deeplake 3.9.52 incompatible with NumPy 2.x (NEP 50): TypeError in get_incompatible_dtype

[LuigiGonnella]

Bug Report

Environment

• deeplake==3.9.52
• numpy>=2.0 (tested on 2.4.2)

Description

NumPy 2.x (NEP 50) removed support for passing raw Python scalars (int, float, bool) as the first argument to np.can_cast. This causes a TypeError when DeepLake attempts to validate scalar samples in get_incompatible_dtype inside deeplake/util/casting.py.

Error

TypeError: can_cast() does not support Python ints, floats, and complex because the result used to depend on the value.

The offending line in deeplake/util/casting.py:

if np.can_cast(samples, dtype)  # samples is a raw Python float/int

Fix

A fix is available in this PR:
#3143

References

• NumPy 2.0 migration guide
• NEP 50 — Promotion rules for Python scalars