Sourcery refactored main branch #1
Conversation
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
| version += "+" + sha[:7] | ||
| version += f"+{sha[:7]}" |
There was a problem hiding this comment.
Function _get_version refactored with the following changes:
- Use f-string instead of string concatenation (
use-fstring-for-concatenation)
| match = re.match(r"^\s*URL\s+(https:\/\/.+)$", line) | ||
| if match: | ||
| url = match.group(1) | ||
| yield url | ||
| if match := re.match(r"^\s*URL\s+(https:\/\/.+)$", line): | ||
| yield match[1] |
There was a problem hiding this comment.
Function _parse_url refactored with the following changes:
- Use named expression to simplify assignment and conditional (
use-named-expression) - Inline variable that is immediately yielded (
inline-immediately-yielded-variable) - Replace m.group(x) with m[x] for re.Match objects (
use-getitem-for-re-match-groups)
| with open("README.md") as f: | ||
| long_description = f.read() | ||
|
|
||
| long_description = Path("README.md").read_text() |
There was a problem hiding this comment.
Function _main refactored with the following changes:
- Simplify basic file reads with
pathlib(path-read)
| w = [] | ||
| base_workflow_name = f"{prefix}binary_{os_type}_{btype}_py{python_version}_{cu_version}" | ||
| w.append(generate_base_workflow(base_workflow_name, python_version, cu_version, filter_branch, os_type, btype)) | ||
|
|
||
| w = [ | ||
| generate_base_workflow( | ||
| base_workflow_name, | ||
| python_version, | ||
| cu_version, | ||
| filter_branch, | ||
| os_type, | ||
| btype, | ||
| ) | ||
| ] |
There was a problem hiding this comment.
Function build_workflow_pair refactored with the following changes:
- Move assignment closer to its usage within a block (
move-assign-in-block) - Merge append into list declaration (
merge-list-append)
| d["subfolder"] = "" if os_type == "macos" else cu_version + "/" | ||
| d["subfolder"] = "" if os_type == "macos" else f"{cu_version}/" |
There was a problem hiding this comment.
Function generate_upload_workflow refactored with the following changes:
- Use f-string instead of string concatenation (
use-fstring-for-concatenation)
| file_text = speaker_id + "-" + chapter_id + self.base_dataset._ext_txt | ||
| file_text = f"{speaker_id}-{chapter_id}{self.base_dataset._ext_txt}" |
There was a problem hiding this comment.
Function CustomDataset._target_length refactored with the following changes:
- Use f-string instead of string concatenation [×3] (
use-fstring-for-concatenation)
| dataloader = torch.utils.data.DataLoader( | ||
| return torch.utils.data.DataLoader( | ||
| dataset, | ||
| batch_size=None, | ||
| collate_fn=self._train_collate_fn, | ||
| num_workers=10, | ||
| shuffle=True, | ||
| ) | ||
| return dataloader |
There was a problem hiding this comment.
Function LibriSpeechRNNTModule.train_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader( | ||
| return torch.utils.data.DataLoader( | ||
| dataset, | ||
| batch_size=None, | ||
| collate_fn=self._valid_collate_fn, | ||
| num_workers=10, | ||
| ) | ||
| return dataloader |
There was a problem hiding this comment.
Function LibriSpeechRNNTModule.val_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader(dataset, batch_size=1, collate_fn=self._test_collate_fn) | ||
| return dataloader | ||
| return torch.utils.data.DataLoader( | ||
| dataset, batch_size=1, collate_fn=self._test_collate_fn | ||
| ) |
There was a problem hiding this comment.
Function LibriSpeechRNNTModule.test_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader( | ||
| return torch.utils.data.DataLoader( | ||
| dataset, | ||
| batch_size=None, | ||
| collate_fn=self._train_collate_fn, | ||
| num_workers=10, | ||
| shuffle=True, | ||
| ) | ||
| return dataloader |
There was a problem hiding this comment.
Function MuSTCRNNTModule.train_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader( | ||
| return torch.utils.data.DataLoader( | ||
| dataset, | ||
| batch_size=None, | ||
| collate_fn=self._valid_collate_fn, | ||
| num_workers=10, | ||
| ) | ||
| return dataloader |
There was a problem hiding this comment.
Function MuSTCRNNTModule.val_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader(dataset, batch_size=1, collate_fn=self._test_collate_fn) | ||
| return dataloader | ||
| return torch.utils.data.DataLoader( | ||
| dataset, batch_size=1, collate_fn=self._test_collate_fn | ||
| ) |
There was a problem hiding this comment.
Function MuSTCRNNTModule.test_common_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader(dataset, batch_size=1, collate_fn=self._test_collate_fn) | ||
| return dataloader | ||
| return torch.utils.data.DataLoader( | ||
| dataset, batch_size=1, collate_fn=self._test_collate_fn | ||
| ) |
There was a problem hiding this comment.
Function MuSTCRNNTModule.test_he_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader(dataset, batch_size=1, collate_fn=self._test_collate_fn) | ||
| return dataloader | ||
| return torch.utils.data.DataLoader( | ||
| dataset, batch_size=1, collate_fn=self._test_collate_fn | ||
| ) |
There was a problem hiding this comment.
Function MuSTCRNNTModule.dev_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| assert len(idx_target_lengths) > 0 | ||
| assert idx_target_lengths |
There was a problem hiding this comment.
Function CustomDataset.__init__ refactored with the following changes:
- Simplify sequence length comparison (
simplify-len-comparison)
| else: | ||
| scaling_factor = self.anneal_factor ** (self._step_count - self.force_anneal_step) | ||
| return [scaling_factor * base_lr for base_lr in self.base_lrs] | ||
| scaling_factor = self.anneal_factor ** (self._step_count - self.force_anneal_step) | ||
| return [scaling_factor * base_lr for base_lr in self.base_lrs] |
There was a problem hiding this comment.
Function WarmupLR.get_lr refactored with the following changes:
- Remove unnecessary else after guard condition (
remove-unnecessary-else)
| nbest_batch = list(zip(hypos_str, hypos_score, hypos_ids)) | ||
|
|
||
| return nbest_batch | ||
| return list(zip(hypos_str, hypos_score, hypos_ids)) |
There was a problem hiding this comment.
Function post_process_hypos refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| file_text = speaker_id + "-" + chapter_id + librispeech_dataset._ext_txt | ||
| file_text = f"{speaker_id}-{chapter_id}{librispeech_dataset._ext_txt}" |
There was a problem hiding this comment.
Function get_sample_lengths refactored with the following changes:
- Use f-string instead of string concatenation [×3] (
use-fstring-for-concatenation)
| dataloader = torch.utils.data.DataLoader( | ||
| return torch.utils.data.DataLoader( | ||
| dataset, | ||
| num_workers=self.num_workers, | ||
| batch_size=None, | ||
| shuffle=self.train_shuffle, | ||
| ) | ||
| return dataloader |
There was a problem hiding this comment.
Function LibriSpeechDataModule.train_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader(dataset, batch_size=None, num_workers=self.num_workers) | ||
| return dataloader | ||
| return torch.utils.data.DataLoader( | ||
| dataset, batch_size=None, num_workers=self.num_workers | ||
| ) |
There was a problem hiding this comment.
Function LibriSpeechDataModule.val_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| dataloader = torch.utils.data.DataLoader(dataset, batch_size=None) | ||
| return dataloader | ||
| return torch.utils.data.DataLoader(dataset, batch_size=None) |
There was a problem hiding this comment.
Function LibriSpeechDataModule.test_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| filename = "librispeech_clean_100_{}".format(idx) | ||
| filename = f"librispeech_clean_100_{idx}" | ||
| actual = sample[0][2] | ||
| predicted = model(batch) | ||
| hypout.append("{} ({})\n".format(predicted.upper().strip(), filename)) | ||
| refout.append("{} ({})\n".format(actual.upper().strip(), filename)) | ||
| hypout.append(f"{predicted.upper().strip()} ({filename})\n") | ||
| refout.append(f"{actual.upper().strip()} ({filename})\n") |
There was a problem hiding this comment.
Function run_eval refactored with the following changes:
- Replace call to format with f-string [×3] (
use-fstring-for-formatting)
| else: | ||
| scaling_factor = self.anneal_factor ** (self._step_count - self.force_anneal_step) | ||
| return [scaling_factor * base_lr for base_lr in self.base_lrs] | ||
| scaling_factor = self.anneal_factor ** (self._step_count - self.force_anneal_step) | ||
| return [scaling_factor * base_lr for base_lr in self.base_lrs] |
There was a problem hiding this comment.
Function WarmupLR.get_lr refactored with the following changes:
- Remove unnecessary else after guard condition (
remove-unnecessary-else)
| nbest_batch = list(zip(hypos_str, hypos_score, hypos_ids)) | ||
|
|
||
| return nbest_batch | ||
| return list(zip(hypos_str, hypos_score, hypos_ids)) |
There was a problem hiding this comment.
Function post_process_hypos refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| rareset = set() | ||
| for line in fin: | ||
| rareset.add(line.strip().upper()) | ||
|
|
||
| rareset = {line.strip().upper() for line in fin} |
There was a problem hiding this comment.
Lines 18-111 refactored with the following changes:
- Convert for loop into set comprehension (
set-comprehension) - Replace call to format with f-string [×8] (
use-fstring-for-formatting) - Simplify dictionary access using default get (
default-get)
| dataloader = DataLoader( | ||
| return DataLoader( | ||
| dataset, | ||
| batch_sampler=sampler, | ||
| collate_fn=CollateFnLibriLightLimited(), | ||
| num_workers=10, | ||
| ) | ||
| return dataloader |
There was a problem hiding this comment.
Function HuBERTFineTuneModule.val_dataloader refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| args = parser.parse_args() | ||
| return args | ||
| return parser.parse_args() |
There was a problem hiding this comment.
Function _parse_args refactored with the following changes:
- Inline variable that is immediately returned (
inline-immediately-returned-variable)
| if args.use_gpu: | ||
| device = torch.device("cuda") | ||
| else: | ||
| device = torch.device("cpu") | ||
|
|
||
| device = torch.device("cuda") if args.use_gpu else torch.device("cpu") |
There was a problem hiding this comment.
Function main refactored with the following changes:
- Replace if statement with if expression (
assign-if-exp)
| if len(filtered_length_idx) == 0: | ||
| if not filtered_length_idx: |
There was a problem hiding this comment.
Function BucketizeBatchSampler.__init__ refactored with the following changes:
- Simplify sequence length comparison (
simplify-len-comparison)
| buckets = {k: v for k, v in sorted(buckets.items())} | ||
| buckets = dict(sorted(buckets.items())) |
There was a problem hiding this comment.
Function BucketizeBatchSampler._get_buckets refactored with the following changes:
- Replace identity comprehension with call to collection constructor (
identity-comprehension)
Branch
mainrefactored by Sourcery.If you're happy with these changes, merge this Pull Request using the Squash and merge strategy.
See our documentation here.
Run Sourcery locally
Reduce the feedback loop during development by using the Sourcery editor plugin:
Review changes via command line
To manually merge these changes, make sure you're on the
mainbranch, then run:Help us improve this pull request!