Skip to content

ModelVersion#

Representation of all metadata about a specific model version.

Initialization#

Initialize with the init_model_version() function or the class constructor.

import neptune

run = neptune.init_model_version(model="MODEL-ID")
from neptune import ModelVersion

model_version = ModelVersion(model="MODEL-ID")
If Neptune can't find your project name or API token

As a best practice, you should save your Neptune API token and project name as environment variables:

export NEPTUNE_API_TOKEN="h0dHBzOi8aHR0cHM6Lkc78ghs74kl0jv...Yh3Kb8"
export NEPTUNE_PROJECT="ml-team/classification"

Alternatively, you can pass the information when using a function that takes api_token and project as arguments:

run = neptune.init_run( # (1)!
    api_token="h0dHBzOi8aHR0cHM6Lkc78ghs74kl0jv...Yh3Kb8",  # your token here
    project="ml-team/classification",  # your full project name here
)
  1. Also works for init_model(), init_model_version(), init_project(), and integrations that create Neptune runs underneath the hood, such as NeptuneLogger or NeptuneCallback.

  2. API token: In the bottom-left corner, expand the user menu and select Get my API token.

  3. Project name: You can copy the path from the project details ( Edit project details).

If you haven't registered, you can log anonymously to a public project:

api_token=neptune.ANONYMOUS_API_TOKEN
project="common/quickstarts"

Make sure not to publish sensitive data through your code!

You can use the model_version object to:

  • Log and fetch metadata about a specific version of the model.
  • Manage the lifecycle stage of the model version.

Parameters

Name      Type Default     Description
with_id str, optional None The Neptune identifier of an existing model version to resume, such as "CLS-PRE-3". The identifier is stored in the object's sys/id field. If omitted or None is passed, a new model version is created.
name str, optional "Untitled" A custom name for the model version. You can use it as a human-readable ID and add it to the model versions table as a column (sys/name).
model str, optional None Identifier of the model for which the new version should be created. Required when creating a new model version. The identifier is stored in the model's sys/id field.
project str, optional None Name of a project in the form workspace-name/project-name. If None, the value of the NEPTUNE_PROJECT environment variable is used.
api_token str, optional None Your Neptune API token (or a service account's API token). If None, the value of the NEPTUNE_API_TOKEN environment variable is used.

To keep your token secure, avoid placing it in source code. Instead, save it as an environment variable.

mode str, optional async Connection mode in which the logging will work. Possible values are async, sync, read-only, and debug.

If you leave it out, the value of the NEPTUNE_MODE environment variable is used. If that's not set, the default async is used.

flush_period float, optional 5 (seconds) In asynchronous (default) connection mode, how often Neptune should trigger disk flushing.
proxies dict, optional None Argument passed to HTTP calls made via the Requests library. For details on proxies, see the Requests documentation.
async_lag_callback NeptuneObjectCallback, optional None Custom callback function which is called if the lag between a queued operation and its synchronization with the server exceeds the duration defined by async_lag_threshold. The callback should take a ModelVersion object as the argument and can contain any custom code, such as calling stop() on the object.

Note: Instead of using this argument, you can use Neptune's default callback by setting the NEPTUNE_ENABLE_DEFAULT_ASYNC_LAG_CALLBACK environment variable to TRUE.

async_lag_threshold float, optional 1800.0 (seconds) Duration between the queueing and synchronization of an operation. If a lag callback (default callback enabled via environment variable or custom callback passed to the async_lag_callback argument) is enabled, the callback is called when this duration is exceeded.
async_no_progress_callback NeptuneObjectCallback, optional None Custom callback function which is called if there has been no synchronization progress whatsoever for the duration defined by async_no_progress_threshold. The callback should take a ModelVersion object as the argument and can contain any custom code, such as calling stop() on the object.

Note: Instead of using this argument, you can use Neptune's default callback by setting the NEPTUNE_ENABLE_DEFAULT_ASYNC_NO_PROGRESS_CALLBACK environment variable to TRUE.

async_no_progress_threshold float, optional 300.0 (seconds) For how long there has been no synchronization progress. If a no-progress callback (default callback enabled via environment variable or custom callback passed to the async_no_progress_callback argument) is enabled, the callback is called when this duration is exceeded.

Returns

ModelVersion object that is used to manage the model version and log metadata to it.

Examples

Connect to an existing model version with the identifier CLS-PRE-8
from neptune import ModelVersion

model_version = ModelVersion(with_id="CLS-PRE-8")

Field lookup: []#

You can access the field of a model version through a dict-like field lookup: model_version[field_path].

This way, you can

  • store metadata:

    model_version["model/binary"].upload("model.pt")
    model_version["validation/dataset"].track_files("s3://datasets/validation")
    model_version["validation/metrics/acc"] = 0.98
    
  • fetch already logged metadata – for example, validation metrics for the model version:

    model_version = neptune.init_model_version(with_id="CLS-TREE-3")
    val_acc = model_version["validation/metrics/acc"].fetch()
    if val_acc >= ACC_THRESHOLD:
        model_version.change_stage("staging")
    

Returns

The returned type depends on the field type and whether a field is stored under the given path.

Path Example Returns
Field exists - The returned type matches the type of the field
Field does not exist - Handler object
Path is namespace and has field

Path: "model"

Field "model/binary" exists

Namespace handler object

Example

import neptune

model_version = neptune.init_model_version(model="CLS-PRE")

# Create new Float field
model_version["validation/acc"] = 50
# Update the value of the field
model_version["validation/acc"] = 100
# Error - it's no longer possible to store a File under a Float field
model_version["validation/acc"].upload("acc.bmp")  # Error

# Create new Series fields
model_version["train/logs"].append("Model registry, day 1:")
# Continue logging to existing Series fields
model_version["train/logs"].append("A model version is born")

# If you access a namespace handler, you can interact with it like an object
val_ns = model_version["validation"]
val_ns["accuracy"] = 0.76  # Stores 0.76 under path "validation/accuracy"

Assignment: =#

Convenience alias for assign().


assign()#

Assign values to multiple fields from a dictionary. You can use this method to store multiple pieces of metadata with a single command.

Parameters

Name Type Default Description
value dict None A dictionary with values to assign, where keys (str) become the paths of the fields. The dictionary can be nested, in which case the path will be a combination of all keys.
wait Boolean, optional False By default, logged metadata is sent to the server in the background. With this option set to True, Neptune first sends all data before executing the call. See Connection modes.

Example

import neptune

model_version = neptune.init_model_version(model="CLS-TREE")

# Assign multiple fields from a dictionary
model_info = {"size_limit": 50.0, "size_units": "MB"}
model_version["model"] = model_info

# You can also store metadata piece by piece
model_version["model/size_limit"] = 50.0
model_version["model/size_units"] = "MB"

# Dictionaries can be nested
model_info = {"size": {"limit": 50.0}}
model_version["model"] = model_version
# This will store the number 50.0 under path "model/size/limit"

change_stage()#

Changes the stage of the model version. The available stages are:

  • None
  • Staging
  • Production
  • Archived

In order to change the stage of the model version directly, you need contributor permissions within the project.

This method is always synchronous, which means that Neptune will wait for all other calls to reach the Neptune servers before executing it.

Parameters

Name Type Default Description
stage str None The new stage of the model version. Possible values are "none", "staging", "production", and "archived".

Example

import neptune

model_version = neptune.init_model_version(with_id="CLS-TREE-3")

# If the model is good enough, promote it to the staging
val_acc = model_version["validation/metrics/acc"].fetch()

if val_acc >= ACC_THRESHOLD:
    model_version.change_stage("staging")

del#

Completely removes the field or namespace and all associated metadata stored under the path.

If you need to free up storage or reduce the number of fields, can use del to remove unneeded metadata that takes up space, such as notebook training checkpoints.

See also: pop().

Examples

import neptune

model_version = neptune.init_model_version(with_id="CLS-TREE-3")

# Delete field with path "model/binary"
del model_version["model/binary"]

# You can also delete the whole namespace
del model_version["model"]

exists()#

Checks if there is a field or namespace under the specified path.

Info

This method checks the local representation of the model version. If the field was created by another process or the metadata has not reached the Neptune servers, it may not be possible to fetch. In this case you can:

  • Call sync() on the model_version object to synchronize the local representation with the server.
  • Call wait() on the model_version object to wait for all logging calls to finish.

Parameters

Name Type Default Description
path str - Path to check for the existence of a field or namespace

Examples

import neptune

model_version = neptune.init_model_version(with_id="CLS-PRE-8")

# If an old dataset exists, remove it
if model_version.exists("dataset/v0.4"):
    del model_version["dataset/v0.4"]

Info

When working in asynchronous (default) mode, the metadata you track may not be immediately available to fetch from the server, even if it appears in the local representation.

To work around this, you can call wait() on the model_version object.

import neptune

model_version = neptune.init_model_version(with_id="CLS-TREE-3")

model_version["model/binary"].upload("model.pt")

# The path exists in the local representation
if model_version.exists("model/binary"):
    # However, the tracking call may have not reached Neptune servers yet
    model_version["model/binary"].download()  # Error: the field does not exist

model_version.wait()

fetch()#

Fetches the values of all single-value fields (that are not of type File) as a dictionary.

The result preserves the hierarchical structure of the model metadata.

Returns

dict containing the values of all non-File single-value fields.

Example

import neptune

model_version = neptune.init_model_version(with_id="CLS-TREE-3")

# Fetch all the validation metrics
val_metrics = model_version["validation/metrics"].fetch()

get_structure()#

Returns the metadata structure of a ModelVersion object in the form of a dictionary.

This method can be used to traverse the metadata structure programmatically when using Neptune in automated workflows.

See also: print_structure().

Warning

The returned object is a shallow copy of the internal structure. Any modifications to it may result in tracking malfunction.

Returns

dict with the metadata structure of the model version.

Example

>>> import neptune
>>> model_version = neptune.init_model_version(with_id="CLS-PRE-8")
>>> model_version.get_structure()
{'model': {'binary': <neptune.attributes.atoms.file.File object at 0x000001C8EF8A1690>, 'parameters': {'eta': <neptune.attributes.atoms.float.Float object at 0x000001C8EF8A1780>, 'gamma': <neptune.attributes.atoms.float.Float object at 0x000001C8EF8A1840>, 'max_depth': <neptune.attributes.atoms.integer.Integer object at 0x000001C8EF8A1900>}}, ... }}

get_url()#

Returns a direct link to model version in Neptune. The same link is printed in the console once the model_version object has been initialized.

Returns

str with the URL of the model version in Neptune.

Example

>>> import neptune
>>> model_version = neptune.init_model_version(with_id="CLS-PRE-3")
>>> model_version.get_url()
https://app.neptune.ai/ml-team/classification/m/CLS-PRE/v/CLS-PRE-3

pop()#

Completely removes the field or namespace and all associated metadata stored under the path.

See also del.

Parameters

Name Type Default Description
path str - Path of the field or namespace to be removed.
wait Boolean, optional False By default, logged metadata is sent to the server in the background. With this option set to True, Neptune first sends all data before executing the call. See Connection modes.

Examples

import neptune

model_version = neptune.init_model_version(with_id="CLS-TREE-3")

# Delete a field along with its data
model_version.pop("model/binary")

You can invoke pop() directly on fields and namespaces.

# The following line
model_version.pop("model/binary")
# is equiavlent to this line
model_version["model/binary"].pop()
# or this line
model_version["model"].pop("binary")

# You can also batch-delete the whole namespace
model_version["model"].pop()

Pretty-prints the structure of the model metadata. Paths are ordered lexicographically and the structure is colored.

See also: get_structure().

Example

>>> import neptune
>>> model_version = neptune.init_model_version(with_id="CLS-PRE-8")
>>> model_version.print_structure()
'model':
    'binary': File
    'parameters':
        'eta': Float
        'gamma': Float
        'max_depth': Integer
'sys':
    'creation_time': Datetime
    'id': String
    'model_id': String
    'modification_time': Datetime
    'monitoring_time': Integer
    'name': String
    'owner': String
    'ping_time': Datetime
    'running_time': Float
    'size': Float
    'stage': String
    'state': RunState
    'tags': StringSet
    'trashed': Boolean
'train':
    'acc': Float
    'dataset': Artifact
'val':
    'acc': Float
    'dataset': Artifact

stop()#

Stops the connection to Neptune and synchronizes all data.

When using context managers, Neptune automatically calls stop() when exiting the ModelVersion context.

Warning

Always call stop() in interactive environments, such as a Python interpreter or Jupyter notebook. The connection to Neptune is not stopped when the cell has finished executing, but rather when the entire notebook stops.

If you're running a script, the connection is stopped automatically when the script finishes executing. However, it's a best practice to call stop() when the connection is no longer needed.

Parameters

Name Type Default Description
seconds int or float, optional None Wait for the specified time for all logging calls to finish before stopping the connection. If None, wait for all logging calls to finish.

Examples

If you initializing the connection from a Python script, Neptune stops it automatically when the script finishes executing.

import neptune

model_version = neptune.init_model_version(model="CLS-TREE")

[...] # Your code

# stop() is automatically called at the end for every Neptune object

If you are initializing multiple connections from one script, it's a good practice to call stop() for any unneeded connections:

import neptune

for model_version_id in model_versions:
    model_version = neptune.init_model_version(version=model_version_id)
    [...] # Your code
    model_version.stop()

Using with statement and context manager:

for model_version_id in model_versions:
    with neptune.init_model_version(version=model_version_id) as model_version:
        [...] # Your code
        # stop() is automatically called
        # when code execution exits the with statement

sync()#

Synchronizes the local representation of the model version with Neptune servers.

Parameters

Name Type Default Description
wait Boolean, optional False By default, logged metadata is sent to the server in the background. With this option set to True, Neptune first sends all data before executing the call. See Connection modes.

wait()#

Wait for all the logging calls to finish.

Parameters

Name Type Default Description
disk_only Boolean, optional False If True, the process will wait only for the data to be saved locally from memory, but will not wait for it to reach Neptune servers.