saliency

library0.2.1pypypi✓ verified 89d ago

Framework-agnostic library for computing saliency maps (e.g., integrated gradients, SmoothGrad, XRAI) for deep learning models. Current version: 0.2.1. Release cadence is low, with updates driven by research contributions.

ai-ml

pip install saliency

github ↗package ↗

INSTALL

IMPORT

SIG · SALIENCY

saliency

ai-mlpythonv0.2.1

Install

10.6s avg

Import

1599ms

Disk

307MB

Pass rate

8/ 10

Env Coverage8 / 10

glibc

3.9–3.13

musl

3.9–3.13

Install & Compatibility

Where this runs

tested against v0.2.1 · pip install

no network on importno background threads

Install × environment matrix

Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux

musl

glibc

py 3.10

✕ build_error

✓ 10.4s

py 3.11

✓ —

✓ 10.2s

py 3.12

✓ —

✓ 10.4s

py 3.13

✓ —

✓ 9.9s

py 3.9

✕ build_error

✓ 12s

307MB installed

● package 307MB

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

core

✓ from saliency import core

✗ import saliency

The top-level module does not expose submodules directly; use explicit import.

IntegratedGradients

✓ from saliency.core import IntegratedGradients

✗ from saliency import IntegratedGradients

Classes are in submodules like core, not directly in saliency.

SmoothGrad

✓ from saliency.core import SmoothGrad

✗ from saliency import SmoothGrad

Same as above; use correct submodule path.

XRAI

✓ from saliency.xrai import XRAI

XRAI is in its own submodule 'xrai', not in core.

visualize

✓ from saliency.core import visualize

✗ from saliency import visualize

Visualization utilities are in core.

Compute integrated gradients on a simple Keras model.

import tensorflow as tf from saliency.core import IntegratedGradients, visualize # Build a simple model model = tf.keras.Sequential([ tf.keras.layers.Dense(10, activation='relu', input_shape=(4,)), tf.keras.layers.Dense(1, activation='sigmoid') ]) # Dummy input and baseline x_input = tf.constant([[1.0, 2.0, 3.0, 4.0]]) baseline = tf.zeros_like(x_input) # Call model wrapper def model_fn(x): return model(x) # Compute integrated gradients ig = IntegratedGradients() attributions = ig.GetMask(x_input, model_fn, baseline, x_steps=25) print(attributions)

Debug

Known issues

breakingIn version 0.2.0, the API was overhauled: previous methods like `saliency.IntegratedGradients` moved to `saliency.core`. Code using old import paths will break.

fix

Update imports: from saliency.core import IntegratedGradients (and others). Also update method calls (e.g., GetMask instead of GetMask).

affects: <0.2.0

deprecatedThe method `GetMask` in IntegratedGradients may be deprecated in future in favor of `compute_saliency` or similar naming. Check CHANGELOG.

fix

Monitor repository for updated API; aim to use any newer method names once released.

affects: >=0.2.0

gotchaXRAI method requires both positive and negative attributions; using only positive attributions will produce incorrect masks.

fix

Ensure you pass the full attributions tensor (including negative values) to XRAI, not a rectified version.

affects: all

gotchaThe library expects models to output logits (pre-softmax) for gradient calculations. Using softmax outputs may lead to vanishing gradients.

fix

Always use a model that returns logits, or modify the model function to return logits before activation.

affects: all

Errors

Common errors & fixes

ModuleNotFoundError: No module named 'saliency.core'

Older version installed (<0.2.0) where core module didn't exist.

fix

Upgrade to latest: pip install --upgrade saliency

AttributeError: module 'saliency' has no attribute 'IntegratedGradients'

Importing directly from top-level saliency instead of submodule.

fix

Use: from saliency.core import IntegratedGradients

TypeError: GetMask() missing 1 required positional argument: 'x_steps'

x_steps parameter is required since version 0.2.0 (old default removed).

fix

Provide x_steps explicitly, e.g., GetMask(..., x_steps=25)

Upgrade

Version history

0.2.1latest on PyPI · released Mar 20, 2024

Audit

Dependencies

numpyrequiredCore dependency for array operations.

pillowrequiredImage loading and visualization.

tensorflowoptionalDefault backend for model integration.

torchoptionalAlternative backend for PyTorch models.

Agent activity

16 hits · last 30 days

node

OpenAI (training)

Resources

githubhttps://github.com/pair-code/saliency ↗

packagehttps://pypi.org/project/saliency/ ↗