=beam_alpha,
max_output_length=max_output_length)
# sort outputs back to original order
all_outputs.extend(output[sort_reverse_index])
valid_attention_scores.extend(
attention_scores[sort_reverse_index]
if attention_scores is not None else [])
assert len(all_outputs) == len(data)
if loss_function is not None and total_ntokens > 0:
# total validation loss
valid_loss = total_loss
# exponent of token-level negative log prob
valid_ppl = torch.exp(total_loss / total_ntokens)
else:
valid_loss = -1
valid_ppl = -1
# decode back to symbols
decoded_valid = model.trg_vocab.arrays_to_sentences(arrays=all_outputs,
cut_at_eos=True)
# evaluate with metric on full dataset
join_char = " " if level in ["word", "bpe"] else ""
valid_sources = [join_char.join(s) for s in data.src]
valid_references = [join_char.join(t) for t in data.trg]
valid_hypotheses = [join_char.join(t) for t in decoded_valid]
# post-process
if level == "bpe" and postprocess:
valid_sources = [bpe_postprocess(s) for s in valid_sources]
valid_references = [bpe_postprocess(v) for v in valid_references]
valid_hypotheses = [bpe_postprocess(v) for v in valid_hypotheses]
# if references are given, evaluate against them
if valid_references:
assert len(valid_hypotheses) == len(valid_references)
current_valid_score = 0
if eval_metric.lower() == 'bleu':
# this version does not use any tokenization
current_valid_score = bleu(valid_hypotheses, valid_references)
elif eval_metric.lower() == 'chrf':
current_valid_score = chrf(valid_hypotheses, valid_references)
elif eval_metric.lower() == 'token_accuracy':
current_valid_score = token_accuracy(valid_hypotheses,
valid_references,
level=level)
elif eval_metric.lower() == 'sequence_accuracy':
current_valid_score = sequence_accuracy(
valid_hypotheses, valid_references)
else:
current_valid_score = -1
return current_valid_score, valid_loss, valid_ppl, valid_sources, \
valid_sources_raw, valid_references, valid_hypotheses, \
decoded_valid, valid_attention_scores
def validate_on_data(model: Model,
data: Dataset,
batch_size: int,
use_cuda: bool,
max_output_length: int,
trg_level: str,
eval_metrics: Optional[Sequence[str]],
loss_function: torch.nn.Module = None,
beam_size: int = 0,
force_prune_size: int = 5,
beam_alpha: int = 0,
batch_type: str = "sentence",
save_attention: bool = False,
validate_by_label: bool = False,
forced_sparsity: bool = False,
method=None,
max_hyps=1,
break_at_p: float = 1.0,
break_at_argmax: bool = False,
short_depth: int = 0):
"""
Generate translations for the given data.
If `loss_function` is not None and references are given,
also compute the loss.
:param model:
:param data: dataset for validation
:param batch_size: validation batch size
:param use_cuda:
:param max_output_length: maximum length for generated hypotheses
:param trg_level: target segmentation level
:param eval_metrics:
:param loss_function: loss function that computes a scalar loss
for given inputs and targets
:param beam_size: beam size for validation (default 0 is greedy)
:param beam_alpha: beam search alpha for length penalty (default 0)
:param batch_type: validation batch type (sentence or token)
:return:
- current_valid_scores: current validation score [eval_metric],
- valid_references: validation references,
- valid_hypotheses: validation_hypotheses,
- decoded_valid: raw validation hypotheses (before post-processing),
- valid_attention_scores: attention scores for validation hypotheses
"""
if beam_size > 0:
force_prune_size = beam_size
if validate_by_label:
assert isinstance(data, TSVDataset) and data.label_columns
valid_scores = defaultdict(float) # container for scores
stats = defaultdict(float)
valid_iter = make_data_iter(dataset=data,
batch_size=batch_size,
batch_type=batch_type,
shuffle=False,
train=False,
use_cuda=use_cuda)
pad_index = model.trg_vocab.stoi[PAD_TOKEN]
model.eval() # disable dropout
force_objectives = loss_function is not None or forced_sparsity
# possible tasks are: force w/ gold, force w/ empty, search
scorer = partial(len_penalty, alpha=beam_alpha) if beam_alpha > 0 else None
confidences = []
corrects = []
with torch.no_grad():
all_outputs = []
valid_attention_scores = defaultdict(list)
for valid_batch in iter(valid_iter):
batch = Batch(valid_batch, pad_index)
rev_index = batch.sort_by_src_lengths()
encoder_output, _ = model.encode(batch)
empty_probs = None
if force_objectives and not isinstance(model, EnsembleModel):
# compute all the logits.
logits = model.force_decode(batch, encoder_output)[0]
bsz, gold_len, vocab_size = logits.size()
gold, gold_lengths, _ = batch["trg"]
prediction_steps = gold_lengths.sum().item() - bsz
assert gold.size(0) == bsz
if loss_function is not None:
gold_pred = gold[:, 1:].contiguous().view(-1)
batch_loss = loss_function(
logits.view(-1, logits.size(-1)), gold_pred)
valid_scores["loss"] += batch_loss
if forced_sparsity:
# compute probabilities
out = logits.view(-1, vocab_size)
if isinstance(model, EnsembleModel):
probs = out
else:
probs = model.decoder.gen_func(out, dim=-1)
# Compute numbers derived from the distributions.
# This includes support size, entropy, and calibration
non_pad = (gold[:, 1:] != pad_index).view(-1)
real_probs = probs[non_pad]
n_supported = real_probs.gt(0).sum().item()
pred_ps, pred_ix = real_probs.max(dim=-1)
real_gold = gold[:, 1:].contiguous().view(-1)[non_pad]
real_correct = pred_ix.eq(real_gold)
corrects.append(real_correct)
confidences.append(pred_ps)
beam_probs, _ = real_probs.topk(force_prune_size, dim=-1)
pruned_mass = 1 - beam_probs.sum(dim=-1)
stats["force_pruned_mass"] += pruned_mass.sum().item()
# compute stuff with the empty sequence
empty_probs = probs.view(bsz, gold_len,
vocab_size)[:, 0, model.eos_index]
assert empty_probs.size() == gold_lengths.size()
empty_possible = empty_probs.gt(0).sum().item()
empty_mass = empty_probs.sum().item()
stats["eos_supported"] += empty_possible
stats["eos_mass"] += empty_mass
stats["n_supp"] += n_supported
stats["n_pred"] += prediction_steps
short_scores = None
if short_depth > 0:
# we call run_batch again with the short depth. We don't
# really care what the hypotheses are, we only want the
# scores
_, _, short_scores = model.run_batch(
batch=batch,
beam_size=beam_size, # can this be removed?
scorer=scorer, # should be none
max_output_length=short_depth,
method="dfs",
max_hyps=max_hyps,
encoder_output=encoder_output,
return_scores=True)
# run as during inference to produce translations
# todo: return_scores for greedy
output, attention_scores, beam_scores = model.run_batch(
batch=batch,
beam_size=beam_size,
scorer=scorer,
max_output_length=max_output_length,
method=method,
max_hyps=max_hyps,
encoder_output=encoder_output,
return_scores=True,
break_at_argmax=break_at_argmax,
break_at_p=break_at_p)
stats["hyp_length"] += output.ne(model.pad_index).sum().item()
if beam_scores is not None and empty_probs is not None:
# I need to expand this to handle stuff up to length m.
# note that although you can compute the probability of the
# empty sequence without any extra computation, you *do* need
# to do extra decoding if you want to get the most likely
# sequence with length <= m.
empty_better = empty_probs.log().gt(beam_scores).sum().item()
stats["empty_better"] += empty_better
if short_scores is not None:
short_better = short_scores.gt(beam_scores).sum().item()
stats["short_better"] += short_better
# sort outputs back to original order
all_outputs.extend(output[rev_index])
if save_attention and attention_scores is not None:
# beam search currently does not support attention logging
for k, v in attention_scores.items():
valid_attention_scores[k].extend(v[rev_index])
assert len(all_outputs) == len(data)
ref_length = sum(len(d.trg) for d in data)
valid_scores["length_ratio"] = stats["hyp_length"] / ref_length
assert len(corrects) == len(confidences)
if corrects:
valid_scores["ece"] = expected_calibration_error(corrects, confidences)
if stats["n_pred"] > 0:
valid_scores["ppl"] = math.exp(valid_scores["loss"] / stats["n_pred"])
if forced_sparsity and stats["n_pred"] > 0:
valid_scores["support"] = stats["n_supp"] / stats["n_pred"]
valid_scores["empty_possible"] = stats["eos_supported"] / len(
all_outputs)
valid_scores["empty_prob"] = stats["eos_mass"] / len(all_outputs)
valid_scores[
"force_pruned_mass"] = stats["force_pruned_mass"] / stats["n_pred"]
if beam_size > 0:
valid_scores["empty_better"] = stats["empty_better"] / len(
all_outputs)
if short_depth > 0:
score_name = "depth_{}_better".format(short_depth)
valid_scores[score_name] = stats["short_better"] / len(
all_outputs)
# postprocess
raw_hyps = model.trg_vocab.arrays_to_sentences(all_outputs)
valid_hyps = postprocess(raw_hyps, trg_level)
valid_refs = postprocess(data.trg, trg_level)
# evaluate
eval_funcs = {
"bleu": bleu,
"chrf": chrf,
"token_accuracy": partial(token_accuracy, level=trg_level),
"sequence_accuracy": sequence_accuracy,
"wer": word_error_rate,
"cer": partial(character_error_rate, level=trg_level),
"levenshtein_distance": partial(levenshtein_distance, level=trg_level)
}
selected_eval_metrics = {name: eval_funcs[name] for name in eval_metrics}
decoding_scores, scores_by_label = evaluate_decoding(
data, valid_refs, valid_hyps, selected_eval_metrics, validate_by_label)
valid_scores.update(decoding_scores)
return valid_scores, valid_refs, valid_hyps, \
raw_hyps, valid_attention_scores, scores_by_label
def get_rl_loss_for_batch(self, batch: Batch, loss_function,
rl_weight: float, beta_entropy: float,
use_cuda: bool, max_output_length: int,
level: str) -> Tensor:
"""
Generate translations for the given data.
If `loss_function` is not None and references are given,
also compute the loss.
:param model: model module
:param logger: logger
:param data: dataset for validation
:param batch_size: validation batch size
:param use_cuda: if True, use CUDA
:param max_output_length: maximum length for generated hypotheses
:param level: segmentation level, one of "char", "bpe", "word"
:param eval_metric: evaluation metric, e.g. "bleu"
:param loss_function: loss function that computes a scalar loss
for given inputs and targets
:param beam_size: beam size for validation.
If <2 then greedy decoding (default).
:param beam_alpha: beam search alpha for length penalty,
disabled if set to -1 (default).
:param batch_type: validation batch type (sentence or token)
:return:
- current_valid_score: current validation score [eval_metric],
- valid_loss: validation loss,
- valid_ppl:, validation perplexity,
- valid_sources: validation sources,
- valid_sources_raw: raw validation sources (before post-processing),
- valid_references: validation references,
- valid_hypotheses: validation_hypotheses,
- decoded_valid: raw validation hypotheses (before post-processing),
- valid_attention_scores: attention scores for validation hypotheses
"""
# sort batch now by src length and keep track of order
sort_reverse_index = batch.sort_by_src_lengths()
if rl_weight != 1:
# note not yet normalized by tokens
#TODO: Check here to make sure that we do not delete mle_loss here
breakpoint()
loss = self.get_loss_for_batch(batch, loss_function)
# for logging
mle_loss = loss.cpu().detach().item()
else:
loss = 0
mle_loss = 0
if rl_weight > 0.0:
# run as during inference to produce translations & RL score
output, transposed_log_probs, entropy = self.run_rl_batch(
batch=batch, max_output_length=max_output_length)
# sort outputs back to original order
output = output[sort_reverse_index]
log_probs = torch.stack(transposed_log_probs).T[
sort_reverse_index] # T x B -> B x T as Tensor
# decode back to symbols
decoded_src = self.src_vocab.arrays_to_sentences(arrays=batch.src,
cut_at_eos=True)
decoded_trg = self.trg_vocab.arrays_to_sentences(arrays=batch.trg,
cut_at_eos=True)
decoded_hyp = self.trg_vocab.arrays_to_sentences(arrays=output,
cut_at_eos=True)
# evaluate with metric on full dataset
join_char = " " if level in ["word", "bpe"] else ""
train_sources = [join_char.join(s) for s in decoded_src]
train_references = [join_char.join(t) for t in decoded_trg]
train_hypotheses = [join_char.join(t) for t in decoded_hyp]
# post-process
if level == "bpe":
train_sources = [bpe_postprocess(s) for s in train_sources]
train_references = [
bpe_postprocess(v) for v in train_references
]
train_hypotheses = [
bpe_postprocess(v) for v in train_hypotheses
]
# if references are given, evaluate against them
assert len(train_hypotheses) == len(train_references)
reinforce_scores = self.bleurt_scorer(references=train_references,
hypotheses=train_hypotheses)
reinforce_scores = torch.tensor(reinforce_scores).unsqueeze(-1)
if use_cuda:
reinforce_scores = reinforce_scores.cuda()
log_probs = log_probs.cuda()
reward_adjusted_log_probs = torch.mul(log_probs, reinforce_scores)
# minimize the log-adjusted cost and maximize entropy (or "multiply entropy by -1 and minimize")
# note this is not normalized by the number of tokens yet
batch_rl_loss = reward_adjusted_log_probs.sum(
) - beta_entropy * entropy
loss = loss * (1 - rl_weight) + rl_weight * batch_rl_loss
batch_rl_loss = batch_rl_loss.cpu().detach().item()
entropy = entropy.cpu().detach().item()
mean_bleurt = torch.mean(reinforce_scores).item()
else:
batch_rl_loss = 0
entropy = 0
mean_bleurt = 0
return loss, batch_rl_loss, mle_loss, entropy, mean_bleurt
def validate_on_data(model,
data,
batch_size,
use_cuda,
max_output_length,
level,
eval_metric,
criterion,
beam_size=0,
beam_alpha=-1):
"""
Generate translations for the given data.
If `criterion` is not None and references are given, also compute the loss.
:param model:
:param data:
:param batch_size:
:param use_cuda:
:param max_output_length:
:param level:
:param eval_metric:
:param criterion:
:param beam_size:
:param beam_alpha:
:return:
"""
valid_iter = make_data_iter(dataset=data,
batch_size=batch_size,
shuffle=False,
train=False)
valid_sources_raw = [s for s in data.src]
pad_index = model.src_vocab.stoi[PAD_TOKEN]
# disable dropout
model.eval()
# don't track gradients during validation
with torch.no_grad():
all_outputs = []
valid_attention_scores = []
total_loss = 0
total_ntokens = 0
for valid_i, valid_batch in enumerate(iter(valid_iter), 1):
# run as during training to get validation loss (e.g. xent)
batch = Batch(valid_batch, pad_index, use_cuda=use_cuda)
# sort batch now by src length and keep track of order
sort_reverse_index = batch.sort_by_src_lengths()
# TODO save computation: forward pass is computed twice
# run as during training with teacher forcing
if criterion is not None and batch.trg is not None:
batch_loss = model.get_loss_for_batch(batch,
criterion=criterion)
total_loss += batch_loss
total_ntokens += batch.ntokens
# run as during inference to produce translations
output, attention_scores = model.run_batch(
batch=batch,
beam_size=beam_size,
beam_alpha=beam_alpha,
max_output_length=max_output_length)
# sort outputs back to original order
all_outputs.extend(output[sort_reverse_index])
valid_attention_scores.extend(
attention_scores[sort_reverse_index]
if attention_scores is not None else [])
assert len(all_outputs) == len(data)
if criterion is not None and total_ntokens > 0:
# total validation loss
valid_loss = total_loss
# exponent of token-level negative log prob
valid_ppl = torch.exp(total_loss / total_ntokens)
else:
valid_loss = -1
valid_ppl = -1
# decode back to symbols
decoded_valid = arrays_to_sentences(arrays=all_outputs,
vocabulary=model.trg_vocab,
cut_at_eos=True)
# evaluate with metric on full dataset
join_char = " " if level in ["word", "bpe"] else ""
valid_sources = [join_char.join(s) for s in data.src]
valid_references = [join_char.join(t) for t in data.trg]
valid_hypotheses = [join_char.join(t) for t in decoded_valid]
# post-process
if level == "bpe":
valid_sources = [bpe_postprocess(s) for s in valid_sources]
valid_references = [bpe_postprocess(v) for v in valid_references]
valid_hypotheses = [bpe_postprocess(v) for v in valid_hypotheses]
# if references are given, evaluate against them
if len(valid_references) > 0:
assert len(valid_hypotheses) == len(valid_references)
current_valid_score = 0
if eval_metric.lower() == 'bleu':
# this version does not use any tokenization
current_valid_score = bleu(valid_hypotheses, valid_references)
elif eval_metric.lower() == 'chrf':
current_valid_score = chrf(valid_hypotheses, valid_references)
elif eval_metric.lower() == 'token_accuracy':
current_valid_score = token_accuracy(valid_hypotheses,
valid_references,
level=level)
elif eval_metric.lower() == 'sequence_accuracy':
current_valid_score = sequence_accuracy(
valid_hypotheses, valid_references)
else:
current_valid_score = -1
return current_valid_score, valid_loss, valid_ppl, valid_sources, \
valid_sources_raw, valid_references, valid_hypotheses, \
decoded_valid, \
valid_attention_scores
def validate_on_data(model: Model, data: Dataset,
batch_size: int,
use_cuda: bool, max_output_length: int,
src_level: str,
trg_level: str,
eval_metrics: Optional[Sequence[str]],
attn_metrics: Optional[Sequence[str]],
loss_function: torch.nn.Module = None,
beam_size: int = 0, beam_alpha: int = 0,
batch_type: str = "sentence",
save_attention: bool = False,
log_sparsity: bool = False,
apply_mask: bool = True # hmm
) \
-> (float, float, float, List[str], List[List[str]], List[str],
List[str], List[List[str]], List[np.array]):
"""
Generate translations for the given data.
If `loss_function` is not None and references are given,
also compute the loss.
:param model: model module
:param data: dataset for validation
:param batch_size: validation batch size
:param use_cuda: if True, use CUDA
:param max_output_length: maximum length for generated hypotheses
:param src_level: source segmentation level, one of "char", "bpe", "word"
:param trg_level: target segmentation level, one of "char", "bpe", "word"
:param eval_metrics: evaluation metric, e.g. "bleu"
:param loss_function: loss function that computes a scalar loss
for given inputs and targets
:param beam_size: beam size for validation.
If 0 then greedy decoding (default).
:param beam_alpha: beam search alpha for length penalty,
disabled if set to 0 (default).
:param batch_type: validation batch type (sentence or token)
:return:
- current_valid_score: current validation score [eval_metric],
- valid_loss: validation loss,
- valid_ppl:, validation perplexity,
- valid_sources: validation sources,
- valid_sources_raw: raw validation sources (before post-processing),
- valid_references: validation references,
- valid_hypotheses: validation_hypotheses,
- decoded_valid: raw validation hypotheses (before post-processing),
- valid_attention_scores: attention scores for validation hypotheses
"""
eval_funcs = {
"bleu": bleu,
"chrf": chrf,
"token_accuracy": partial(token_accuracy, level=trg_level),
"sequence_accuracy": sequence_accuracy,
"wer": wer,
"cer": partial(character_error_rate, level=trg_level)
}
selected_eval_metrics = {name: eval_funcs[name] for name in eval_metrics}
valid_iter = make_data_iter(
dataset=data, batch_size=batch_size, batch_type=batch_type,
shuffle=False, train=False)
valid_sources_raw = [s for s in data.src]
pad_index = model.src_vocab.stoi[PAD_TOKEN]
# disable dropout
model.eval()
# don't track gradients during validation
scorer = partial(len_penalty, alpha=beam_alpha) if beam_alpha > 0 else None
with torch.no_grad():
all_outputs = []
valid_attention_scores = defaultdict(list)
total_loss = 0
total_ntokens = 0
total_nseqs = 0
total_attended = defaultdict(int)
greedy_steps = 0
greedy_supported = 0
for valid_batch in iter(valid_iter):
# run as during training to get validation loss (e.g. xent)
batch = Batch(valid_batch, pad_index, use_cuda=use_cuda)
# sort batch now by src length and keep track of order
sort_reverse_index = batch.sort_by_src_lengths()
# run as during training with teacher forcing
if loss_function is not None and batch.trg is not None:
batch_loss = model.get_loss_for_batch(
batch, loss_function=loss_function)
total_loss += batch_loss
total_ntokens += batch.ntokens
total_nseqs += batch.nseqs
# run as during inference to produce translations
output, attention_scores, probs = model.run_batch(
batch=batch, beam_size=beam_size, scorer=scorer,
max_output_length=max_output_length, log_sparsity=log_sparsity,
apply_mask=apply_mask)
if log_sparsity:
lengths = torch.LongTensor((output == model.trg_vocab.stoi[EOS_TOKEN]).argmax(axis=1)).unsqueeze(1)
batch_greedy_steps = lengths.sum().item()
greedy_steps += lengths.sum().item()
ix = torch.arange(output.shape[1]).unsqueeze(0).expand(output.shape[0], -1)
mask = ix <= lengths
supp = probs.exp().gt(0).sum(dim=-1).cpu() # batch x len
supp = torch.where(mask, supp, torch.tensor(0)).sum()
greedy_supported += supp.float().item()
# sort outputs back to original order
all_outputs.extend(output[sort_reverse_index])
if attention_scores is not None:
# is attention_scores ever None?
if save_attention:
# beam search currently does not support attention logging
for k, v in attention_scores.items():
valid_attention_scores[k].extend(v[sort_reverse_index])
if attn_metrics:
# add to total_attended
for k, v in attention_scores.items():
total_attended[k] += (v > 0).sum()
assert len(all_outputs) == len(data)
if log_sparsity:
print(greedy_supported / greedy_steps)
valid_scores = dict()
if loss_function is not None and total_ntokens > 0:
# total validation loss
valid_loss = total_loss
valid_scores["loss"] = total_loss
valid_scores["ppl"] = torch.exp(total_loss / total_ntokens)
# decode back to symbols
decoded_valid = model.trg_vocab.arrays_to_sentences(arrays=all_outputs,
cut_at_eos=True)
# evaluate with metric on full dataset
src_join_char = " " if src_level in ["word", "bpe"] else ""
trg_join_char = " " if trg_level in ["word", "bpe"] else ""
valid_sources = [src_join_char.join(s) for s in data.src]
valid_references = [trg_join_char.join(t) for t in data.trg]
valid_hypotheses = [trg_join_char.join(t) for t in decoded_valid]
if attn_metrics:
decoded_ntokens = sum(len(t) for t in decoded_valid)
for attn_metric in attn_metrics:
assert attn_metric == "support"
for attn_name, tot_attended in total_attended.items():
score_name = attn_name + "_" + attn_metric
# this is not the right denominator
valid_scores[score_name] = tot_attended / decoded_ntokens
# post-process
if src_level == "bpe":
valid_sources = [bpe_postprocess(s) for s in valid_sources]
if trg_level == "bpe":
valid_references = [bpe_postprocess(v) for v in valid_references]
valid_hypotheses = [bpe_postprocess(v) for v in valid_hypotheses]
languages = [language for language in data.language]
by_language = defaultdict(list)
seqs = zip(valid_references, valid_hypotheses) if valid_references else valid_hypotheses
if languages:
examples = zip(languages, seqs)
for lang, seq in examples:
by_language[lang].append(seq)
else:
by_language[None].extend(seqs)
# if references are given, evaluate against them
# incorrect if-condition?
# scores_by_lang = {name: dict() for name in selected_eval_metrics}
scores_by_lang = dict()
if valid_references and eval_metrics is not None:
assert len(valid_hypotheses) == len(valid_references)
for eval_metric, eval_func in selected_eval_metrics.items():
score_by_lang = dict()
for lang, pairs in by_language.items():
lang_hyps, lang_refs = zip(*pairs)
lang_score = eval_func(lang_hyps, lang_refs)
score_by_lang[lang] = lang_score
score = sum(score_by_lang.values()) / len(score_by_lang)
valid_scores[eval_metric] = score
scores_by_lang[eval_metric] = score_by_lang
if not languages:
scores_by_lang = None
return valid_scores, valid_sources, \
valid_sources_raw, valid_references, valid_hypotheses, \
decoded_valid, valid_attention_scores, scores_by_lang, by_language
def validate_on_data(model: Model,
data: Dataset,
batch_size: int,
use_cuda: bool,
max_output_length: int,
level: str,
eval_metric: Optional[str],
loss_function: torch.nn.Module = None,
beam_size: int = 0,
beam_alpha: int = -1,
batch_type: str = "sentence",
kb_task = None,
valid_kb: Dataset = None,
valid_kb_lkp: list = [],
valid_kb_lens:list=[],
valid_kb_truvals: Dataset = None,
valid_data_canon: Dataset = None,
report_on_canonicals: bool = False,
) \
-> (float, float, float, List[str], List[List[str]], List[str],
List[str], List[List[str]], List[np.array]):
"""
Generate translations for the given data.
If `loss_function` is not None and references are given,
also compute the loss.
:param model: model module
:param data: dataset for validation
:param batch_size: validation batch size
:param use_cuda: if True, use CUDA
:param max_output_length: maximum length for generated hypotheses
:param level: segmentation level, one of "char", "bpe", "word"
:param eval_metric: evaluation metric, e.g. "bleu"
:param loss_function: loss function that computes a scalar loss
for given inputs and targets
:param beam_size: beam size for validation.
If 0 then greedy decoding (default).
:param beam_alpha: beam search alpha for length penalty,
disabled if set to -1 (default).
:param batch_type: validation batch type (sentence or token)
:param kb_task: is not None if kb_task should be executed
:param valid_kb: MonoDataset holding the loaded valid kb data
:param valid_kb_lkp: List with valid example index to corresponding kb indices
:param valid_kb_len: List with amount of triples per kb
:param valid_data_canon: TranslationDataset of valid data but with canonized target data (for loss reporting)
:return:
- current_valid_score: current validation score [eval_metric],
- valid_loss: validation loss,
- valid_ppl:, validation perplexity,
- valid_sources: validation sources,
- valid_sources_raw: raw validation sources (before post-processing),
- valid_references: validation references,
- valid_hypotheses: validation_hypotheses,
- decoded_valid: raw validation hypotheses (before post-processing),
- valid_attention_scores: attention scores for validation hypotheses
- valid_ent_f1: TODO FIXME
"""
print(f"\n{'-'*10} ENTER VALIDATION {'-'*10}\n")
print(f"\n{'-'*10} VALIDATION DEBUG {'-'*10}\n")
print("---data---")
print(dir(data[0]))
print([[
getattr(example, attr) for attr in dir(example)
if hasattr(getattr(example, attr), "__iter__") and "kb" in attr
or "src" in attr or "trg" in attr
] for example in data[:3]])
print(batch_size)
print(use_cuda)
print(max_output_length)
print(level)
print(eval_metric)
print(loss_function)
print(beam_size)
print(beam_alpha)
print(batch_type)
print(kb_task)
print("---valid_kb---")
print(dir(valid_kb[0]))
print([[
getattr(example, attr) for attr in dir(example)
if hasattr(getattr(example, attr), "__iter__") and "kb" in attr
or "src" in attr or "trg" in attr
] for example in valid_kb[:3]])
print(len(valid_kb_lkp), valid_kb_lkp[-5:])
print(len(valid_kb_lens), valid_kb_lens[-5:])
print("---valid_kb_truvals---")
print(len(valid_kb_truvals), valid_kb_lens[-5:])
print([[
getattr(example, attr) for attr in dir(example)
if hasattr(getattr(example, attr), "__iter__") and "kb" in attr
or "src" in attr or "trg" in attr or "trv" in attr
] for example in valid_kb_truvals[:3]])
print("---valid_data_canon---")
print(len(valid_data_canon), valid_data_canon[-5:])
print([[
getattr(example, attr) for attr in dir(example)
if hasattr(getattr(example, attr), "__iter__") and "kb" in attr
or "src" in attr or "trg" in attr or "trv" or "can" in attr
] for example in valid_data_canon[:3]])
print(report_on_canonicals)
print(f"\n{'-'*10} END VALIDATION DEBUG {'-'*10}\n")
if not kb_task:
valid_iter = make_data_iter(dataset=data,
batch_size=batch_size,
batch_type=batch_type,
shuffle=False,
train=False)
else:
# knowledgebase version of make data iter and also provide canonized target data
# data: for bleu/ent f1
# canon_data: for loss
valid_iter = make_data_iter_kb(data,
valid_kb,
valid_kb_lkp,
valid_kb_lens,
valid_kb_truvals,
batch_size=batch_size,
batch_type=batch_type,
shuffle=False,
train=False,
canonize=model.canonize,
canon_data=valid_data_canon)
valid_sources_raw = data.src
pad_index = model.src_vocab.stoi[PAD_TOKEN]
# disable dropout
model.eval()
# don't track gradients during validation
with torch.no_grad():
all_outputs = []
valid_attention_scores = []
valid_kb_att_scores = []
total_loss = 0
total_ntokens = 0
total_nseqs = 0
for valid_batch in iter(valid_iter):
# run as during training to get validation loss (e.g. xent)
batch = Batch(valid_batch, pad_index, use_cuda=use_cuda) \
if not kb_task else \
Batch_with_KB(valid_batch, pad_index, use_cuda=use_cuda)
assert hasattr(batch, "kbsrc") == bool(kb_task)
# sort batch now by src length and keep track of order
if not kb_task:
sort_reverse_index = batch.sort_by_src_lengths()
else:
sort_reverse_index = list(range(batch.src.shape[0]))
# run as during training with teacher forcing
if loss_function is not None and batch.trg is not None:
ntokens = batch.ntokens
if hasattr(batch, "trgcanon") and batch.trgcanon is not None:
ntokens = batch.ntokenscanon # normalize loss with num canonical tokens for perplexity
# do a loss calculation without grad updates just to report valid loss
# we can only do this when batch.trg exists, so not during actual translation/deployment
batch_loss = model.get_loss_for_batch(
batch, loss_function=loss_function)
# keep track of metrics for reporting
total_loss += batch_loss
total_ntokens += ntokens # gold target tokens
total_nseqs += batch.nseqs
# run as during inference to produce translations
output, attention_scores, kb_att_scores = model.run_batch(
batch=batch,
beam_size=beam_size,
beam_alpha=beam_alpha,
max_output_length=max_output_length)
# sort outputs back to original order
all_outputs.extend(output[sort_reverse_index])
valid_attention_scores.extend(
attention_scores[sort_reverse_index]
if attention_scores is not None else [])
valid_kb_att_scores.extend(kb_att_scores[sort_reverse_index]
if kb_att_scores is not None else [])
assert len(all_outputs) == len(data)
if loss_function is not None and total_ntokens > 0:
# total validation loss
valid_loss = total_loss
# exponent of token-level negative log likelihood
# can be seen as 2^(cross_entropy of model on valid set); normalized by num tokens;
# see https://en.wikipedia.org/wiki/Perplexity#Perplexity_per_word
valid_ppl = torch.exp(valid_loss / total_ntokens)
else:
valid_loss = -1
valid_ppl = -1
# decode back to symbols
decoding_vocab = model.trg_vocab if not kb_task else model.trv_vocab
decoded_valid = decoding_vocab.arrays_to_sentences(arrays=all_outputs,
cut_at_eos=True)
print(f"decoding_vocab.itos: {decoding_vocab.itos}")
print(decoded_valid)
# evaluate with metric on full dataset
join_char = " " if level in ["word", "bpe"] else ""
valid_sources = [join_char.join(s) for s in data.src]
# TODO replace valid_references with uncanonicalized dev.car data ... requires writing new Dataset in data.py
valid_references = [join_char.join(t) for t in data.trg]
valid_hypotheses = [join_char.join(t) for t in decoded_valid]
# post-process
if level == "bpe":
valid_sources = [bpe_postprocess(s) for s in valid_sources]
valid_references = [bpe_postprocess(v) for v in valid_references]
valid_hypotheses = [bpe_postprocess(v) for v in valid_hypotheses]
# if references are given, evaluate against them
if valid_references:
assert len(valid_hypotheses) == len(valid_references)
print(list(zip(valid_sources, valid_references, valid_hypotheses)))
current_valid_score = 0
if eval_metric.lower() == 'bleu':
# this version does not use any tokenization
current_valid_score = bleu(valid_hypotheses, valid_references)
elif eval_metric.lower() == 'chrf':
current_valid_score = chrf(valid_hypotheses, valid_references)
elif eval_metric.lower() == 'token_accuracy':
current_valid_score = token_accuracy(valid_hypotheses,
valid_references,
level=level)
elif eval_metric.lower() == 'sequence_accuracy':
current_valid_score = sequence_accuracy(
valid_hypotheses, valid_references)
if kb_task:
valid_ent_f1, valid_ent_mcc = calc_ent_f1_and_ent_mcc(
valid_hypotheses,
valid_references,
vocab=model.trv_vocab,
c_fun=model.canonize,
report_on_canonicals=report_on_canonicals)
else:
valid_ent_f1, valid_ent_mcc = -1, -1
else:
current_valid_score = -1
print(f"\n{'-'*10} EXIT VALIDATION {'-'*10}\n")
return current_valid_score, valid_loss, valid_ppl, valid_sources, \
valid_sources_raw, valid_references, valid_hypotheses, \
decoded_valid, valid_attention_scores, valid_kb_att_scores, \
valid_ent_f1, valid_ent_mcc
