fix: Fix Ivy Failing Test: paddle - elementwise.multiply #28056

sgalpha01 · 2024-01-26T02:46:42Z

PR Description

paddle doesn't support bfloat16, that's why tests were failing.

❯ pytest ivy_tests/test_ivy/test_functional/test_core/test_elementwise.py::test_multiply --verbose
Use Database in ReadOnly Mode with local caching !
===================================================================== test session starts =====================================================================
platform linux -- Python 3.10.0, pytest-7.4.4, pluggy-1.3.0 -- /home/r0gue_shinobi/.miniconda3/envs/ivy_dev/bin/python3.10
cachedir: .pytest_cache
hypothesis profile 'ivy_profile' -> max_examples=25, print_blob=True, deadline=timedelta(milliseconds=500000), suppress_health_check=[HealthCheck.too_slow, HealthCheck.filter_too_much, HealthCheck.data_too_large], database=MultiplexedDatabase(DirectoryBasedExampleDatabase(PosixPath('/home/r0gue_shinobi/Desktop/ivy/.hypothesis/examples')), ReadOnlyDatabase(RedisExampleDatabase(Redis<ConnectionPool<Connection<host=redis-17011.c259.us-central1-2.gce.cloud.redislabs.com,port=17011,db=0>>>, expire_after=datetime.timedelta(days=8))))
metadata: {'Python': '3.10.0', 'Platform': 'Linux-6.6.6-76060606-generic-x86_64-with-glibc2.35', 'Packages': {'pytest': '7.4.4', 'pluggy': '1.3.0'}, 'Plugins': {'anyio': '4.2.0', 'hypothesis': '6.96.1', 'metadata': '3.0.0', 'json-report': '1.5.0'}}
backend(s): all
device: cpu
number of Hypothesis examples: 25
rootdir: /home/r0gue_shinobi/Desktop/ivy
plugins: anyio-4.2.0, hypothesis-6.96.1, metadata-3.0.0, json-report-1.5.0
collected 5 items                                                                                                                                             

ivy_tests/test_ivy/test_functional/test_core/test_elementwise.py::test_multiply[cpu-numpy-False-False] PASSED                                           [ 20%]
ivy_tests/test_ivy/test_functional/test_core/test_elementwise.py::test_multiply[cpu-jax-False-False] PASSED                                             [ 40%]
ivy_tests/test_ivy/test_functional/test_core/test_elementwise.py::test_multiply[cpu-tensorflow-False-False] PASSED                                      [ 60%]
ivy_tests/test_ivy/test_functional/test_core/test_elementwise.py::test_multiply[cpu-torch-False-False] PASSED                                           [ 80%]
ivy_tests/test_ivy/test_functional/test_core/test_elementwise.py::test_multiply[cpu-paddle-False-False] PASSED                                          [100%]

====================================================================== 100.0% of 5 passed ==================================================================================================

Related Issue

Closes #28055

Checklist

Did you add a function?
Did you add the tests?
Did you run your tests and are your tests passing?
Did pre-commit not fail on any check?
Did you follow the steps we provided?

Socials

samthakur587

Hii @sgalpha01 can you please remove the changes you made in importing modules.

samthakur587 · 2024-02-05T10:43:31Z

ivy/functional/backends/paddle/elementwise.py

 from ivy import promote_types_of_inputs
 from ivy.func_wrapper import (
-    with_unsupported_device_and_dtypes,
    with_supported_device_and_dtypes,


Suggested change

with_supported_device_and_dtypes,

with_supported_device_and_dtypes, you don't need to change this.

Hey @samthakur587, the changes are simply reordering the functions imported in ascending order, which was automatically done by black. No imports are removed or added. Please let me know if you want me to revert the imports to their initial state.

can you please just update your branch to latest commit. i want to check it's still coming or not. just for review. 😄

Suggested change

with_supported_device_and_dtypes,

with_supported_device_and_dtypes,

Hello @samthakur587, I updated my branch to the latest commit. I'll now make the required changes you requested in subsequent messages.

samthakur587 · 2024-02-05T14:01:08Z

ivy/functional/backends/paddle/elementwise.py

-            "float16",
-        )
-    },
+    {"2.6.0 and below": ("int8", "uint8", "int16", "float16", "bfloat16")},
    backend_version,


it's good and working fine. but i think it's looks more understandable if you add @supported_dtype(float32, float64, int32, int64 ,complex).

and can you please implement code for complex dtype at paddle backend.

Suggested change

backend_version,

backend_version,

Adding test support for bool also as it works in the main library.

>>> x = paddle.to_tensor([[True, False], [False, True]]) >>> y = paddle.to_tensor([[False, False], [True, False]]) >>> x * y Tensor(shape=[2, 2], dtype=bool, place=Place(cpu), stop_gradient=True, [[False, False], [False, False]])

Made the changes to handle complex dtype. Please review it.

Hii @sgalpha01 the changes you made looking good to me. i have tested it's passing all the tests. 👍🏾

…le_elementwise_multiply

samthakur587

one last change and your PR ready to merge. as i have seen that you added the support of bool dtype and ivy test_case by default it uses tf as ground truth backend and tf doesn't support the bool dtype as a result the bool dtype is not tested at any backend. 😄 so you have to make some changes in test_function.

after making this changes check it is passing all the tests at local.

samthakur587 · 2024-02-06T13:19:57Z

change helper dtype to valid and ground truth to torch.

sgalpha01 · 2024-02-06T13:50:29Z

@samthakur587, I made the changes in the test file and ran the tests locally. Please review it.

samthakur587

LGTM 🚀

fix: Fix Ivy Failing Test: paddle - elementwise.multiply

7dfff16

ivy-leaves added the Paddle Paddle Backend label Jan 26, 2024

ivy-seed assigned NiteshK84 Jan 26, 2024

ivy-leaves added the Priority label Jan 26, 2024

samthakur587 self-requested a review February 4, 2024 21:04

samthakur587 suggested changes Feb 5, 2024

View reviewed changes

samthakur587 reviewed Feb 5, 2024

View reviewed changes

samthakur587 self-assigned this Feb 5, 2024

samthakur587 reviewed Feb 5, 2024

View reviewed changes

sgalpha01 and others added 4 commits February 5, 2024 22:50

Merge branch 'main' of https://github.com/sgalpha01/ivy into fix_padd…

ba65764

…le_elementwise_multiply

added complex dtype support for elementwise multiply - paddle backend

3327c7d

Merge branch 'main' of https://github.com/sgalpha01/ivy into fix_padd…

28e70e9

…le_elementwise_multiply

Merge branch 'main' of https://github.com/sgalpha01/ivy into fix_padd…

f90f198

…le_elementwise_multiply

samthakur587 suggested changes Feb 6, 2024

View reviewed changes

updated test to handle bool dtype for elementwise multiply

13834ef

samthakur587 approved these changes Feb 8, 2024

View reviewed changes

samthakur587 merged commit d38a312 into ivy-llc:main Feb 8, 2024
50 of 141 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix: Fix Ivy Failing Test: paddle - elementwise.multiply #28056

fix: Fix Ivy Failing Test: paddle - elementwise.multiply #28056

sgalpha01 commented Jan 26, 2024

samthakur587 left a comment

samthakur587 Feb 5, 2024

sgalpha01 Feb 5, 2024

samthakur587 Feb 5, 2024

sgalpha01 Feb 5, 2024

samthakur587 Feb 5, 2024

samthakur587 Feb 5, 2024

sgalpha01 Feb 5, 2024

samthakur587 Feb 6, 2024

samthakur587 left a comment

samthakur587 commented Feb 6, 2024

sgalpha01 commented Feb 6, 2024

samthakur587 left a comment

	with_supported_device_and_dtypes,
	with_supported_device_and_dtypes, you don't need to change this.

fix: Fix Ivy Failing Test: paddle - elementwise.multiply #28056

fix: Fix Ivy Failing Test: paddle - elementwise.multiply #28056

Conversation

sgalpha01 commented Jan 26, 2024

PR Description

Related Issue

Checklist

Socials

samthakur587 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

samthakur587 left a comment

Choose a reason for hiding this comment

samthakur587 commented Feb 6, 2024

sgalpha01 commented Feb 6, 2024

samthakur587 left a comment

Choose a reason for hiding this comment