Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fixed circular imports in init.py and autograd.py #5 #6

Closed
wants to merge 1 commit into from
Closed

Fixed circular imports in init.py and autograd.py #5 #6

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion python/needle/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
from . import init
from .autograd import Tensor, cpu, all_devices
from . import ops
from .ops import *
from . import init
from . import data
from . import nn
from . import optim
5 changes: 3 additions & 2 deletions python/needle/autograd.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,9 +1,10 @@
"""Core data structures."""
import needle
from typing import List, Optional, NamedTuple, Tuple, Union
from typing import List, Optional, NamedTuple, Tuple, Union, Dict
from collections import namedtuple
import numpy
from needle import init

from . import init

# needle version
LAZY_MODE = False
Expand Down
24 changes: 11 additions & 13 deletions python/needle/init.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,27 +1,25 @@
import math
import needle as ndl

from .autograd import Tensor, cpu


def rand(*shape, low=0.0, high=1.0, device=None, dtype="float32", requires_grad=False):
""" Generate random numbers uniform between low and high """
device = ndl.cpu() if device is None else device
device = cpu() if device is None else device
array = device.rand(*shape) * (high - low) + low
return ndl.Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)
return Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)


def randn(*shape, mean=0.0, std=1.0, device=None, dtype="float32", requires_grad=False):
""" Generate random normal with specified mean and std deviation """
device = ndl.cpu() if device is None else device
device = cpu() if device is None else device
array = device.randn(*shape) * std + mean
return ndl.Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)
return Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)


def constant(*shape, c=1.0, device=None, dtype="float32", requires_grad=False):
""" Generate constant Tensor """
device = ndl.cpu() if device is None else device
device = cpu() if device is None else device
array = device.ones(*shape, dtype=dtype) * c # note: can change dtype
return ndl.Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)
return Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)


def ones(*shape, device=None, dtype="float32", requires_grad=False):
Expand All @@ -36,15 +34,15 @@ def zeros(*shape, device=None, dtype="float32", requires_grad=False):

def randb(*shape, p=0.5, device=None, dtype="bool", requires_grad=False):
""" Generate binary random Tensor """
device = ndl.cpu() if device is None else device
device = cpu() if device is None else device
array = device.rand(*shape) <= p
return ndl.Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)
return Tensor(array, device=device, dtype=dtype, requires_grad=requires_grad)


def one_hot(n, i, device=None, dtype="float32", requires_grad=False):
""" Generate one-hot encoding Tensor """
device = ndl.cpu() if device is None else device
return ndl.Tensor(device.one_hot(n,i.numpy(), dtype=dtype), device=device, requires_grad=requires_grad)
device = cpu() if device is None else device
return Tensor(device.one_hot(n,i.numpy(), dtype=dtype), device=device, requires_grad=requires_grad)


def xavier_uniform(fan_in, fan_out, gain=1.0, **kwargs):
Expand Down