I have a list of parameters and I would like to sum all the elements in the list
import torch
from torch import nn
a = nn.Parameter(torch.rand(1))
b = nn.Parameter(torch.rand(1))
my_list = [a, b]
torch.sum(*my_list)
I receive the error
Traceback (most recent call last):
File "<input>", line 8, in <module>
TypeError: sum() received an invalid combination of arguments - got (Parameter, Parameter), but expected one of:
* (Tensor input, *, torch.dtype dtype)
* (Tensor input, tuple of ints dim, bool keepdim, *, torch.dtype dtype, Tensor out)
* (Tensor input, tuple of names dim, bool keepdim, *, torch.dtype dtype, Tensor out)
I was wondering if there is a way to perform operations like torch.sum on Parameters?
You may want to access a for example which is a nn.Parameter()
with a.data that returns a torch tensor. I hope that helps.
try to print a's .data field to get some intuition:
print(a.data)
you need to concatenate tensors using torch.stack and then sum.
import torch
from torch import nn
a=nn.Parameter(torch.rand(1))
b=nn.Parameter(torch.rand(1))
l=[a,b]
l1=torch.stack(l,dim=0)
torch.sum(l)
Related
This question already has an answer here:
scipy curve_fit doesn't like math module
(1 answer)
Closed 2 years ago.
I am trying to calculate probabilities from Gaussian Distribution by varying the standard deviation(std).
I am expecting by using the Gaussian Quadrature with 21 points, integrating over a range from -1 to +1 with mean = 0 and setting std= 1 and std =2 will yield p = 0.68 and p = 0.95 respectively (picture attached).
import scipy.integrate as integrate
import math as m
#mean=0
#varying the sigma
def f(sigma,x):
return m.exp(-1*(x**2)/(2*sigma**2))/(sigma*m.sqrt(2*m.pi))
def prob_at_nsigma(sigma):
value = 0.
ans,anserr =integrate.quadrature(f,-1,1,args=(sigma ,),maxiter=21)
value = ans
return value
print(prob_at_nsigma(1))
And I get the following errors which I don't see why the "divide by zero" and "only size-1 arrays can be converted to Python scalars" arise:
runfile('C:/pythonExe/ntu_cp/untitled0.py', wdir='C:/pythonExe/ntu_cp')
C:\pythonExe\ntu_cp\untitled0.py:14: RuntimeWarning: divide by zero encountered in true_divide
return m.exp(-1*(x**2)/(2*sigma**2))/(sigma*m.sqrt(2*m.pi))
C:\pythonExe\ntu_cp\untitled0.py:14: RuntimeWarning: invalid value encountered in true_divide
return m.exp(-1*(x**2)/(2*sigma**2))/(sigma*m.sqrt(2*m.pi))
Traceback (most recent call last):
File "C:\pythonExe\ntu_cp\untitled0.py", line 23, in <module>
print(prob_at_nsigma(1))
File "C:\pythonExe\ntu_cp\untitled0.py", line 19, in prob_at_nsigma
ans,anserr =integrate.quadrature(f,-1,1,args=(sigma ,),maxiter=21)
File "C:\Users\cztee\anaconda3\lib\site-packages\scipy\integrate\_quadrature.py", line 238, in quadrature
newval = fixed_quad(vfunc, a, b, (), n)[0]
File "C:\Users\cztee\anaconda3\lib\site-packages\scipy\integrate\_quadrature.py", line 119, in fixed_quad
return (b-a)/2.0 * np.sum(w*func(y, *args), axis=-1), None
File "C:\Users\cztee\anaconda3\lib\site-packages\scipy\integrate\_quadrature.py", line 149, in vfunc
return func(x, *args)
File "C:\pythonExe\ntu_cp\untitled0.py", line 14, in f
return m.exp(-1*(x**2)/(2*sigma**2))/(sigma*m.sqrt(2*m.pi))
TypeError: only size-1 arrays can be converted to Python scalars
Appreciate any help. Thanks!
Gaussian distribution
Two problems in your code:
As mentioned, you should use numpy instead of math module.
Your f function's argument order is wrong. Scipy assumes that the first variable is x followed by other parameters. So it should be f(x, sigma), not f(sigma, x).
Solution:
import numpy as np
import scipy.integrate as integrate
def f(x, sigma):
return np.exp(-1*(x**2)/(2*sigma**2))/(sigma*np.sqrt(2*np.pi))
def prob_at_nsigma(sigma):
value = 0.
ans, anserr = integrate.quadrature(f, -1, 1, args=(sigma,), maxiter=21)
return ans
print(prob_at_nsigma(1))
# 0.6826894922280757
I can create a numpy array from a python list as follows:
>>> a = [1,2,3]
>>> b = np.array(a).reshape(3,1)
>>> print(b)
[[1]
[2]
[3]]
However, I don't know what causes error in the following code:
Code :
>>> a = [1,2,3]
>>> b = np.full((3,1), a)
Error :
ValueError Traceback (most recent call last)
<ipython-input-275-1ab6c109dda4> in <module>()
1 a = [1,2,3]
----> 2 b = np.full((3,1), a)
3 print(b)
/usr/local/lib/python3.6/dist-packages/numpy/core/numeric.py in full(shape, fill_value, dtype, order)
324 dtype = array(fill_value).dtype
325 a = empty(shape, dtype, order)
--> 326 multiarray.copyto(a, fill_value, casting='unsafe')
327 return a
328
<__array_function__ internals> in copyto(*args, **kwargs)
ValueError: could not broadcast input array from shape (3) into shape (3,1)
Even though the list a has 3 elements inside it and I expect a 3x1 numpy array, the full() method fails to deliver it.
I referred the broadcasting article of numpy too. However, they are much more focused towards the arithmetic operation perspective, hence I couldn't obtain anything useful from there.
So it would be great if you can help me to understand the difference in b/w. the above mentioned array creation methods and the cause of the error too.
Numpy is unable to broadcast the two shapes together because your list is interpreted as a 'row vector' (np.array(a).shape = (3,)) while you are asking for a 'column vector' (shape = (3, 1)). If you are set on using np.full, then you can shape your list as a column vector initially:
>>> import numpy as np
>>>
>>> a = [[1],[2],[3]]
>>> b = np.full((3,1), a)
Another option is to convert a into a numpy array ahead of time and add a new axis to match the desired output shape.
>>> a = [1,2,3]
>>> a = np.array(a)[:, np.newaxis]
>>> b = np.full((3,1), a)
I'm trying to convert input_image which is a tensor to numpy array.Following the already answered questions here and several others that suggested to use input_image.eval() or equivalently sess.run() for this conversion, I did the same, but it throws an error and apparently expects a feed_dict value for the sess.run(). But since here I'm not trying to run an operation dependent on unknown values, I don't see the need for the feed_dict here because all I'm doing here is just conversion.
Besides, just so as to check I also tried converting a tf.constant([1,2,3]) value right above it using the same method and it got successfully compiled despite its data type being the same as input_image. Here's my code which is the part of larger script:
def call(self, x):
input_image = Input(shape=(None, None, 3))
print(input_image.shape)
print(type(tf.constant([1,2,3])))
print(type(input_image))
print(type(K.get_session().run(tf.constant([1,2,3]))))
print(type(K.get_session().run(input_image)))
and here's the error:
(?, ?, ?, 3)
<class 'tensorflow.python.framework.ops.Tensor'>
<class 'tensorflow.python.framework.ops.Tensor'>
<class 'numpy.ndarray'>
Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1365, in _do_call
return fn(*args)
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1350, in _run_fn
target_list, run_metadata)
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1443, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
(0) Invalid argument: You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,?,?,3]
[[{{node input_1}}]]
[[input_1/_1051]]
(1) Invalid argument: You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,?,?,3]
[[{{node input_1}}]]
0 successful operations.
0 derived errors ignored.
I wonder why the former would work and the latter won't.
There is no such thing as "converting" a symbolic tensor to a numpy array, as the latter cannot hold the same kind of information as the former.
When you use eval() or session.run(), what you are doing is evaluating a symbolic expression to get a numerical result, which is a numpy array, but this is not a conversion. Evaluating an expression might or might not require additional input data (that's what the feed_dict is for), depending on the expression.
Evaluating a constant (tf.constant) does not require any input data, but evaluating your other expression does require the input data, so you cannot "convert" this to a numpy array.
Just adding to (or elaborating on) what #MatiasValdenegro said,
TensorFlow follows something called graph execution (or define-then-run). In other words, when you write a TensorFlow program it defines something called a data-flow graph which shows how the operations you defined are related to each other. And then you execute bits and pieces of that graph depending on the results you're after.
Let's consider two examples. (I am switching to a simple TensorFlow program instead of Keras bits as it makes things more clear - After all K.get_session() returns a Session object).
Example 1
Say you have the following program.
import tensorflow as tf
a = tf.placeholder(shape=[2,2], dtype=tf.float32)
b = tf.constant(1, dtype=tf.float32)
c = a * b
# Wrong: This is what you're doing essentially when you do sess.run(input_image)
with tf.Session() as sess:
print(sess.run(c))
# Right: You need to feed values that c is dependent on
with tf.Session() as sess:
print(sess.run(c, feed_dict={a: np.array([[1,2],[2,3]])}))
Whenever a resulting tensor (e.g. c) is dependent on a placeholder you cannot execute it and get the result without feeding values to all the dependent placeholders.
Example 2
When you define a tf.constant(1) this is not dependent on anything. In other words you don't need a feed_dict and can directly run eval() or sess.run() on it.
Update: Further explanation on why you need a feed_dict for input_image
TLDR: You need a feed_dict because your resulting Tensor is produced by an Input layer.
Your input_image is basically the resulting tensor you get by feeding something to the Input layer. Usually in Keras, you are not exposed to the internal placeholder level details. But you would do that via using model.fit() or model.evaluate(). You can see that Keras Input layer in fact uses a placeholder by analysing this line.
Hope I made my point clear that you do need to feed in a value to the placeholder to successfully evaluate the output of an Input layer. Because that basically holds a placeholder.
Update 2: How to feed to your Input layer
So, appears you can use feed_dict with Keras Input layer in the following manner. Instead of defining shape argument you straight away pass a placeholder to the tensor argument, which will bypass the internal placeholder creation in the layer.
from tensorflow.keras.layers import InputLayer
import numpy as np
import tensorflow.keras.backend as K
x = tf.placeholder(shape=[None, None, None, 3], dtype=tf.float32)
input_image = Input(tensor=x)
arr = np.array([[[[1,1,1]]]])
print(arr.shape)
print(K.get_session().run(input_image, feed_dict={x: arr}))
I am trying to use the sklearn MinMaxScaler to rescale a python column like below:
scaler = MinMaxScaler()
y = scaler.fit(df['total_amount'])
But got the following errors:
Traceback (most recent call last):
File "/Users/edamame/workspace/git/my-analysis/experiments/my_seq.py", line 54, in <module>
y = scaler.fit(df['total_amount'])
File "/Users/edamame/workspace/git/my-analysis/venv/lib/python3.4/site-packages/sklearn/preprocessing/data.py", line 308, in fit
return self.partial_fit(X, y)
File "/Users/edamame/workspace/git/my-analysis/venv/lib/python3.4/site-packages/sklearn/preprocessing/data.py", line 334, in partial_fit
estimator=self, dtype=FLOAT_DTYPES)
File "/Users/edamame/workspace/git/my-analysis/venv/lib/python3.4/site-packages/sklearn/utils/validation.py", line 441, in check_array
"if it contains a single sample.".format(array))
ValueError: Expected 2D array, got 1D array instead:
array=[3.180000e+00 2.937450e+03 6.023850e+03 2.216292e+04 1.074589e+04
:
0.000000e+00 0.000000e+00 9.000000e+01 1.260000e+03].
Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample.
Any idea what was wrong?
The input to MinMaxScaler needs to be array-like, with shape [n_samples, n_features]. So you can apply it on the column as a dataframe rather than a series (using double square brackets instead of single):
y = scaler.fit(df[['total_amount']])
Though from your description, it sounds like you want fit_transform rather than just fit (but I could be wrong):
y = scaler.fit_transform(df[['total_amount']])
A little more explanation:
If your dataframe had 100 rows, consider the difference in shape when you transform a column to an array:
>>> np.array(df[['total_amount']]).shape
(100, 1)
>>> np.array(df['total_amount']).shape
(100,)
The first returns a shape that matches [n_samples, n_features] (as required by MinMaxScaler), whereas the second does not.
Try to do with this way:
import pandas as pd
from sklearn import preprocessing
x = df.values #returns a numpy array
min_max_scaler = preprocessing.MinMaxScaler()
x_scaled = min_max_scaler.fit_transform(x)
df = pd.DataFrame(x_scaled)
Using the following function i am trying to generate index from the data:
Function:
import numpy as np
from sklearn.decomposition import PCA
def pca_index(data,components=1,indx=1):
corrs = np.asarray(data.cov())
pca = PCA(n_components = components).fit(corrs)
trns = pca.transform(data)
index=np.dot(trns[0:indx],pca.explained_variance_ratio_[0:indx])
return index
Index: generation from principal components
index = pca_index(data=mydata,components=3,indx=2)
Following error is being generated when i am calling the function:
Traceback (most recent call last):
File "<ipython-input-411-35115ef28e61>", line 1, in <module>
index = pca_index(data=mydata,components=3,indx=2)
File "<ipython-input-410-49c0174a047a>", line 15, in pca_index
index=np.dot(trns[0:indx],pca.explained_variance_ratio_[0:indx])
ValueError: shapes (2,3) and (2,) not aligned: 3 (dim 1) != 2 (dim 0)
Can anyone help with the error.
According to my understanding there is some error at the following point when i am passing the subscript indices as variable (indx):
trns[0:indx],pca.explained_variance_ratio_[0:**indx**]
In np.dot you are trying to multiply a matrix having dimensions (2,3) with a matrix having dimensions (2,), i.e. a vector.
However, you can only multiply NxM to MxP, e.g. (3,2) to (2,1) or (2,3) to (3,1).
In your example the second matrix have dimensions of (2,) which, in numpy terms, is similar but not the same as (2,1). You can reshape a vector into a matrix with vector.reshape([2,1])
You might also transpose you first matrix, thus converting its dimensions from (2,3) to (3,2).
However, make sure that you multiply appropriate matrices as the result will differ from you might expect.