metadata
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:412178
- loss:MultipleNegativesRankingLoss
base_model: answerdotai/ModernBERT-base
widget:
- source_sentence: |-
Clip off all parts from all bounding boxes that are outside of the image.
Returns
-------
imgaug.BoundingBoxesOnImage
Bounding boxes, clipped to fall within the image dimensions.
sentences:
- |-
def model_best(y1, y2, samples=1000, progressbar=True):
"""
Bayesian Estimation Supersedes the T-Test
This model runs a Bayesian hypothesis comparing if y1 and y2 come
from the same distribution. Returns are assumed to be T-distributed.
In addition, computes annual volatility and Sharpe of in and
out-of-sample periods.
This model replicates the example used in:
Kruschke, John. (2012) Bayesian estimation supersedes the t
test. Journal of Experimental Psychology: General.
Parameters
----------
y1 : array-like
Array of returns (e.g. in-sample)
y2 : array-like
Array of returns (e.g. out-of-sample)
samples : int, optional
Number of posterior samples to draw.
Returns
-------
model : pymc.Model object
PyMC3 model containing all random variables.
trace : pymc3.sampling.BaseTrace object
A PyMC3 trace object that contains samples for each parameter
of the posterior.
See Also
--------
plot_stoch_vol : plotting of tochastic volatility model
"""
y = np.concatenate((y1, y2))
mu_m = np.mean(y)
mu_p = 0.000001 * 1 / np.std(y)**2
sigma_low = np.std(y) / 1000
sigma_high = np.std(y) * 1000
with pm.Model() as model:
group1_mean = pm.Normal('group1_mean', mu=mu_m, tau=mu_p,
testval=y1.mean())
group2_mean = pm.Normal('group2_mean', mu=mu_m, tau=mu_p,
testval=y2.mean())
group1_std = pm.Uniform('group1_std', lower=sigma_low,
upper=sigma_high, testval=y1.std())
group2_std = pm.Uniform('group2_std', lower=sigma_low,
upper=sigma_high, testval=y2.std())
nu = pm.Exponential('nu_minus_two', 1 / 29., testval=4.) + 2.
returns_group1 = pm.StudentT('group1', nu=nu, mu=group1_mean,
lam=group1_std**-2, observed=y1)
returns_group2 = pm.StudentT('group2', nu=nu, mu=group2_mean,
lam=group2_std**-2, observed=y2)
diff_of_means = pm.Deterministic('difference of means',
group2_mean - group1_mean)
pm.Deterministic('difference of stds',
group2_std - group1_std)
pm.Deterministic('effect size', diff_of_means /
pm.math.sqrt((group1_std**2 +
group2_std**2) / 2))
pm.Deterministic('group1_annual_volatility',
returns_group1.distribution.variance**.5 *
np.sqrt(252))
pm.Deterministic('group2_annual_volatility',
returns_group2.distribution.variance**.5 *
np.sqrt(252))
pm.Deterministic('group1_sharpe', returns_group1.distribution.mean /
returns_group1.distribution.variance**.5 *
np.sqrt(252))
pm.Deterministic('group2_sharpe', returns_group2.distribution.mean /
returns_group2.distribution.variance**.5 *
np.sqrt(252))
trace = pm.sample(samples, progressbar=progressbar)
return model, trace
- |-
def clip_out_of_image(self):
"""
Clip off all parts from all bounding boxes that are outside of the image.
Returns
-------
imgaug.BoundingBoxesOnImage
Bounding boxes, clipped to fall within the image dimensions.
"""
bbs_cut = [bb.clip_out_of_image(self.shape)
for bb in self.bounding_boxes if bb.is_partly_within_image(self.shape)]
return BoundingBoxesOnImage(bbs_cut, shape=self.shape)
- |-
def _initPermanence(self, potential, connectedPct):
"""
Initializes the permanences of a column. The method
returns a 1-D array the size of the input, where each entry in the
array represents the initial permanence value between the input bit
at the particular index in the array, and the column represented by
the 'index' parameter.
Parameters:
----------------------------
:param potential: A numpy array specifying the potential pool of the column.
Permanence values will only be generated for input bits
corresponding to indices for which the mask value is 1.
:param connectedPct: A value between 0 or 1 governing the chance, for each
permanence, that the initial permanence value will
be a value that is considered connected.
"""
# Determine which inputs bits will start out as connected
# to the inputs. Initially a subset of the input bits in a
# column's potential pool will be connected. This number is
# given by the parameter "connectedPct"
perm = numpy.zeros(self._numInputs, dtype=realDType)
for i in xrange(self._numInputs):
if (potential[i] < 1):
continue
if (self._random.getReal64() <= connectedPct):
perm[i] = self._initPermConnected()
else:
perm[i] = self._initPermNonConnected()
# Clip off low values. Since we use a sparse representation
# to store the permanence values this helps reduce memory
# requirements.
perm[perm < self._synPermTrimThreshold] = 0
return perm
- source_sentence: |-
Perform a weighted average over dicts that are each on a different node
Input: local_name2valcount: dict mapping key -> (value, count)
Returns: key -> mean
sentences:
- >-
def MotionBlur(k=5, angle=(0, 360), direction=(-1.0, 1.0), order=1,
name=None, deterministic=False, random_state=None):
"""
Augmenter that sharpens images and overlays the result with the original image.
dtype support::
See ``imgaug.augmenters.convolutional.Convolve``.
Parameters
----------
k : int or tuple of int or list of int or imgaug.parameters.StochasticParameter, optional
Kernel size to use.
* If a single int, then that value will be used for the height
and width of the kernel.
* If a tuple of two ints ``(a, b)``, then the kernel size will be
sampled from the interval ``[a..b]``.
* If a list, then a random value will be sampled from that list per image.
* If a StochasticParameter, then ``N`` samples will be drawn from
that parameter per ``N`` input images, each representing the kernel
size for the nth image.
angle : number or tuple of number or list of number or imgaug.parameters.StochasticParameter, optional
Angle of the motion blur in degrees (clockwise, relative to top center direction).
* If a number, exactly that value will be used.
* If a tuple ``(a, b)``, a random value from the range ``a <= x <= b`` will
be sampled per image.
* If a list, then a random value will be sampled from that list per image.
* If a StochasticParameter, a value will be sampled from the
parameter per image.
direction : number or tuple of number or list of number or imgaug.parameters.StochasticParameter, optional
Forward/backward direction of the motion blur. Lower values towards -1.0 will point the motion blur towards
the back (with angle provided via `angle`). Higher values towards 1.0 will point the motion blur forward.
A value of 0.0 leads to a uniformly (but still angled) motion blur.
* If a number, exactly that value will be used.
* If a tuple ``(a, b)``, a random value from the range ``a <= x <= b`` will
be sampled per image.
* If a list, then a random value will be sampled from that list per image.
* If a StochasticParameter, a value will be sampled from the
parameter per image.
order : int or iterable of int or imgaug.ALL or imgaug.parameters.StochasticParameter, optional
Interpolation order to use when rotating the kernel according to `angle`.
See :func:`imgaug.augmenters.geometric.Affine.__init__`.
Recommended to be ``0`` or ``1``, with ``0`` being faster, but less continuous/smooth as `angle` is changed,
particularly around multiple of 45 degrees.
name : None or str, optional
See :func:`imgaug.augmenters.meta.Augmenter.__init__`.
deterministic : bool, optional
See :func:`imgaug.augmenters.meta.Augmenter.__init__`.
random_state : None or int or numpy.random.RandomState, optional
See :func:`imgaug.augmenters.meta.Augmenter.__init__`.
Examples
--------
>>> aug = iaa.MotionBlur(k=15)
Create a motion blur augmenter with kernel size of 15x15.
>>> aug = iaa.MotionBlur(k=15, angle=[-45, 45])
Create a motion blur augmenter with kernel size of 15x15 and a blur angle of either -45 or 45 degrees (randomly
picked per image).
"""
# TODO allow (1, None) and set to identity matrix if k == 1
k_param = iap.handle_discrete_param(k, "k", value_range=(3, None), tuple_to_uniform=True, list_to_choice=True,
allow_floats=False)
angle_param = iap.handle_continuous_param(angle, "angle", value_range=None, tuple_to_uniform=True,
list_to_choice=True)
direction_param = iap.handle_continuous_param(direction, "direction", value_range=(-1.0-1e-6, 1.0+1e-6),
tuple_to_uniform=True, list_to_choice=True)
def create_matrices(image, nb_channels, random_state_func):
# avoid cyclic import between blur and geometric
from . import geometric as iaa_geometric
# force discrete for k_sample via int() in case of stochastic parameter
k_sample = int(k_param.draw_sample(random_state=random_state_func))
angle_sample = angle_param.draw_sample(random_state=random_state_func)
direction_sample = direction_param.draw_sample(random_state=random_state_func)
k_sample = k_sample if k_sample % 2 != 0 else k_sample + 1
direction_sample = np.clip(direction_sample, -1.0, 1.0)
direction_sample = (direction_sample + 1.0) / 2.0
matrix = np.zeros((k_sample, k_sample), dtype=np.float32)
matrix[:, k_sample//2] = np.linspace(float(direction_sample), 1.0 - float(direction_sample), num=k_sample)
rot = iaa_geometric.Affine(rotate=angle_sample, order=order)
matrix = (rot.augment_image((matrix * 255).astype(np.uint8)) / 255.0).astype(np.float32)
return [matrix/np.sum(matrix)] * nb_channels
if name is None:
name = "Unnamed%s" % (ia.caller_name(),)
return iaa_convolutional.Convolve(create_matrices, name=name, deterministic=deterministic,
random_state=random_state)
- |-
def rolling_sharpe(returns, rolling_sharpe_window):
"""
Determines the rolling Sharpe ratio of a strategy.
Parameters
----------
returns : pd.Series
Daily returns of the strategy, noncumulative.
- See full explanation in tears.create_full_tear_sheet.
rolling_sharpe_window : int
Length of rolling window, in days, over which to compute.
Returns
-------
pd.Series
Rolling Sharpe ratio.
Note
-----
See https://en.wikipedia.org/wiki/Sharpe_ratio for more details.
"""
return returns.rolling(rolling_sharpe_window).mean() \
/ returns.rolling(rolling_sharpe_window).std() \
* np.sqrt(APPROX_BDAYS_PER_YEAR)
- |-
def mpi_weighted_mean(comm, local_name2valcount):
"""
Perform a weighted average over dicts that are each on a different node
Input: local_name2valcount: dict mapping key -> (value, count)
Returns: key -> mean
"""
all_name2valcount = comm.gather(local_name2valcount)
if comm.rank == 0:
name2sum = defaultdict(float)
name2count = defaultdict(float)
for n2vc in all_name2valcount:
for (name, (val, count)) in n2vc.items():
try:
val = float(val)
except ValueError:
if comm.rank == 0:
warnings.warn('WARNING: tried to compute mean on non-float {}={}'.format(name, val))
else:
name2sum[name] += val * count
name2count[name] += count
return {name : name2sum[name] / name2count[name] for name in name2sum}
else:
return {}
- source_sentence: |-
Generate and return the following encoder related substitution variables:
encoderSpecsStr:
For the base description file, this string defines the default
encoding dicts for each encoder. For example:
'__gym_encoder' : { 'fieldname': 'gym',
'n': 13,
'name': 'gym',
'type': 'SDRCategoryEncoder',
'w': 7},
'__address_encoder' : { 'fieldname': 'address',
'n': 13,
'name': 'address',
'type': 'SDRCategoryEncoder',
'w': 7}
encoderSchemaStr:
For the base description file, this is a list containing a
DeferredDictLookup entry for each encoder. For example:
[DeferredDictLookup('__gym_encoder'),
DeferredDictLookup('__address_encoder'),
DeferredDictLookup('__timestamp_timeOfDay_encoder'),
DeferredDictLookup('__timestamp_dayOfWeek_encoder'),
DeferredDictLookup('__consumption_encoder')],
permEncoderChoicesStr:
For the permutations file, this defines the possible
encoder dicts for each encoder. For example:
'__timestamp_dayOfWeek_encoder': [
None,
{'fieldname':'timestamp',
'name': 'timestamp_timeOfDay',
'type':'DateEncoder'
'dayOfWeek': (7,1)
},
{'fieldname':'timestamp',
'name': 'timestamp_timeOfDay',
'type':'DateEncoder'
'dayOfWeek': (7,3)
},
],
'__field_consumption_encoder': [
None,
{'fieldname':'consumption',
'name': 'consumption',
'type':'AdaptiveScalarEncoder',
'n': 13,
'w': 7,
}
]
Parameters:
--------------------------------------------------
includedFields: item from the 'includedFields' section of the
description JSON object. This is a list of dicts, each
dict defining the field name, type, and optional min
and max values.
retval: (encoderSpecsStr, encoderSchemaStr permEncoderChoicesStr)
sentences:
- |-
def _generateEncoderStringsV1(includedFields):
""" Generate and return the following encoder related substitution variables:
encoderSpecsStr:
For the base description file, this string defines the default
encoding dicts for each encoder. For example:
'__gym_encoder' : { 'fieldname': 'gym',
'n': 13,
'name': 'gym',
'type': 'SDRCategoryEncoder',
'w': 7},
'__address_encoder' : { 'fieldname': 'address',
'n': 13,
'name': 'address',
'type': 'SDRCategoryEncoder',
'w': 7}
encoderSchemaStr:
For the base description file, this is a list containing a
DeferredDictLookup entry for each encoder. For example:
[DeferredDictLookup('__gym_encoder'),
DeferredDictLookup('__address_encoder'),
DeferredDictLookup('__timestamp_timeOfDay_encoder'),
DeferredDictLookup('__timestamp_dayOfWeek_encoder'),
DeferredDictLookup('__consumption_encoder')],
permEncoderChoicesStr:
For the permutations file, this defines the possible
encoder dicts for each encoder. For example:
'__timestamp_dayOfWeek_encoder': [
None,
{'fieldname':'timestamp',
'name': 'timestamp_timeOfDay',
'type':'DateEncoder'
'dayOfWeek': (7,1)
},
{'fieldname':'timestamp',
'name': 'timestamp_timeOfDay',
'type':'DateEncoder'
'dayOfWeek': (7,3)
},
],
'__field_consumption_encoder': [
None,
{'fieldname':'consumption',
'name': 'consumption',
'type':'AdaptiveScalarEncoder',
'n': 13,
'w': 7,
}
]
Parameters:
--------------------------------------------------
includedFields: item from the 'includedFields' section of the
description JSON object. This is a list of dicts, each
dict defining the field name, type, and optional min
and max values.
retval: (encoderSpecsStr, encoderSchemaStr permEncoderChoicesStr)
"""
# ------------------------------------------------------------------------
# First accumulate the possible choices for each encoder
encoderChoicesList = []
for fieldInfo in includedFields:
fieldName = fieldInfo['fieldName']
# Get the list of encoder choices for this field
(choicesList, aggFunction) = _generateEncoderChoicesV1(fieldInfo)
encoderChoicesList.extend(choicesList)
# ------------------------------------------------------------------------
# Generate the string containing the encoder specs and encoder schema. See
# the function comments for an example of the encoderSpecsStr and
# encoderSchemaStr
#
encoderSpecsList = []
for encoderChoices in encoderChoicesList:
# Use the last choice as the default in the base file because the 1st is
# often None
encoder = encoderChoices[-1]
# Check for bad characters
for c in _ILLEGAL_FIELDNAME_CHARACTERS:
if encoder['name'].find(c) >= 0:
raise _ExpGeneratorException("Illegal character in field: %r (%r)" % (
c, encoder['name']))
encoderSpecsList.append("%s: \n%s%s" % (
_quoteAndEscape(encoder['name']),
2*_ONE_INDENT,
pprint.pformat(encoder, indent=2*_INDENT_STEP)))
encoderSpecsStr = ',\n '.join(encoderSpecsList)
# ------------------------------------------------------------------------
# Generate the string containing the permutation encoder choices. See the
# function comments above for an example of the permEncoderChoicesStr
permEncoderChoicesList = []
for encoderChoices in encoderChoicesList:
permEncoderChoicesList.append("%s: %s," % (
_quoteAndEscape(encoderChoices[-1]['name']),
pprint.pformat(encoderChoices, indent=2*_INDENT_STEP)))
permEncoderChoicesStr = '\n'.join(permEncoderChoicesList)
permEncoderChoicesStr = _indentLines(permEncoderChoicesStr, 1,
indentFirstLine=False)
# Return results
return (encoderSpecsStr, permEncoderChoicesStr)
- |-
def shift(self, top=None, right=None, bottom=None, left=None):
"""
Shift/move the line strings from one or more image sides.
Parameters
----------
top : None or int, optional
Amount of pixels by which to shift all bounding boxes from the
top.
right : None or int, optional
Amount of pixels by which to shift all bounding boxes from the
right.
bottom : None or int, optional
Amount of pixels by which to shift all bounding boxes from the
bottom.
left : None or int, optional
Amount of pixels by which to shift all bounding boxes from the
left.
Returns
-------
imgaug.augmentables.lines.LineStringsOnImage
Shifted line strings.
"""
lss_new = [ls.shift(top=top, right=right, bottom=bottom, left=left)
for ls in self.line_strings]
return LineStringsOnImage(lss_new, shape=self.shape)
- |-
def cross_entropy_reward_loss(logits, actions, rewards, name=None):
"""Calculate the loss for Policy Gradient Network.
Parameters
----------
logits : tensor
The network outputs without softmax. This function implements softmax inside.
actions : tensor or placeholder
The agent actions.
rewards : tensor or placeholder
The rewards.
Returns
--------
Tensor
The TensorFlow loss function.
Examples
----------
>>> states_batch_pl = tf.placeholder(tf.float32, shape=[None, D])
>>> network = InputLayer(states_batch_pl, name='input')
>>> network = DenseLayer(network, n_units=H, act=tf.nn.relu, name='relu1')
>>> network = DenseLayer(network, n_units=3, name='out')
>>> probs = network.outputs
>>> sampling_prob = tf.nn.softmax(probs)
>>> actions_batch_pl = tf.placeholder(tf.int32, shape=[None])
>>> discount_rewards_batch_pl = tf.placeholder(tf.float32, shape=[None])
>>> loss = tl.rein.cross_entropy_reward_loss(probs, actions_batch_pl, discount_rewards_batch_pl)
>>> train_op = tf.train.RMSPropOptimizer(learning_rate, decay_rate).minimize(loss)
"""
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=actions, logits=logits, name=name)
return tf.reduce_sum(tf.multiply(cross_entropy, rewards))
- source_sentence: |-
Translate an index into coordinates, using the given coordinate system.
Similar to ``numpy.unravel_index``.
:param index: (int) The index of the point. The coordinates are expressed as a
single index by using the dimensions as a mixed radix definition. For
example, in dimensions 42x10, the point [1, 4] is index
1*420 + 4*10 = 460.
:param dimensions (list of ints) The coordinate system.
:returns: (list) of coordinates of length ``len(dimensions)``.
sentences:
- |-
def coordinatesFromIndex(index, dimensions):
"""
Translate an index into coordinates, using the given coordinate system.
Similar to ``numpy.unravel_index``.
:param index: (int) The index of the point. The coordinates are expressed as a
single index by using the dimensions as a mixed radix definition. For
example, in dimensions 42x10, the point [1, 4] is index
1*420 + 4*10 = 460.
:param dimensions (list of ints) The coordinate system.
:returns: (list) of coordinates of length ``len(dimensions)``.
"""
coordinates = [0] * len(dimensions)
shifted = index
for i in xrange(len(dimensions) - 1, 0, -1):
coordinates[i] = shifted % dimensions[i]
shifted = shifted / dimensions[i]
coordinates[0] = shifted
return coordinates
- |-
def step(self, observation, **extra_feed):
"""
Compute next action(s) given the observation(s)
Parameters:
----------
observation observation data (either single or a batch)
**extra_feed additional data such as state or mask (names of the arguments should match the ones in constructor, see __init__)
Returns:
-------
(action, value estimate, next state, negative log likelihood of the action under current policy parameters) tuple
"""
a, v, state, neglogp = self._evaluate([self.action, self.vf, self.state, self.neglogp], observation, **extra_feed)
if state.size == 0:
state = None
return a, v, state, neglogp
- |-
def pretty_eta(seconds_left):
"""Print the number of seconds in human readable format.
Examples:
2 days
2 hours and 37 minutes
less than a minute
Paramters
---------
seconds_left: int
Number of seconds to be converted to the ETA
Returns
-------
eta: str
String representing the pretty ETA.
"""
minutes_left = seconds_left // 60
seconds_left %= 60
hours_left = minutes_left // 60
minutes_left %= 60
days_left = hours_left // 24
hours_left %= 24
def helper(cnt, name):
return "{} {}{}".format(str(cnt), name, ('s' if cnt > 1 else ''))
if days_left > 0:
msg = helper(days_left, 'day')
if hours_left > 0:
msg += ' and ' + helper(hours_left, 'hour')
return msg
if hours_left > 0:
msg = helper(hours_left, 'hour')
if minutes_left > 0:
msg += ' and ' + helper(minutes_left, 'minute')
return msg
if minutes_left > 0:
return helper(minutes_left, 'minute')
return 'less than a minute'
- source_sentence: Validates control dictionary for the experiment context
sentences:
- >-
def load_file_list(path=None, regx='\.jpg', printable=True,
keep_prefix=False):
r"""Return a file list in a folder by given a path and regular expression.
Parameters
----------
path : str or None
A folder path, if `None`, use the current directory.
regx : str
The regx of file name.
printable : boolean
Whether to print the files infomation.
keep_prefix : boolean
Whether to keep path in the file name.
Examples
----------
>>> file_list = tl.files.load_file_list(path=None, regx='w1pre_[0-9]+\.(npz)')
"""
if path is None:
path = os.getcwd()
file_list = os.listdir(path)
return_list = []
for _, f in enumerate(file_list):
if re.search(regx, f):
return_list.append(f)
# return_list.sort()
if keep_prefix:
for i, f in enumerate(return_list):
return_list[i] = os.path.join(path, f)
if printable:
logging.info('Match file list = %s' % return_list)
logging.info('Number of files = %d' % len(return_list))
return return_list
- |-
def getCompletingSwarms(self):
"""Return the list of all completing swarms.
Parameters:
---------------------------------------------------------------------
retval: list of active swarm Ids
"""
swarmIds = []
for swarmId, info in self._state['swarms'].iteritems():
if info['status'] == 'completing':
swarmIds.append(swarmId)
return swarmIds
- |-
def __validateExperimentControl(self, control):
""" Validates control dictionary for the experiment context"""
# Validate task list
taskList = control.get('tasks', None)
if taskList is not None:
taskLabelsList = []
for task in taskList:
validateOpfJsonValue(task, "opfTaskSchema.json")
validateOpfJsonValue(task['taskControl'], "opfTaskControlSchema.json")
taskLabel = task['taskLabel']
assert isinstance(taskLabel, types.StringTypes), \
"taskLabel type: %r" % type(taskLabel)
assert len(taskLabel) > 0, "empty string taskLabel not is allowed"
taskLabelsList.append(taskLabel.lower())
taskLabelDuplicates = filter(lambda x: taskLabelsList.count(x) > 1,
taskLabelsList)
assert len(taskLabelDuplicates) == 0, \
"Duplcate task labels are not allowed: %s" % taskLabelDuplicates
return
pipeline_tag: sentence-similarity
library_name: sentence-transformers
SentenceTransformer based on answerdotai/ModernBERT-base
This is a sentence-transformers model finetuned from answerdotai/ModernBERT-base on the code_search_net dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
Model Details
Model Description
- Model Type: Sentence Transformer
- Base model: answerdotai/ModernBERT-base
- Maximum Sequence Length: 4096 tokens
- Output Dimensionality: 768 dimensions
- Similarity Function: Cosine Similarity
- Training Dataset:
- code_search_net
Model Sources
- Documentation: Sentence Transformers Documentation
- Repository: Sentence Transformers on GitHub
- Hugging Face: Sentence Transformers on Hugging Face
Full Model Architecture
SentenceTransformer(
(0): Transformer({'max_seq_length': 4096, 'do_lower_case': False}) with Transformer model: ModernBertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)
Usage
Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
pip install -U sentence-transformers
Then you can load this model and run inference.
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("juanwisz/modernbert-python-code-retrieval")
# Run inference
sentences = [
'Validates control dictionary for the experiment context',
'def __validateExperimentControl(self, control):\n """ Validates control dictionary for the experiment context"""\n # Validate task list\n taskList = control.get(\'tasks\', None)\n if taskList is not None:\n taskLabelsList = []\n\n for task in taskList:\n validateOpfJsonValue(task, "opfTaskSchema.json")\n validateOpfJsonValue(task[\'taskControl\'], "opfTaskControlSchema.json")\n\n taskLabel = task[\'taskLabel\']\n\n assert isinstance(taskLabel, types.StringTypes), \\\n "taskLabel type: %r" % type(taskLabel)\n assert len(taskLabel) > 0, "empty string taskLabel not is allowed"\n\n taskLabelsList.append(taskLabel.lower())\n\n taskLabelDuplicates = filter(lambda x: taskLabelsList.count(x) > 1,\n taskLabelsList)\n assert len(taskLabelDuplicates) == 0, \\\n "Duplcate task labels are not allowed: %s" % taskLabelDuplicates\n\n return',
'def load_file_list(path=None, regx=\'\\.jpg\', printable=True, keep_prefix=False):\n r"""Return a file list in a folder by given a path and regular expression.\n\n Parameters\n ----------\n path : str or None\n A folder path, if `None`, use the current directory.\n regx : str\n The regx of file name.\n printable : boolean\n Whether to print the files infomation.\n keep_prefix : boolean\n Whether to keep path in the file name.\n\n Examples\n ----------\n >>> file_list = tl.files.load_file_list(path=None, regx=\'w1pre_[0-9]+\\.(npz)\')\n\n """\n if path is None:\n path = os.getcwd()\n file_list = os.listdir(path)\n return_list = []\n for _, f in enumerate(file_list):\n if re.search(regx, f):\n return_list.append(f)\n # return_list.sort()\n if keep_prefix:\n for i, f in enumerate(return_list):\n return_list[i] = os.path.join(path, f)\n\n if printable:\n logging.info(\'Match file list = %s\' % return_list)\n logging.info(\'Number of files = %d\' % len(return_list))\n return return_list',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
Training Details
Training Dataset
code_search_net
- Dataset: code_search_net
- Size: 412,178 training samples
- Columns:
queryandpositive - Approximate statistics based on the first 1000 samples:
query positive type string string details - min: 4 tokens
- mean: 73.72 tokens
- max: 2258 tokens
- min: 46 tokens
- mean: 300.87 tokens
- max: 3119 tokens
- Samples:
query positive Extracts the list of arguments that start with any of the specified prefix valuesdef findArgs(args, prefixes):
"""
Extracts the list of arguments that start with any of the specified prefix values
"""
return list([
arg for arg in args
if len([p for p in prefixes if arg.lower().startswith(p.lower())]) > 0
])Removes any arguments in the supplied list that are contained in the specified blacklistdef stripArgs(args, blacklist):
"""
Removes any arguments in the supplied list that are contained in the specified blacklist
"""
blacklist = [b.lower() for b in blacklist]
return list([arg for arg in args if arg.lower() not in blacklist])Executes a child process and captures its outputdef capture(command, input=None, cwd=None, shell=False, raiseOnError=False):
"""
Executes a child process and captures its output
"""
# Attempt to execute the child process
proc = subprocess.Popen(command, stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE, cwd=cwd, shell=shell, universal_newlines=True)
(stdout, stderr) = proc.communicate(input)
# If the child process failed and we were asked to raise an exception, do so
if raiseOnError == True and proc.returncode != 0:
raise Exception(
'child process ' + str(command) +
' failed with exit code ' + str(proc.returncode) +
'\nstdout: "' + stdout + '"' +
'\nstderr: "' + stderr + '"'
)
return CommandOutput(proc.returncode, stdout, stderr) - Loss:
MultipleNegativesRankingLosswith these parameters:{ "scale": 20.0, "similarity_fct": "cos_sim" }
Evaluation Dataset
code_search_net
- Dataset: code_search_net
- Size: 23,107 evaluation samples
- Columns:
queryandpositive - Approximate statistics based on the first 1000 samples:
query positive type string string details - min: 5 tokens
- mean: 168.27 tokens
- max: 2118 tokens
- min: 48 tokens
- mean: 467.9 tokens
- max: 4096 tokens
- Samples:
query positive Train a deepq model.
Parameters
-------
env: gym.Env
environment to train on
network: string or a function
neural network to use as a q function approximator. If string, has to be one of the names of registered models in baselines.common.models
(mlp, cnn, conv_only). If a function, should take an observation tensor and return a latent variable tensor, which
will be mapped to the Q function heads (see build_q_func in baselines.deepq.models for details on that)
seed: int or None
prng seed. The runs with the same seed "should" give the same results. If None, no seeding is used.
lr: float
learning rate for adam optimizer
total_timesteps: int
number of env steps to optimizer for
buffer_size: int
size of the replay buffer
exploration_fraction: float
fraction of entire training period over which the exploration rate is annealed
exploration_final_eps: float
final value of ra...def learn(env,
network,
seed=None,
lr=5e-4,
total_timesteps=100000,
buffer_size=50000,
exploration_fraction=0.1,
exploration_final_eps=0.02,
train_freq=1,
batch_size=32,
print_freq=100,
checkpoint_freq=10000,
checkpoint_path=None,
learning_starts=1000,
gamma=1.0,
target_network_update_freq=500,
prioritized_replay=False,
prioritized_replay_alpha=0.6,
prioritized_replay_beta0=0.4,
prioritized_replay_beta_iters=None,
prioritized_replay_eps=1e-6,
param_noise=False,
callback=None,
load_path=None,
**network_kwargs
):
"""Train a deepq model.
Parameters
-------
env: gym.Env
environment to train on
network: string or a function
neural network to use as a q function approximator. If string, has to be one of the ...Save model to a pickle located atpathdef save_act(self, path=None):
"""Save model to a pickle located atpath"""
if path is None:
path = os.path.join(logger.get_dir(), "model.pkl")
with tempfile.TemporaryDirectory() as td:
save_variables(os.path.join(td, "model"))
arc_name = os.path.join(td, "packed.zip")
with zipfile.ZipFile(arc_name, 'w') as zipf:
for root, dirs, files in os.walk(td):
for fname in files:
file_path = os.path.join(root, fname)
if file_path != arc_name:
zipf.write(file_path, os.path.relpath(file_path, td))
with open(arc_name, "rb") as f:
model_data = f.read()
with open(path, "wb") as f:
cloudpickle.dump((model_data, self._act_params), f)CNN from Nature paper.def nature_cnn(unscaled_images, **conv_kwargs):
"""
CNN from Nature paper.
"""
scaled_images = tf.cast(unscaled_images, tf.float32) / 255.
activ = tf.nn.relu
h = activ(conv(scaled_images, 'c1', nf=32, rf=8, stride=4, init_scale=np.sqrt(2),
**conv_kwargs))
h2 = activ(conv(h, 'c2', nf=64, rf=4, stride=2, init_scale=np.sqrt(2), **conv_kwargs))
h3 = activ(conv(h2, 'c3', nf=64, rf=3, stride=1, init_scale=np.sqrt(2), **conv_kwargs))
h3 = conv_to_fc(h3)
return activ(fc(h3, 'fc1', nh=512, init_scale=np.sqrt(2))) - Loss:
MultipleNegativesRankingLosswith these parameters:{ "scale": 20.0, "similarity_fct": "cos_sim" }
Training Hyperparameters
Non-Default Hyperparameters
eval_strategy: epochper_device_train_batch_size: 4gradient_accumulation_steps: 4learning_rate: 2e-05num_train_epochs: 10warmup_steps: 1000fp16: True
All Hyperparameters
Click to expand
overwrite_output_dir: Falsedo_predict: Falseeval_strategy: epochprediction_loss_only: Trueper_device_train_batch_size: 4per_device_eval_batch_size: 8per_gpu_train_batch_size: Noneper_gpu_eval_batch_size: Nonegradient_accumulation_steps: 4eval_accumulation_steps: Nonetorch_empty_cache_steps: Nonelearning_rate: 2e-05weight_decay: 0.0adam_beta1: 0.9adam_beta2: 0.999adam_epsilon: 1e-08max_grad_norm: 1.0num_train_epochs: 10max_steps: -1lr_scheduler_type: linearlr_scheduler_kwargs: {}warmup_ratio: 0.0warmup_steps: 1000log_level: passivelog_level_replica: warninglog_on_each_node: Truelogging_nan_inf_filter: Truesave_safetensors: Truesave_on_each_node: Falsesave_only_model: Falserestore_callback_states_from_checkpoint: Falseno_cuda: Falseuse_cpu: Falseuse_mps_device: Falseseed: 42data_seed: Nonejit_mode_eval: Falseuse_ipex: Falsebf16: Falsefp16: Truefp16_opt_level: O1half_precision_backend: autobf16_full_eval: Falsefp16_full_eval: Falsetf32: Nonelocal_rank: 0ddp_backend: Nonetpu_num_cores: Nonetpu_metrics_debug: Falsedebug: []dataloader_drop_last: Falsedataloader_num_workers: 0dataloader_prefetch_factor: Nonepast_index: -1disable_tqdm: Falseremove_unused_columns: Truelabel_names: Noneload_best_model_at_end: Falseignore_data_skip: Falsefsdp: []fsdp_min_num_params: 0fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap: Noneaccelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed: Nonelabel_smoothing_factor: 0.0optim: adamw_torchoptim_args: Noneadafactor: Falsegroup_by_length: Falselength_column_name: lengthddp_find_unused_parameters: Noneddp_bucket_cap_mb: Noneddp_broadcast_buffers: Falsedataloader_pin_memory: Truedataloader_persistent_workers: Falseskip_memory_metrics: Trueuse_legacy_prediction_loop: Falsepush_to_hub: Falseresume_from_checkpoint: Nonehub_model_id: Nonehub_strategy: every_savehub_private_repo: Nonehub_always_push: Falsegradient_checkpointing: Falsegradient_checkpointing_kwargs: Noneinclude_inputs_for_metrics: Falseinclude_for_metrics: []eval_do_concat_batches: Truefp16_backend: autopush_to_hub_model_id: Nonepush_to_hub_organization: Nonemp_parameters:auto_find_batch_size: Falsefull_determinism: Falsetorchdynamo: Noneray_scope: lastddp_timeout: 1800torch_compile: Falsetorch_compile_backend: Nonetorch_compile_mode: Nonedispatch_batches: Nonesplit_batches: Noneinclude_tokens_per_second: Falseinclude_num_input_tokens_seen: Falseneftune_noise_alpha: Noneoptim_target_modules: Nonebatch_eval_metrics: Falseeval_on_start: Falseuse_liger_kernel: Falseeval_use_gather_object: Falseaverage_tokens_across_devices: Falseprompts: Nonebatch_sampler: batch_samplermulti_dataset_batch_sampler: proportional
Training Logs
Click to expand
| Epoch | Step | Training Loss | Validation Loss |
|---|---|---|---|
| 0.0078 | 200 | 0.634 | - |
| 0.0155 | 400 | 0.0046 | - |
| 0.0233 | 600 | 0.0009 | - |
| 0.0311 | 800 | 0.0004 | - |
| 0.0388 | 1000 | 0.0001 | - |
| 0.0466 | 1200 | 0.0002 | - |
| 0.0543 | 1400 | 0.0001 | - |
| 0.0621 | 1600 | 0.0001 | - |
| 0.0699 | 1800 | 0.0001 | - |
| 0.0776 | 2000 | 0.0 | - |
| 0.0854 | 2200 | 0.0 | - |
| 0.0932 | 2400 | 0.0 | - |
| 0.1009 | 2600 | 0.0 | - |
| 0.1087 | 2800 | 0.0005 | - |
| 0.1165 | 3000 | 0.0005 | - |
| 0.1242 | 3200 | 0.0002 | - |
| 0.1320 | 3400 | 0.0 | - |
| 0.1397 | 3600 | 0.0 | - |
| 0.1475 | 3800 | 0.0 | - |
| 0.1553 | 4000 | 0.0001 | - |
| 0.1630 | 4200 | 0.0 | - |
| 0.1708 | 4400 | 0.0001 | - |
| 0.1786 | 4600 | 0.0001 | - |
| 0.1863 | 4800 | 0.0 | - |
| 0.1941 | 5000 | 0.0 | - |
| 0.2019 | 5200 | 0.0 | - |
| 0.2096 | 5400 | 0.0 | - |
| 0.2174 | 5600 | 0.0 | - |
| 0.2251 | 5800 | 0.0 | - |
| 0.2329 | 6000 | 0.0004 | - |
| 0.2407 | 6200 | 0.0 | - |
| 0.2484 | 6400 | 0.0001 | - |
| 0.2562 | 6600 | 0.0 | - |
| 0.2640 | 6800 | 0.0 | - |
| 0.2717 | 7000 | 0.0 | - |
| 0.2795 | 7200 | 0.0 | - |
| 0.2873 | 7400 | 0.0 | - |
| 0.2950 | 7600 | 0.0 | - |
| 0.3028 | 7800 | 0.0 | - |
| 0.3105 | 8000 | 0.0 | - |
| 0.3183 | 8200 | 0.0 | - |
| 0.3261 | 8400 | 0.0004 | - |
| 0.3338 | 8600 | 0.0 | - |
| 0.3416 | 8800 | 0.0 | - |
| 0.3494 | 9000 | 0.0 | - |
| 0.3571 | 9200 | 0.0 | - |
| 0.3649 | 9400 | 0.0 | - |
| 0.3727 | 9600 | 0.0 | - |
| 0.3804 | 9800 | 0.0 | - |
| 0.3882 | 10000 | 0.0 | - |
| 0.3959 | 10200 | 0.0 | - |
| 0.4037 | 10400 | 0.0 | - |
| 0.4115 | 10600 | 0.0 | - |
| 0.4192 | 10800 | 0.0 | - |
| 0.4270 | 11000 | 0.0 | - |
| 0.4348 | 11200 | 0.0 | - |
| 0.4425 | 11400 | 0.0 | - |
| 0.4503 | 11600 | 0.0 | - |
| 0.4581 | 11800 | 0.0 | - |
| 0.4658 | 12000 | 0.0 | - |
| 0.4736 | 12200 | 0.0 | - |
| 0.4813 | 12400 | 0.0 | - |
| 0.4891 | 12600 | 0.0005 | - |
| 0.4969 | 12800 | 0.0 | - |
| 0.5046 | 13000 | 0.0 | - |
| 0.5124 | 13200 | 0.0001 | - |
| 0.5202 | 13400 | 0.0 | - |
| 0.5279 | 13600 | 0.0 | - |
| 0.5357 | 13800 | 0.0 | - |
| 0.5435 | 14000 | 0.0 | - |
| 0.5512 | 14200 | 0.0 | - |
| 0.5590 | 14400 | 0.0004 | - |
| 0.5667 | 14600 | 0.0 | - |
| 0.5745 | 14800 | 0.0 | - |
| 0.5823 | 15000 | 0.0 | - |
| 0.5900 | 15200 | 0.0 | - |
| 0.5978 | 15400 | 0.0 | - |
| 0.6056 | 15600 | 0.0 | - |
| 0.6133 | 15800 | 0.0 | - |
| 0.6211 | 16000 | 0.0 | - |
| 0.6289 | 16200 | 0.0 | - |
| 0.6366 | 16400 | 0.0006 | - |
| 0.6444 | 16600 | 0.0 | - |
| 0.6521 | 16800 | 0.0005 | - |
| 0.6599 | 17000 | 0.0 | - |
| 0.6677 | 17200 | 0.0 | - |
| 0.6754 | 17400 | 0.0 | - |
| 0.6832 | 17600 | 0.0 | - |
| 0.6910 | 17800 | 0.0 | - |
| 0.6987 | 18000 | 0.0005 | - |
| 0.7065 | 18200 | 0.0001 | - |
| 0.7143 | 18400 | 0.0 | - |
| 0.7220 | 18600 | 0.0 | - |
| 0.7298 | 18800 | 0.0 | - |
| 0.7375 | 19000 | 0.0 | - |
| 0.7453 | 19200 | 0.0 | - |
| 0.7531 | 19400 | 0.0 | - |
| 0.7608 | 19600 | 0.0 | - |
| 0.7686 | 19800 | 0.0001 | - |
| 0.7764 | 20000 | 0.0 | - |
| 0.7841 | 20200 | 0.0 | - |
| 0.7919 | 20400 | 0.0 | - |
| 0.7997 | 20600 | 0.0004 | - |
| 0.8074 | 20800 | 0.0 | - |
| 0.8152 | 21000 | 0.0 | - |
| 0.8229 | 21200 | 0.0 | - |
| 0.8307 | 21400 | 0.0009 | - |
| 0.8385 | 21600 | 0.0 | - |
| 0.8462 | 21800 | 0.0 | - |
| 0.8540 | 22000 | 0.0 | - |
| 0.8618 | 22200 | 0.0 | - |
| 0.8695 | 22400 | 0.0002 | - |
| 0.8773 | 22600 | 0.0 | - |
| 0.8851 | 22800 | 0.0 | - |
| 0.8928 | 23000 | 0.0001 | - |
| 0.9006 | 23200 | 0.0 | - |
| 0.9083 | 23400 | 0.0 | - |
| 0.9161 | 23600 | 0.0 | - |
| 0.9239 | 23800 | 0.0 | - |
| 0.9316 | 24000 | 0.0 | - |
| 0.9394 | 24200 | 0.0 | - |
| 0.9472 | 24400 | 0.0 | - |
| 0.9549 | 24600 | 0.0 | - |
| 0.9627 | 24800 | 0.0 | - |
| 0.9704 | 25000 | 0.0 | - |
| 0.9782 | 25200 | 0.0 | - |
| 0.9860 | 25400 | 0.0 | - |
| 0.9937 | 25600 | 0.0 | - |
| 1.0 | 25762 | - | 0.0001 |
| 1.0015 | 25800 | 0.0005 | - |
| 1.0092 | 26000 | 0.0 | - |
| 1.0170 | 26200 | 0.0 | - |
| 1.0248 | 26400 | 0.0 | - |
| 1.0325 | 26600 | 0.0 | - |
| 1.0403 | 26800 | 0.0 | - |
| 1.0481 | 27000 | 0.0 | - |
| 1.0558 | 27200 | 0.0 | - |
| 1.0636 | 27400 | 0.0 | - |
| 1.0713 | 27600 | 0.0 | - |
| 1.0791 | 27800 | 0.0 | - |
| 1.0869 | 28000 | 0.0 | - |
| 1.0946 | 28200 | 0.0 | - |
| 1.1024 | 28400 | 0.0 | - |
| 1.1102 | 28600 | 0.0 | - |
| 1.1179 | 28800 | 0.0 | - |
| 1.1257 | 29000 | 0.0 | - |
| 1.1335 | 29200 | 0.0 | - |
| 1.1412 | 29400 | 0.0 | - |
| 1.1490 | 29600 | 0.0 | - |
| 1.1567 | 29800 | 0.0 | - |
| 1.1645 | 30000 | 0.0 | - |
| 1.1723 | 30200 | 0.0 | - |
| 1.1800 | 30400 | 0.0 | - |
| 1.1878 | 30600 | 0.0 | - |
| 1.1956 | 30800 | 0.0 | - |
| 1.2033 | 31000 | 0.0 | - |
| 1.2111 | 31200 | 0.0 | - |
| 1.2189 | 31400 | 0.0 | - |
| 1.2266 | 31600 | 0.0004 | - |
| 1.2344 | 31800 | 0.0004 | - |
| 1.2421 | 32000 | 0.0 | - |
| 1.2499 | 32200 | 0.0 | - |
| 1.2577 | 32400 | 0.0 | - |
| 1.2654 | 32600 | 0.0 | - |
| 1.2732 | 32800 | 0.0 | - |
| 1.2810 | 33000 | 0.0 | - |
| 1.2887 | 33200 | 0.0 | - |
| 1.2965 | 33400 | 0.0 | - |
| 1.3043 | 33600 | 0.0 | - |
| 1.3120 | 33800 | 0.0 | - |
| 1.3198 | 34000 | 0.0 | - |
| 1.3275 | 34200 | 0.0 | - |
| 1.3353 | 34400 | 0.0 | - |
| 1.3431 | 34600 | 0.0 | - |
| 1.3508 | 34800 | 0.0004 | - |
| 1.3586 | 35000 | 0.0005 | - |
| 1.3664 | 35200 | 0.0004 | - |
| 1.3741 | 35400 | 0.0011 | - |
| 1.3819 | 35600 | 0.0 | - |
| 1.3897 | 35800 | 0.0 | - |
| 1.3974 | 36000 | 0.0 | - |
| 1.4052 | 36200 | 0.0 | - |
| 1.4129 | 36400 | 0.0 | - |
| 1.4207 | 36600 | 0.0 | - |
| 1.4285 | 36800 | 0.0 | - |
| 1.4362 | 37000 | 0.0 | - |
| 1.4440 | 37200 | 0.0001 | - |
| 1.4518 | 37400 | 0.0 | - |
| 1.4595 | 37600 | 0.0 | - |
| 1.4673 | 37800 | 0.0 | - |
| 1.4751 | 38000 | 0.0 | - |
| 1.4828 | 38200 | 0.0004 | - |
| 1.4906 | 38400 | 0.0003 | - |
| 1.4983 | 38600 | 0.0 | - |
| 1.5061 | 38800 | 0.0 | - |
| 1.5139 | 39000 | 0.0 | - |
| 1.5216 | 39200 | 0.0 | - |
| 1.5294 | 39400 | 0.0004 | - |
| 1.5372 | 39600 | 0.0004 | - |
| 1.5449 | 39800 | 0.0 | - |
| 1.5527 | 40000 | 0.0 | - |
| 1.5605 | 40200 | 0.0 | - |
| 1.5682 | 40400 | 0.0 | - |
| 1.5760 | 40600 | 0.0009 | - |
| 1.5837 | 40800 | 0.0 | - |
| 1.5915 | 41000 | 0.0009 | - |
| 1.5993 | 41200 | 0.0 | - |
| 1.6070 | 41400 | 0.0 | - |
| 1.6148 | 41600 | 0.0 | - |
| 1.6226 | 41800 | 0.0 | - |
| 1.6303 | 42000 | 0.0 | - |
| 1.6381 | 42200 | 0.0 | - |
| 1.6459 | 42400 | 0.0 | - |
| 1.6536 | 42600 | 0.0 | - |
| 1.6614 | 42800 | 0.0 | - |
| 1.6691 | 43000 | 0.0 | - |
| 1.6769 | 43200 | 0.0 | - |
| 1.6847 | 43400 | 0.0 | - |
| 1.6924 | 43600 | 0.0 | - |
| 1.7002 | 43800 | 0.0 | - |
| 1.7080 | 44000 | 0.0 | - |
| 1.7157 | 44200 | 0.0 | - |
| 1.7235 | 44400 | 0.0 | - |
| 1.7313 | 44600 | 0.0 | - |
| 1.7390 | 44800 | 0.0 | - |
| 1.7468 | 45000 | 0.0 | - |
| 1.7545 | 45200 | 0.0 | - |
| 1.7623 | 45400 | 0.0 | - |
| 1.7701 | 45600 | 0.0 | - |
| 1.7778 | 45800 | 0.0 | - |
| 1.7856 | 46000 | 0.0 | - |
| 1.7934 | 46200 | 0.0 | - |
| 1.8011 | 46400 | 0.0 | - |
| 1.8089 | 46600 | 0.0 | - |
| 1.8167 | 46800 | 0.0 | - |
| 1.8244 | 47000 | 0.0 | - |
| 1.8322 | 47200 | 0.0 | - |
| 1.8399 | 47400 | 0.0 | - |
| 1.8477 | 47600 | 0.0 | - |
| 1.8555 | 47800 | 0.0004 | - |
| 1.8632 | 48000 | 0.0 | - |
| 1.8710 | 48200 | 0.0 | - |
| 1.8788 | 48400 | 0.0 | - |
| 1.8865 | 48600 | 0.0 | - |
| 1.8943 | 48800 | 0.0 | - |
| 1.9021 | 49000 | 0.0004 | - |
| 1.9098 | 49200 | 0.0 | - |
| 1.9176 | 49400 | 0.0 | - |
| 1.9253 | 49600 | 0.0004 | - |
| 1.9331 | 49800 | 0.0 | - |
| 1.9409 | 50000 | 0.0 | - |
| 1.9486 | 50200 | 0.0 | - |
| 1.9564 | 50400 | 0.0 | - |
| 1.9642 | 50600 | 0.0004 | - |
| 1.9719 | 50800 | 0.0 | - |
| 1.9797 | 51000 | 0.0 | - |
| 1.9875 | 51200 | 0.0 | - |
| 1.9952 | 51400 | 0.0004 | - |
| 2.0 | 51524 | - | 0.0001 |
| 2.0030 | 51600 | 0.0 | - |
| 2.0107 | 51800 | 0.0 | - |
| 2.0185 | 52000 | 0.0 | - |
| 2.0262 | 52200 | 0.0 | - |
| 2.0340 | 52400 | 0.0004 | - |
| 2.0418 | 52600 | 0.0004 | - |
| 2.0495 | 52800 | 0.0 | - |
| 2.0573 | 53000 | 0.0008 | - |
| 2.0651 | 53200 | 0.0 | - |
| 2.0728 | 53400 | 0.0 | - |
| 2.0806 | 53600 | 0.0 | - |
| 2.0883 | 53800 | 0.0 | - |
| 2.0961 | 54000 | 0.0 | - |
| 2.1039 | 54200 | 0.0 | - |
| 2.1116 | 54400 | 0.0 | - |
| 2.1194 | 54600 | 0.0 | - |
| 2.1272 | 54800 | 0.0 | - |
| 2.1349 | 55000 | 0.0 | - |
| 2.1427 | 55200 | 0.0 | - |
| 2.1505 | 55400 | 0.0 | - |
| 2.1582 | 55600 | 0.0 | - |
| 2.1660 | 55800 | 0.0 | - |
| 2.1737 | 56000 | 0.0 | - |
| 2.1815 | 56200 | 0.0 | - |
| 2.1893 | 56400 | 0.0 | - |
| 2.1970 | 56600 | 0.0 | - |
| 2.2048 | 56800 | 0.0 | - |
| 2.2126 | 57000 | 0.0 | - |
| 2.2203 | 57200 | 0.0 | - |
| 2.2281 | 57400 | 0.0 | - |
| 2.2359 | 57600 | 0.0 | - |
| 2.2436 | 57800 | 0.0 | - |
| 2.2514 | 58000 | 0.0004 | - |
| 2.2591 | 58200 | 0.0 | - |
| 2.2669 | 58400 | 0.0004 | - |
| 2.2747 | 58600 | 0.0 | - |
| 2.2824 | 58800 | 0.0 | - |
| 2.2902 | 59000 | 0.0 | - |
| 2.2980 | 59200 | 0.0 | - |
| 2.3057 | 59400 | 0.0 | - |
| 2.3135 | 59600 | 0.0 | - |
| 2.3213 | 59800 | 0.0004 | - |
| 2.3290 | 60000 | 0.0 | - |
| 2.3368 | 60200 | 0.0004 | - |
| 2.3445 | 60400 | 0.0 | - |
| 2.3523 | 60600 | 0.0 | - |
| 2.3601 | 60800 | 0.0 | - |
| 2.3678 | 61000 | 0.0 | - |
| 2.3756 | 61200 | 0.0 | - |
| 2.3834 | 61400 | 0.0 | - |
| 2.3911 | 61600 | 0.0 | - |
| 2.3989 | 61800 | 0.0 | - |
| 2.4067 | 62000 | 0.0005 | - |
| 2.4144 | 62200 | 0.0 | - |
| 2.4222 | 62400 | 0.0 | - |
| 2.4299 | 62600 | 0.0 | - |
| 2.4377 | 62800 | 0.0 | - |
| 2.4455 | 63000 | 0.0 | - |
| 2.4532 | 63200 | 0.0 | - |
| 2.4610 | 63400 | 0.0 | - |
| 2.4688 | 63600 | 0.0 | - |
| 2.4765 | 63800 | 0.0 | - |
| 2.4843 | 64000 | 0.0 | - |
| 2.4921 | 64200 | 0.0 | - |
| 2.4998 | 64400 | 0.0 | - |
| 2.5076 | 64600 | 0.0 | - |
| 2.5153 | 64800 | 0.0 | - |
| 2.5231 | 65000 | 0.0 | - |
| 2.5309 | 65200 | 0.0 | - |
| 2.5386 | 65400 | 0.0 | - |
| 2.5464 | 65600 | 0.0004 | - |
| 2.5542 | 65800 | 0.0 | - |
| 2.5619 | 66000 | 0.0 | - |
| 2.5697 | 66200 | 0.0 | - |
| 2.5775 | 66400 | 0.0 | - |
| 2.5852 | 66600 | 0.0 | - |
| 2.5930 | 66800 | 0.0 | - |
| 2.6007 | 67000 | 0.0 | - |
| 2.6085 | 67200 | 0.0 | - |
| 2.6163 | 67400 | 0.0 | - |
| 2.6240 | 67600 | 0.0 | - |
| 2.6318 | 67800 | 0.0 | - |
| 2.6396 | 68000 | 0.0 | - |
| 2.6473 | 68200 | 0.0 | - |
| 2.6551 | 68400 | 0.0 | - |
| 2.6629 | 68600 | 0.0 | - |
| 2.6706 | 68800 | 0.0004 | - |
| 2.6784 | 69000 | 0.0 | - |
| 2.6861 | 69200 | 0.0 | - |
| 2.6939 | 69400 | 0.0 | - |
| 2.7017 | 69600 | 0.0004 | - |
| 2.7094 | 69800 | 0.0004 | - |
| 2.7172 | 70000 | 0.0 | - |
| 2.7250 | 70200 | 0.0 | - |
| 2.7327 | 70400 | 0.0 | - |
| 2.7405 | 70600 | 0.0 | - |
| 2.7483 | 70800 | 0.0 | - |
| 2.7560 | 71000 | 0.0004 | - |
| 2.7638 | 71200 | 0.0 | - |
| 2.7715 | 71400 | 0.0 | - |
| 2.7793 | 71600 | 0.0 | - |
| 2.7871 | 71800 | 0.0 | - |
| 2.7948 | 72000 | 0.0 | - |
| 2.8026 | 72200 | 0.0 | - |
| 2.8104 | 72400 | 0.0 | - |
| 2.8181 | 72600 | 0.0 | - |
| 2.8259 | 72800 | 0.0 | - |
| 2.8337 | 73000 | 0.0004 | - |
| 2.8414 | 73200 | 0.0 | - |
| 2.8492 | 73400 | 0.0 | - |
| 2.8569 | 73600 | 0.0 | - |
| 2.8647 | 73800 | 0.0004 | - |
| 2.8725 | 74000 | 0.0 | - |
| 2.8802 | 74200 | 0.0 | - |
| 2.8880 | 74400 | 0.0 | - |
| 2.8958 | 74600 | 0.0 | - |
| 2.9035 | 74800 | 0.0 | - |
| 2.9113 | 75000 | 0.0 | - |
| 2.9191 | 75200 | 0.0 | - |
| 2.9268 | 75400 | 0.0004 | - |
| 2.9346 | 75600 | 0.0 | - |
| 2.9423 | 75800 | 0.0 | - |
| 2.9501 | 76000 | 0.0 | - |
| 2.9579 | 76200 | 0.0 | - |
| 2.9656 | 76400 | 0.0 | - |
| 2.9734 | 76600 | 0.0004 | - |
| 2.9812 | 76800 | 0.0 | - |
| 2.9889 | 77000 | 0.0 | - |
| 2.9967 | 77200 | 0.0 | - |
| 3.0 | 77286 | - | 0.0000 |
Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.3.1
- Transformers: 4.48.0
- PyTorch: 2.5.1+cu121
- Accelerate: 1.2.1
- Datasets: 3.2.0
- Tokenizers: 0.21.0
Citation
BibTeX
ModernBERT
@misc{warner2024smarterbetterfasterlonger,
title={Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference},
author={Benjamin Warner and Antoine Chaffin and Benjamin Clavié and Orion Weller and Oskar Hallström and Said Taghadouini and Alexis Gallagher and Raja Biswas and Faisal Ladhak and Tom Aarsen and Nathan Cooper and Griffin Adams and Jeremy Howard and Iacopo Poli},
year={2024},
eprint={2412.13663},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2412.13663},
}
Sentence Transformers
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
MultipleNegativesRankingLoss
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}