def load_annotation_db(self, path):
# Expect two paths, one to questions and one to annotations
assert (
len(path) == 2
), "VQACPv2 requires 2 paths; one to questions and one to annotations"
with PathManager.open(path[0]) as f:
path_0 = json.load(f)
with PathManager.open(path[1]) as f:
path_1 = json.load(f)
if "annotations" in path[0]:
annotations = path_0
questions = path_1
else:
annotations = path_1
questions = path_0
# Convert to linear format
data = []
question_dict = {}
for question in questions:
question_dict[question["question_id"]] = question["question"]
for annotation in annotations:
annotation["question"] = question_dict[annotation["question_id"]]
answers = []
for answer in annotation["answers"]:
answers.append(answer["answer"])
annotation["answers"] = answers
data.append(copy.deepcopy(annotation))
self.data = data
def test_on_update_end(self):
self.cb.on_train_start()
self.cb.on_update_end(meter=self.trainer.meter, should_log=False)
f = PathManager.open(os.path.join(self.tmpdir, "train.log"))
self.assertFalse(
any("time_since_start" in line for line in f.readlines()))
self.cb.on_update_end(meter=self.trainer.meter, should_log=True)
f = PathManager.open(os.path.join(self.tmpdir, "train.log"))
self.assertTrue(
any("time_since_start" in line for line in f.readlines()))
def test_log_writer(self) -> None:
self.writer.info(self._tmpfile_write_contents)
f = PathManager.open(
glob.glob(os.path.join(self._tmpdir, "logs", "train*"))[0])
self.assertTrue(
any(self._tmpfile_write_contents in line
for line in f.readlines()))
f = PathManager.open(os.path.join(self._tmpdir, "train.log"))
self.assertTrue(
any(self._tmpfile_write_contents in line
for line in f.readlines()))
def __init__(self, in_dim, weights_file, bias_file):
super().__init__()
with PathManager.open(weights_file, "rb") as w:
weights = pickle.load(w)
with PathManager.open(bias_file, "rb") as b:
bias = pickle.load(b)
out_dim = bias.shape[0]
self.lc = nn.Linear(in_dim, out_dim)
self.lc.weight.data.copy_(torch.from_numpy(weights))
self.lc.bias.data.copy_(torch.from_numpy(bias))
self.out_dim = out_dim
def checksum(self, download_path):
"""
Checksum on a given file.
Args:
download_path (string): path to the downloaded file.
"""
if self._hashcode is None:
print(f"[ Checksum not provided, skipping for {self._file_name}]")
return
sha256_hash = hashlib.sha256()
destination = os.path.join(download_path, self._file_name)
if not PathManager.isfile(destination):
# File is not present, nothing to checksum
return
with PathManager.open(destination, "rb") as f:
print(f"[ Starting checksum for {self._file_name}]")
for byte_block in iter(lambda: f.read(65536), b""):
sha256_hash.update(byte_block)
if sha256_hash.hexdigest() != self._hashcode:
# remove_dir(download_path)
raise AssertionError(
f"[ Checksum for {self._file_name} from \n{self._url}\n"
"does not match the expected checksum. Please try again. ]"
)
else:
print(f"[ Checksum successful for {self._file_name}]")
def _load_jsonl(self, path):
with PathManager.open(path, "r") as f:
db = f.readlines()
for idx, line in enumerate(db):
db[idx] = json.loads(line.strip("\n"))
self.data = db
self.start_idx = 0
def download_from_google_drive(gd_id, destination, redownload=True):
"""
Use the requests package to download a file from Google Drive.
"""
download = not PathManager.isfile(destination) or redownload
URL = "https://docs.google.com/uc?export=download"
if not download:
return download
else:
# Check first if link is live
check_header(gd_id, from_google=True)
with requests.Session() as session:
response = session.get(URL, params={"id": gd_id}, stream=True)
token = _get_confirm_token(response)
if token:
response.close()
params = {"id": gd_id, "confirm": token}
response = session.get(URL, params=params, stream=True)
CHUNK_SIZE = 32768
with PathManager.open(destination, "wb") as f:
for chunk in response.iter_content(CHUNK_SIZE):
if chunk: # filter out keep-alive new chunks
f.write(chunk)
response.close()
return download
def _load_json(self, path):
with PathManager.open(path, "r") as f:
data = json.load(f)
self.metadata = data.get("metadata", {})
self.data = data.get("data", [])
if len(self.data) == 0:
raise RuntimeError("Dataset is empty")
def _torch_load(self, file):
# Backwards compatibility to Pythia
_hack_imports()
with PathManager.open(file, "rb") as f:
if "cuda" in str(self.device):
return torch.load(f, map_location=self.device)
else:
return torch.load(f, map_location=lambda storage, loc: storage)
def csv_dump(self, filepath):
with PathManager.open(filepath, "w") as f:
title = self.report[0].keys()
cw = csv.DictWriter(f,
title,
delimiter=",",
quoting=csv.QUOTE_MINIMAL)
cw.writeheader()
cw.writerows(self.report)
def load_feat(feat_path: str, convert_to_tensor: bool = False) -> Any:
with PathManager.open(feat_path, "rb") as f:
if feat_path.endswith("npy"):
feat = np.load(f, allow_pickle=True)
if convert_to_tensor:
feat = torch.from_numpy(feat)
elif feat_path.endswith("pth"):
feat = torch.load(f, map_location=torch.device("cpu"))
else:
raise AssertionError("Unknown feature type")
return feat
def __init__(self, config: Config, *args, **kwargs):
super().__init__()
model_data_dir = get_absolute_path(config.model_data_dir)
if not os.path.isabs(config.weights_file):
weights_file = os.path.join(model_data_dir, config.weights_file)
if not os.path.isabs(config.bias_file):
bias_file = os.path.join(model_data_dir, config.bias_file)
if not PathManager.exists(bias_file) or not PathManager.exists(weights_file):
download_path = download_pretrained_model("detectron.vmb_weights")
weights_file = get_absolute_path(os.path.join(download_path, "fc7_w.pkl"))
bias_file = get_absolute_path(os.path.join(download_path, "fc7_b.pkl"))
with PathManager.open(weights_file, "rb") as w:
weights = pickle.load(w)
with PathManager.open(bias_file, "rb") as b:
bias = pickle.load(b)
out_dim = bias.shape[0]
self.lc = nn.Linear(config.in_dim, out_dim)
self.lc.weight.data.copy_(torch.from_numpy(weights))
self.lc.bias.data.copy_(torch.from_numpy(bias))
self.out_dim = out_dim
def mark_done(path, version_string=None):
"""
Mark this path as prebuilt.
Marks the path as done by adding a '.built' file with the current timestamp
plus a version description string if specified.
Args:
path (str): The file path to mark as built
version_string (str): The version of this dataset
"""
data = {}
data["created_at"] = str(datetime.datetime.today())
data["version"] = version_string
with PathManager.open(os.path.join(path, ".built.json"), "w") as f:
json.dump(data, f)
def checksum(self, file, hashes):
sha256_hash = hashlib.sha256()
destination = file
with PathManager.open(destination, "rb") as f:
print("Starting checksum for {}".format(os.path.basename(file)))
for byte_block in iter(lambda: f.read(65536), b""):
sha256_hash.update(byte_block)
if sha256_hash.hexdigest() not in hashes:
# remove_dir(download_path)
raise AssertionError(
f"Checksum of downloaded file does not match the expected "
+ "checksum. Please try again."
)
else:
print("Checksum successful")
def load_pretrained_model(model_name_or_path, *args, **kwargs):
# If this is a file, then load this directly else download and load
if PathManager.exists(model_name_or_path):
download_path = model_name_or_path
model_name = model_name_or_path
else:
download_path = download_pretrained_model(model_name_or_path, *args,
**kwargs)
model_name = model_name_or_path
configs = glob.glob(os.path.join(download_path, "*.yaml"))
assert len(configs) <= 1, (
"Multiple yaml files with the pretrained model. " +
"MMF doesn't know what to do.")
ckpts = []
allowed_ckpt_types = ("*.ckpt", "*.pth", "*.pt")
for ckpt_type in allowed_ckpt_types:
ckpts.extend(glob.glob(os.path.join(download_path, ckpt_type)))
assert (
len(ckpts) == 1
), "None or multiple checkpoints files. MMF doesn't know what to do."
_hack_imports()
with PathManager.open(ckpts[0], "rb") as f:
ckpt = torch.load(f, map_location=lambda storage, loc: storage)
# If configs are not present, will ckpt provide the config?
if len(configs) == 0:
assert "config" in ckpt, (
"No configs provided with pretrained model "
" while checkpoint also doesn't have configuration.")
config = ckpt["config"]
else:
config = load_yaml(configs[0])
model_config = config.get("model_config", config)
ckpt = ckpt.get("model", ckpt)
# Also handle the case of model_name is path
model_config = model_config.get(
model_name.split(os.path.sep)[-1].split(".")[0])
return {"config": model_config, "checkpoint": ckpt, "full_config": config}
def _load_npy(self, path):
with PathManager.open(path, "rb") as f:
self.db = np.load(f, allow_pickle=True)
self.start_idx = 0
if type(self.db) == dict:
self.metadata = self.db.get("metadata", {})
self.data = self.db.get("data", [])
else:
# TODO: Deprecate support for this
self.metadata = {"version": 1}
self.data = self.db
# Handle old imdb support
if "image_id" not in self.data[0]:
self.start_idx = 1
if len(self.data) == 0:
self.data = self.db
def _download_model(self):
_is_master = is_master()
model_file_path = os.path.join(get_mmf_cache_dir(), "wiki.en.bin")
if not _is_master:
return model_file_path
if PathManager.exists(model_file_path):
logger.info(f"Vectors already present at {model_file_path}.")
return model_file_path
import requests
from tqdm import tqdm
from VisualBERT.mmf.common.constants import FASTTEXT_WIKI_URL
PathManager.mkdirs(os.path.dirname(model_file_path))
response = requests.get(FASTTEXT_WIKI_URL, stream=True)
with PathManager.open(model_file_path, "wb") as f:
pbar = tqdm(
total=int(response.headers["Content-Length"]) / 4096,
miniters=50,
disable=not _is_master,
)
idx = 0
for data in response.iter_content(chunk_size=4096):
if data:
if idx % 50 == 0:
pbar.update(len(data))
f.write(data)
idx += 1
pbar.close()
logger.info(f"fastText bin downloaded at {model_file_path}.")
return model_file_path
def built(path, version_string=None):
"""
Check if '.built' flag has been set for that task.
If a version_string is provided, this has to match, or the version
is regarded as not built.
Version_string are generally the dataset version + the date the file was
last updated. If this doesn't match, dataset will be mark not built. This makes
sure that if we update our features or anything else features are updated
for the end user.
"""
if version_string:
fname = os.path.join(path, ".built.json")
if not PathManager.isfile(fname):
return False
else:
with PathManager.open(fname, "r") as read:
text = json.load(read)
return text.get("version", None) == version_string
else:
return PathManager.isfile(os.path.join(path, ".built.json"))
def load_str_list(fname):
with PathManager.open(fname) as f:
lines = f.readlines()
lines = [line.strip() for line in lines]
return lines
def json_dump(self, filepath):
with PathManager.open(filepath, "w") as f:
json.dump(self.report, f)
def download(url, path, fname, redownload=True, disable_tqdm=False):
"""
Download file using `requests`.
If ``redownload`` is set to false, then will not download tar file again if it is
present (default ``True``).
Returns whether download actually happened or not
"""
outfile = os.path.join(path, fname)
download = not PathManager.isfile(outfile) or redownload
retry = 5
exp_backoff = [2**r for r in reversed(range(retry))]
pbar = None
if download:
# First test if the link is actually downloadable
check_header(url)
if not disable_tqdm:
print("[ Downloading: " + url + " to " + outfile + " ]")
pbar = tqdm.tqdm(unit="B",
unit_scale=True,
desc=f"Downloading {fname}",
disable=disable_tqdm)
while download and retry >= 0:
resume_file = outfile + ".part"
resume = PathManager.isfile(resume_file)
if resume:
resume_pos = os.path.getsize(resume_file)
mode = "ab"
else:
resume_pos = 0
mode = "wb"
response = None
with requests.Session() as session:
try:
header = ({
"Range": "bytes=%d-" % resume_pos,
"Accept-Encoding": "identity"
} if resume else {})
response = session.get(url,
stream=True,
timeout=5,
headers=header)
# negative reply could be 'none' or just missing
if resume and response.headers.get("Accept-Ranges",
"none") == "none":
resume_pos = 0
mode = "wb"
CHUNK_SIZE = 32768
total_size = int(response.headers.get("Content-Length", -1))
# server returns remaining size if resuming, so adjust total
total_size += resume_pos
pbar.total = total_size
done = resume_pos
with PathManager.open(resume_file, mode) as f:
for chunk in response.iter_content(CHUNK_SIZE):
if chunk: # filter out keep-alive new chunks
f.write(chunk)
if total_size > 0:
done += len(chunk)
if total_size < done:
# don't freak out if content-length was too small
total_size = done
pbar.total = total_size
pbar.update(len(chunk))
break
except (
requests.exceptions.ConnectionError,
requests.exceptions.ReadTimeout,
):
retry -= 1
pbar.clear()
if retry >= 0:
print("Connection error, retrying. (%d retries left)" %
retry)
time.sleep(exp_backoff[retry])
else:
print("Retried too many times, stopped retrying.")
finally:
if response:
response.close()
if retry < 0:
raise RuntimeWarning(
"Connection broken too many times. Stopped retrying.")
if download and retry > 0:
pbar.update(done - pbar.n)
if done < total_size:
raise RuntimeWarning("Received less data than specified in " +
"Content-Length header for " + url +
". There may be a download problem.")
move(resume_file, outfile)
if pbar:
pbar.close()
return download
def _cached_log_stream(filename):
return PathManager.open(filename, "a")
def __init__(self,
vocab_file=None,
embedding_dim=300,
data_dir=None,
*args,
**kwargs):
"""Vocab class to be used when you want to train word embeddings from
scratch based on a custom vocab. This will initialize the random
vectors for the vocabulary you pass. Get the vectors using
`get_vectors` function. This will also create random embeddings for
some predefined words like PAD - <pad>, SOS - <s>, EOS - </s>,
UNK - <unk>.
Parameters
----------
vocab_file : str
Path of the vocabulary file containing one word per line
embedding_dim : int
Size of the embedding
"""
self.type = "base"
self.word_dict = {}
self.itos = {}
self.itos[self.PAD_INDEX] = self.PAD_TOKEN
self.itos[self.SOS_INDEX] = self.SOS_TOKEN
self.itos[self.EOS_INDEX] = self.EOS_TOKEN
self.itos[self.UNK_INDEX] = self.UNK_TOKEN
self.word_dict[self.SOS_TOKEN] = self.SOS_INDEX
self.word_dict[self.EOS_TOKEN] = self.EOS_INDEX
self.word_dict[self.PAD_TOKEN] = self.PAD_INDEX
self.word_dict[self.UNK_TOKEN] = self.UNK_INDEX
index = len(self.itos.keys())
self.total_predefined = len(self.itos.keys())
if vocab_file is not None:
if not os.path.isabs(vocab_file) and data_dir is not None:
vocab_file = os.path.join(data_dir, vocab_file)
vocab_file = get_absolute_path(vocab_file)
if not PathManager.exists(vocab_file):
raise RuntimeError("Vocab not found at " + vocab_file)
with PathManager.open(vocab_file, "r") as f:
for line in f:
self.itos[index] = line.strip()
self.word_dict[line.strip()] = index
index += 1
self.word_dict[self.SOS_TOKEN] = self.SOS_INDEX
self.word_dict[self.EOS_TOKEN] = self.EOS_INDEX
self.word_dict[self.PAD_TOKEN] = self.PAD_INDEX
self.word_dict[self.UNK_TOKEN] = self.UNK_INDEX
# Return unk index by default
self.stoi = defaultdict(self.get_unk_index)
self.stoi.update(self.word_dict)
self.vectors = torch.FloatTensor(self.get_size(), embedding_dim)
def test_on_test_end(self):
self.cb.on_test_end(report=self.report, meter=self.trainer.meter)
f = PathManager.open(os.path.join(self.tmpdir, "train.log"))
self.assertTrue(
any("Finished run in" in line for line in f.readlines()))
def save_config(self):
cfg_file = os.path.join(self.ckpt_foldername, "config.yaml")
with PathManager.open(cfg_file, "w") as f:
f.write(self.config.pretty(resolve=True))
def default_loader(path):
with PathManager.open(path, "rb") as f:
img = Image.open(f)
return img.convert("RGB")
def save(self, update, iteration=None, update_best=False):
# Only save in main process
if not is_master():
return
if not iteration:
iteration = update
ckpt_filepath = os.path.join(self.models_foldername,
"model_%d.ckpt" % update)
best_ckpt_filepath = os.path.join(self.ckpt_foldername,
self.ckpt_prefix + "best.ckpt")
current_ckpt_filepath = os.path.join(self.ckpt_foldername,
self.ckpt_prefix + "current.ckpt")
best_iteration = (self.trainer.early_stop_callback.early_stopping.
best_monitored_iteration)
best_update = (self.trainer.early_stop_callback.early_stopping.
best_monitored_update)
best_metric = (self.trainer.early_stop_callback.early_stopping.
best_monitored_value)
model = self.trainer.model
data_parallel = registry.get("data_parallel") or registry.get(
"distributed")
fp16_scaler = getattr(self.trainer, "scaler", None)
fp16_scaler_dict = None
if fp16_scaler is not None:
fp16_scaler_dict = fp16_scaler.state_dict()
if data_parallel is True:
model = model.module
ckpt = {
"model": model.state_dict(),
"optimizer": self.trainer.optimizer.state_dict(),
"best_iteration": best_iteration,
"current_iteration": iteration,
"current_epoch": self.trainer.current_epoch,
"num_updates": update,
"best_update": best_update,
"best_metric_value": best_metric,
"fp16_scaler": fp16_scaler_dict,
# Convert to container to avoid any dependencies
"config": OmegaConf.to_container(self.config, resolve=True),
}
lr_scheduler = self.trainer.lr_scheduler_callback._scheduler
if lr_scheduler is not None:
ckpt["lr_scheduler"] = lr_scheduler.state_dict()
if self.git_repo:
git_metadata_dict = self._get_vcs_fields()
ckpt.update(git_metadata_dict)
with PathManager.open(ckpt_filepath, "wb") as f:
torch.save(ckpt, f)
if update_best:
with PathManager.open(best_ckpt_filepath, "wb") as f:
torch.save(ckpt, f)
# Save current always
with PathManager.open(current_ckpt_filepath, "wb") as f:
torch.save(ckpt, f)
# Remove old checkpoints if max_to_keep is set
if self.max_to_keep > 0:
if len(self.saved_iterations) == self.max_to_keep:
self.remove(self.saved_iterations.pop(0))
self.saved_iterations.append(update)
def test_file_io_open(self):
with PathManager.open(self._tmpfile, mode="r") as f:
s = f.read()
self.assertEqual(s, self._tmpfile_contents)
def finalize(self):
if is_master():
with PathManager.open(self.pth_filepath, "wb") as f:
torch.save(self.trainer.model.state_dict(), f)
