Neurodiffeq Versions Save

A library for solving differential equations using neural networks based on PyTorch, used by multiple research groups around the world, including at Harvard IACS.

v0.3.3

3 years ago
  1. add support for L-BFGS and other optimizers that require a closure for its .step method
  2. improve documentation
  3. improve test cases

v0.3.2

3 years ago
  1. Populated built-in callbacks,
  2. Improved __str__ and __repr__ methods of generators

v0.3.1

3 years ago
  • new condition: DoubleEndedBVP1D
  • more robust MonitorCallback
  • hierarchy issue fix in neurodiffeq.monitors

v0.3.0

3 years ago
  1. Refactored all solve* functions to Solver classes
  2. Renamed some arguments for consistency
  3. Implemented the Swish activation
  4. Cleaned & updated the documentation
  5. Miscellaneous fixes & enhancement

v0.2.2

3 years ago

This release focuses on new APIs and backward compatibility.

  • Addressed argument naming issues #70
  • Addressed ambiguity in FCNN's constructor #71
  • Addressed a potential bug in the diff operator
  • Included new test cases for warnings and errors

v0.2.1

3 years ago

Keep documentation up-to-date with 0.2

v0.2.0

3 years ago
  • all conditions are moved to conditions.py, with custom test cases implemented.
  • better styles for sphinx docs & misc docs fixes
  • fixed travis-ci bug and re-enabled code coverage
  • diff will check tensor shape by default
  • Basic Dash documentation support
  • use safe_diff internally and in test cases (except for test_temporal.py)
  • Functionality to set default dtype and device for tensors
  • Use 64-bit float (a.k.a. double) by default.
  • diff in v0.2.0 checks shape by default.

v0.1.1

3 years ago

Fix a potential bug in neurodiffeq.ode.IVP See #64 for details.

v0.1.0

3 years ago

After the discussion with David, we decided to host the repository on PyPI. Currently, the version is 0.1.0. Features along with breaking changes will be included in future releases.