Unnamed: 0
int64
0
378k
id
int64
49.9k
73.8M
title
stringlengths
15
150
question
stringlengths
37
64.2k
answer
stringlengths
37
44.1k
tags
stringlengths
5
106
score
int64
-10
5.87k
372,000
54,600,851
numpy assignment with masking
<p>I am a newbie to Python. During an exercise I am supposed to use a mask to multiply all values below 100 in the following list by 2:</p> <pre><code>a = np.array([230, 10, 284, 39, 76]) </code></pre> <p>So I wrote the following code:</p> <pre><code>import numpy as np a = np.array([230, 10, 284, 39, 76]) cut = 100 ...
<p>Not exactly sure what you want, if you want to assign to places where <code>a &lt; cut</code> holds (<code>a &lt; cut = [0, 1, 0, 1, 1]</code> is the boolean index), when you assign to <code>a[a &lt; cut]</code>, you assign to the places where the element is 1, meaning on the right side it expects a numpy array of s...
python|numpy
5
372,001
54,635,294
Pandas: check if a value error is in an interval, error message: InvalidOperation: [<class 'decimal.InvalidOperation'>]
<p>My dataframe looks like this:</p> <pre><code>target_price interval 0.001767 [0.00318240, 0.00318624] 0.002978 [0.00318576, 0.00319673] 0.000174 [0.00319581, 0.00319617] 0.002740 [0.00318881, 0.00319617] </code></pre> <p>The code is used: <code>for index,interval in df.iterrows(): if interval.ta...
<p>If it is <code>list</code> </p> <pre><code>df['check']=[y[0]&lt;=x&lt;=y[1] for x , y in zip(df.target_price,df.interval)] Out[43]: [False, False, False, False] </code></pre> <p>If it is interval </p> <pre><code>df['check']=[x in y for x , y in zip(df.target_price,df.interval)] </code></pre> <p>---More info </p>...
pandas
1
372,002
54,616,753
Calculating readmission rate
<p>I am fairly new to Python and I am trying to calculate if a patient was readmitted to the hospital within 30 days or not. </p> <p>The data is in the form of Pandas dataframe with columns for Patient Id, Arrival Date, Departure Date and Status (Discharged, Admitted, Did Not Wait). The question is similar to this pas...
<p>You can use the below:</p> <pre><code>df.groupby('Patient').apply(lambda x : (x['Admission Date'].\ shift(-1)-x['Discharge date']).dt.days.le(30).astype(int)).reset_index(drop=True) </code></pre> <p><strong>Full code</strong>:</p> <p>Considering the df looks like:</p> <pre><code> Visit Patient Admission D...
python|python-3.x|pandas|dataframe
2
372,003
54,310,257
How to zip two lists of tuples by row?
<p>I have two lists like so:</p> <pre><code>list1 = [{'id':'1','id2':'2'},{'id':'2','id2':'3'}] list2 = [{'fname':'a','lname':'b'},{'fname':'c','lname':'d'}] </code></pre> <p>How do I combine the lists into one set of tuples for a pandas dataframe?</p> <p>like so:</p> <pre><code>final_list = [{'id':'1','id2':'2','f...
<p>A "pure" Python answer (ie no Pandas):</p> <pre><code>[{**x[0], **x[1]} for x in zip(list1, list2)] &gt; [{'id': '1', 'id2': '2', 'fname': 'a', 'lname': 'b'}, {'id': '2', 'id2': '3', 'fname': 'c', 'lname': 'd'}] </code></pre> <p>Edited by Scott Boston</p> <pre><code>pd.DataFrame([{**x[0], **x[1]} for x in zi...
python|python-3.x|pandas|list|dataframe
5
372,004
54,291,381
Why is there a problem when loading saved weights on a model
<p>I'm trying to modify a classifier model with many tools (dropout, autoencoder, etc...) to analyse what gets the best results. Thus, I am using the <code>save_weights</code> and <code>load_weights</code> methods. </p> <p>The first time I am launching my model, it works fine. However when loading the weights, the <co...
<p>To restart a training for a model, which was already used by the <code>fit()</code> function, you have to recompile it. </p> <pre><code>model.compile(optimizer='adam', loss = 'categorical_crossentropy', metrics = ['acc']) </code></pre> <p>The reason why is that the model has an optimizer assigned, which is in alre...
tensorflow|keras|google-colaboratory
1
372,005
54,429,176
Pandas DataFrame max, min and mean fails on columns with Nan
<p>I am trying to compute the max, min and mean of every column in a pandas DataFrame. I am however running into some trouble sanitizing my columns.</p> <p>One of my columns contains some "?"s instead of a value I tried to clean this by doing:</p> <pre><code>df = pd.read_csv("Auto.csv") df["horsepower"].replace("?",...
<p>As I mentioned in comment above , <code>dropna</code> will drop the entire row when there are any <code>NaN</code> values in it </p> <pre><code>df = pd.read_csv("Auto.csv") df["horsepower"].replace("?", np.nan, inplace=True) df["horsepower"]=pd.to_numeric(df["horsepower"],errors='coerce') </code></pre> <p>Using <...
python|pandas|numpy
2
372,006
54,457,331
How to type a variable that is passed to numpy.asarray to produce a 2D float array?
<p>I often write functions/methods that take some variable which can come in many forms, i.e., lists of lists, lists of tuples, tuples of tuples, etc. all containing numbers, that I want to convert into a numpy array, kinda like the following:</p> <pre class="lang-py prettyprint-override"><code>import numpy as np def...
<p>Try the <a href="https://github.com/numpy/numpy-stubs" rel="nofollow noreferrer">numpy-stubs: experimental typing stubs for NumPy</a>.</p> <p>It defines the type of the <code>np.array()</code> function like this:</p> <pre class="lang-py prettyprint-override"><code>def array( object: object, dtype: _DtypeLi...
python|numpy|type-hinting|mypy
1
372,007
54,583,072
Pandas read_sql - ignoring error if table does not exist
<p>I have a list of SQL scripts I have narrowed down and want to execute. The data in the list follows this pattern more or less:</p> <pre><code>[DROP TABLE ABC ....;, CREATE TABLE ABC ....;, INSERT INTO TABLE ABC .....;,UPDATE TABLE ABC .....;] </code></pre> <p>Then it repeats itself for the next table. All of thi...
<p>I figured it out, thanks to a combo of comments above. First, my Try/Except was in the incorrect location and also I switched from using pandas read_sql to just using regular session execute and it worked as expected. If table exists, drop it first, if not then create it.</p> <p>Revised code below:</p> <pre><cod...
python|pandas|teradata
1
372,008
54,370,302
Changing the order of entries for a geopandas choropleth map legend
<p>I am plotting a certain categorical value over the map of a city. The line of code I use to plot is the following:</p> <pre><code>fig = plt.figure(figsize=(12, 12)) ax = plt.gca() urban_data.plot(column="category", cmap="viridis", ax=ax, categorical=True, / k=4, legend=True, linewidth=0.5, / ...
<p>The bad news is that categories in legends produced by geopandas are sorted and this is hardcoded (<a href="https://github.com/geopandas/geopandas/blob/ddaa26fdba668a9afb53d9fd86158e6d25dc5ead/geopandas/plotting.py#L449" rel="noreferrer">see source-code here</a>).</p> <p>One solution is hence to have the categorica...
python|matplotlib|legend|geopandas|choropleth
5
372,009
54,514,564
Python and pandas pivot table sum between dates
<p>I have a pivot table which I have created using:</p> <pre><code>df = df[["Ref", # int64 "REGION", # object "COUNTRY", # object "Value_1", # float "Value_2", # float "Value_3", # float "Type", # object "Date", # float64 (may need to convert to date) ]...
<p>First use <a href="http://pandas.pydata.org/pandas-docs/stable/generated/pandas.cut.html" rel="nofollow noreferrer"><code>cut</code></a> with column <code>Year</code> and then aggregate by <a href="http://pandas.pydata.org/pandas-docs/stable/generated/pandas.core.groupby.DataFrameGroupBy.agg.html" rel="nofollow nore...
python-3.x|pandas|pivot-table
2
372,010
54,452,973
Pandas Dataframe TypeError: translate() takes exactly one argument (2 given)
<p>I am trying to delete punctuation and numbers to my pandas dataframe. here is my sample of code : </p> <pre><code>import re import string df.text = df.text.apply(lambda x: x.lower()) df.text = df.text.apply(lambda x: x.translate(None, string.punctuation)) </code></pre> <p>and it gives me error : </p> <blockquote>...
<p>You can use pandas' built-in <a href="https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.str.translate.html" rel="nofollow noreferrer"><code>Series.str.translate</code></a>:</p> <pre><code>In [1]: import pandas as pd In [2]: df = pd.DataFrame({'text': ['f!!o..o!', 'b""a??r', 'b?.?a!.!z']}) I...
python|pandas|python-2.7|dataframe
1
372,011
54,430,260
Pandas: Add a scalar to multiple new columns in an existing dataframe
<p>I recently answered a question where the OP was looking multiple columns with multiple different values to an existing dataframe (<a href="https://stackoverflow.com/a/54421876/6163621">link</a>). And it's fairly succinct, but I don't think very fast.</p> <p>Ultimately I was hoping I could do something like:</p> <...
<p>Why not using <code>assign</code> </p> <pre><code>df.assign(**dict.fromkeys(['b','c'],0)) Out[781]: a b c 0 1 0 0 1 2 0 0 </code></pre> <p>Or create the <code>dict</code> by <code>d=dict(zip([namelist],[valuelist]))</code></p>
python|pandas|dataframe
1
372,012
54,351,019
numpy matrix multiplication raises weird error
<p>The following surprisingly fails:</p> <pre><code># in a loop . try: pressure_book[element] = c @ np.matrix([1] + point).T except TypeError as e: print(c, type(c), d.type) print(point, type(point)) raise e . . </code></pre> <p>And outputs:</p> <pre><code>[[-1.52088384e+08 5.39161089e+03 9.0857665...
<p>It seems that depending on what<sup>1</sup> <code>point</code> contains, the conversion to <code>numpy.matrix</code> with the <code>dtype</code> unspecified might yield <code>object</code>.</p> <p><code>object</code>-<code>dtype</code> arrays \ matrices do not support matrix multiplication.</p> <p>To solve this is...
python|python-3.x|numpy|matrix-multiplication
0
372,013
54,402,611
How to properly build a tensor array dataset from a series of single values - tensorflow newbie
<p>I'm new to tensorflow and the dataset APIs. looks like I'm not feeding the correct lists of dicts to the tensorflow. I get the following output:</p> <pre><code>tensorflow.python.framework.errors_impl.InvalidArgumentError: In[0] is not a matrix. Instead it has shape [] [Op:MatMul] </code></pre> <p>My code is:</p> ...
<p>I have few observations:</p> <p>1) As far as I understood, the problem is with the format of the data you fill in. In the comments you said that your CSV file has a (100, 2) shape. However, you have specified a 10 nodes input layer. So, your neural network is expecting to receive 10 variables as input, but you onl...
python|tensorflow|tensorflow-datasets
0
372,014
54,417,710
add a column in Pandas
<p>I have problems to add a specific column "happiness_average" form variable pd_dictionnary of type "pandas" to another variable <code>pd_dictionnary_treat</code> of type "pandas"</p> <p>I tried to tape <code>pd_dictionnary_treat.append(pd_dictionnary["happiness_average"]</code>) in my code below: </p> <pre><code>i...
<p>Why are you use <code>append</code> function? You can directly add at the time of assigning <code>word</code> column to <code>pd_dictionnary_treat</code>. I think this will help you:</p> <pre><code>import pandas as pd pd_dictionnary=pd.read_csv("/Users/stefanhanssen/Dropbox/hse/word_list/131278/Data_Set_S1.txt",sep...
python|pandas
0
372,015
73,585,171
How to add a timestamp as the last column of an existing Pandas DataFrame
<p>I was looking for the most Pythonic solution available, but didn't see it published anywhere to my liking, so am submitting this as a self-answered question.</p>
<pre><code>df.insert(df.columns.size, 'createts', pd.to_datetime('now').replace(microsecond=0)) </code></pre>
python|pandas|dataframe
0
372,016
73,813,382
Pandas - take maximum/minimum of columns within group
<p>I have a Pandas dataframe df with column names par1,par2,..,par7. I want to select a subset of the total dataframe. Within the group par1,par2,par3 I want to select the rows for which par4 has the lowest value and par5 has the highest value in that group. So from the dataframe:</p> <pre><code>Par1,Par2,Par3,Par4,Par...
<p>Use <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.sort_values.html" rel="nofollow noreferrer"><code>DataFrame.sort_values</code></a> with <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.drop_duplicates.html" rel="nofollow noreferrer"><code>Dat...
pandas
1
372,017
73,785,154
Plotting multiple y-values versus x using Matplotlib
<p>I am trying to plot <code>y1,y2,y3</code> all as a function of <code>x</code> and show all the three on a single plot. But it is showing only two plots. I present the current output.</p> <pre><code>import matplotlib.pyplot as plt import numpy as np x=np.linspace(-1,0,10) print(x) y1=36.554+5.418*np.exp(-1.327*x) y2...
<p>Try this:</p> <pre><code>plt.plot(x, y1, x, y2, x, y3) </code></pre>
python|numpy|matplotlib
1
372,018
73,739,854
How to merge semi-duplicated rows in a Dataframe
<p>I have a dataframe that looks like this:</p> <div class="s-table-container"> <table class="s-table"> <thead> <tr> <th style="text-align: center;">ID</th> <th style="text-align: center;">Name</th> <th style="text-align: center;">app_A</th> <th style="text-align: center;">app_B</th> <th style="text-align: center;">app...
<p>Simple..aggregate <code>app</code> like columns with <code>sum</code> and <code>Total</code> with <code>first</code></p> <pre><code>c = df.filter(like='app_') df.groupby(['ID', 'Name']).agg({**dict.fromkeys(c, 'sum'), 'Total': 'first'}) </code></pre> <h4>Result</h4> <pre><code> app_A app_B app_C ...
python|pandas|dataframe|merge|duplicates
2
372,019
73,799,999
Issues with datetime and isinstance()
<p>I want to find if any variable in a dataset is a datetime. I'm having some serious problems with this seemingly simple task. In the dataset below, 'time' is transformed to datetime using pandas to_datetime(). It becomes a datetime64[ns] type, and, if I'm reading the documentation correct, to_datetime() should return...
<p>IIUC select one value to scalar, in your solution pass <code>Series</code>, so correctly failed:</p> <pre><code>print(isinstance(df.time.iat[0], datetime.datetime)) True </code></pre>
python|pandas|datetime
2
372,020
73,656,968
How to plus a number to a specific cell in pandas dataframe
<p>I am trying to +1 to a cell in pandas dataframe</p> <pre><code>staff['pax'][0]= staff['pax'][0]+1 </code></pre> <p>staff is the dataframe name, pax is the column name while 0 is the row I want to +1. However below is the error..</p> <pre><code>A value is trying to be set on a copy of a slice from a DataFrame See th...
<p>Please refer this documentation <a href="https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.at.html" rel="nofollow noreferrer">https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.at.html</a></p> <pre><code>df.at[0, 'pax'] = df['pax'][0]+1 </code></pre>
python|pandas|dataframe
1
372,021
73,673,387
How to I make a PyQtGraph scrolling graph clear the previous line within a loop
<p>I wish to plot some data from an array with multiple columns, and would like each column to be a different line on the same scrolling graph. As there are many columns, I think it would make sense to plot them within a loop. I'd also like to plot a second scrolling graph with a single line.</p> <p>I can get the sing...
<p>You <em>could</em> clear the plot of all curves each time with <code>.clear()</code>, but that wouldn't be very performant. A better solution would be to keep all the curve objects around and call <code>setData</code> on them each time, like you're doing with the single-curve plot. E.g.</p> <pre class="lang-py prett...
python|numpy|pyqtgraph
0
372,022
73,769,086
How to use multiple background colors for different string value in a particular row of csv file
<p>I have CSV file with multiple columns and I am searching some string in column 5 and highlighting the row for diffrent string. As of now I am able to use just two colours but I want to use more colors for more string into it. Example for</p> <pre><code> x['4']=&quot;abc&quot; ---&gt;I want to use yellow color ...
<p>Create dictionary for background colors by values and pass to <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.io.formats.style.Styler.applymap.html" rel="nofollow noreferrer"><code>Styler.applymap</code></a> for mapping by <code>dict.get</code>, if no match is returned empty value:</p> <pre...
python|pandas|dataframe|csv|data-analysis
0
372,023
73,652,149
Reducing the column to one element?
<p>I see a code like this:</p> <pre><code>boston=datasets.load_boston() X=boston.data[:, None, 6] y= boston.target </code></pre> <p>The site says &quot;To match the X and y dimensionality the X is reduced to 1 element in each row, by the following code&quot;. How is it 1 element? I am a total newbie in this field.</p>...
<p>Take a look at X:</p> <pre><code>from sklearn import datasets boston=datasets.load_boston() X=boston.data[:, None, 6] y= boston.target X[:10] </code></pre> <p>It looks like this:</p> <pre><code>array([[ 65.2], [ 78.9], [ 61.1], [ 45.8], [ 54.2], [ 58.7], [ 66.6], [ ...
pandas|dataframe
0
372,024
73,709,128
Interconversion and difference
<p>I have two types of tensors.</p> <pre><code>1) &lt;class 'tensorflow.python.ops.resource_variable_ops.ResourceVariable'&gt; 2) &lt;class 'tensorflow.python.framework.ops.Tensor'&gt; </code></pre> <p>What is the difference between the two? Can interconversion be done? In particular if I want to convert the second typ...
<p>A <a href="https://www.tensorflow.org/api_docs/python/tf/Tensor" rel="nofollow noreferrer"><code>tf.Tensor</code></a> is basically a multidimensional array of elements, while a <code>tf.ResourceVariable</code> is pretty much like a <a href="https://stackoverflow.com/q/40817665/14774959">fancier</a> <a href="https://...
loops|tensorflow|type-conversion|tensor
1
372,025
73,701,660
how to solve IndexError : single positional indexer is out-of-bounds
<pre><code>CODE:- from datetime import date from datetime import timedelta from nsepy import get_history import pandas as pd import datetime # import matplotlib.pyplot as mp end1 = date.today() start1 = end1 - timedelta(days=365) stock = [ 'RELIANCE'...
<p>The &quot;out-of-bounds&quot; error indicates you're trying to access a part of the dataframe series that doesn't exist. It's most likely caused by df['D_vol'] being less than 90 items long when you try to do</p> <pre><code>df['D_vol'].iloc[-91:-1] </code></pre> <p>Edit: add a length check before the offending line:...
python|pandas|numpy|datetime
1
372,026
73,673,540
Understanding the [:,1] in tf.stack
<p>Hello I am new to python and tensorflow. I read the code about swapping x and y coordiantes from bounding boxes and i am not sure if i understand the code correctly.</p> <pre><code>def swap_xy(boxes): return tf.stack([boxes[:, 1], boxes[:, 0], boxes[:, 3], boxes[:, 2]], axis=-1) </code></pre> <p>with tf.stack I pac...
<p>well this : is use to select all value example</p> <pre><code>lst=[1,2,3,4] print(lst[:2]) #output [1,2] </code></pre>
python|tensorflow
0
372,027
73,825,153
Subtract values from different groups
<p>I have the following DataFrame:</p> <pre><code> A X Time 1 a 10 2 b 17 3 b 20 4 c 21 5 c 36 6 d 40 </code></pre> <p>given by <code>pd.DataFrame({'Time': [1, 2, 3, 4, 5, 6], 'A': ['a', 'b', 'b', 'c', 'c', 'd'], 'X': [10, 17, 20, 21, 36, 40]}).set_index('Time')</code></p> <p>The de...
<p>Aggregate by <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.core.groupby.GroupBy.agg.html" rel="nofollow noreferrer"><code>GroupBy.agg</code></a> with <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.core.groupby.GroupBy.first.html" rel="nofollow noreferrer"><code>...
pandas|numpy
3
372,028
73,638,557
Pandas DataFrame: Based on a string in a row of col 'C', look that value up in col 'A', if found return the value of col 'B' for that matched row
<p>I'm looking for a cleaner solution to this problem than the one I have come up with:</p> <p>Based on the string value of a row in column 'C', look that value up in the string values of column 'A', and if found return the string value of column 'B' for that row.</p> <p>I have a df that looks like this (notice loc[1][...
<p>I prefer to apply a function in these scenarios:</p> <pre><code>def find_value(row: pd.Series) -&gt; pd.Series: val = df.query(&quot;@df['A'] == @row['C']&quot;) return row[&quot;B&quot;] if val.empty else val[&quot;B&quot;].squeeze() df[&quot;BB&quot;] = df.apply(lambda x: find_value(x), axis=1) </code><...
python|pandas|dataframe|merge|lookup
0
372,029
73,733,590
Pandas - drop rows based on two conditions on different columns
<p>Although there are several related questions answered in Pandas, I cannot solve this issue. I have a large dataframe (~ 49000 rows) and want to drop rows the meet two conditions at the same time(~ 120):</p> <ul> <li>For one column: an exact string</li> <li>For another column: a NaN value</li> </ul> <p>My code is ign...
<p>Instead of calling <code>drop</code>, and passing the <code>index</code>, You can create the mask for the condition for which you want to keep the rows, then take only those rows. Also, the logic error seems to be there, you are checking two different condition combined by <code>AND</code> for the same column values...
python|pandas|conditional-statements
2
372,030
73,805,458
PyTorch Datapipes and how does overwriting the datapipe classes work?
<p>Pytorch Datapipes are a new inplace dataset loaders for large data that can be fed into Pytorch models through streaming, for reference these are</p> <ul> <li>Official Doc: <a href="https://pytorch.org/data/main/tutorial.html" rel="nofollow noreferrer">https://pytorch.org/data/main/tutorial.html</a></li> <li>A crash...
<p>It looks like you're trying to chain together a series of torch <code>DataPipe</code>s, namely:</p> <ol> <li><a href="https://pytorch.org/data/0.4/generated/torchdata.datapipes.iter.FileOpener.html" rel="nofollow noreferrer">FileOpener</a> / <code>open_files</code></li> <li><a href="https://pytorch.org/data/0.4/gene...
python|machine-learning|pytorch|dataset|torchdata
1
372,031
73,680,250
Parquet File Encoding - Storing Azure Blob Storage - Failing with Error Error:encoding RLE_DICTIONARY is not supported
<p>I have a pandas dataframe(after pivoted) which i am trying to save the file as parquet and store it in azure blob storage . The file is getting stored in the storage account, however when i try to read the parquet file, i am getting the error - &quot;encoding RLE_DICTIONARY is not supported.&quot; Can some one help ...
<p>After reproducing from my end, I could able to make this work by uploading Iterable data using <code>upload_blob</code>. In my case I'm iterating over the Bytes of the file. Below is the code that worked for me.</p> <pre><code>from azure.storage.blob import BlobServiceClient import pandas as pd storage_account_name...
python|pandas|dataframe|azure-blob-storage|parquet
0
372,032
73,613,734
Use modulo with numbers greater than 64bit integer in numpy/numba
<p>I‘m trying to implement a prime factorization algorithm leveraging the GPU/CUDA for parallelization as a pet project.</p> <p>I‘m using python with numpy and numba for the parallelization part.</p> <p>My problem is now that I hit the 64 bit integer boundary quite fast and I am searching for solutions to work around t...
<p>You can reduce the dividend much more efficiently using modulus thanks to basic congruent identity rules. Indeed:</p> <pre><code>Assuming: a ≡ x [d] b ≡ y [d] c ≡ z [d] Then: a * b + c ≡ x * y + z [d] </code></pre> <p>This is an interesting property if <code>d</code> is small. Additionally, all number can b...
python|numpy|math|cuda|numba
1
372,033
73,621,359
How to obtain source tables from a merged table (Cypher/SQL/Pandas)?
<p>Suppose I have the following table--</p> <pre><code>PersonID | cityID_which_personID_likes | city_longitutde | city_latitude | city_country </code></pre> <p>And I would like to obtain the following two tables--</p> <pre><code>A) PersonID | cityID_which_personID_likes B) cityID | city_longitutde | city_latitude | cit...
<p>you can select a subset of a dataframe like so :</p> <pre><code>df = pd.dataframe({..;}) subset_1= df [[&quot;PersonID&quot; , &quot;cityID_which_personID_likes&quot;]] subset_2= df [[&quot;cityID&quot; , &quot;city_longitutde &quot;, &quot;city_latitude&quot;, &quot;city_country&quot;]] </code></pre> <p>more info ...
sql|pandas|rdbms
0
372,034
73,769,707
Splice two different dataframes based on similar value
<p>I have two dataframes with different dimensions. Lets say it´s displacement measurements but the readings are slightly different values and one has more data. Looks like this:</p> <p>df1</p> <div class="s-table-container"> <table class="s-table"> <thead> <tr> <th>Index</th> <th>displacement</th> </tr> </thead> <tbod...
<p>I used the <strong>merge_asof</strong> function to find the nearest value base on two DataFrames' <strong>displacement</strong> columns, and then filtered the resulting DataFrame by a threshold.</p> <pre><code>df1['displacement'] =df1['displacement'].astype(float) df1 = df1.drop_duplicates('displacement', keep='last...
python|pandas|dataframe|merge
2
372,035
73,805,047
Highlight duplicate rows in Pandas
<p>I'm trying to highlight duplicate rows in several dataframes and export them to excel after. It seems to work when I do it with one dataframe, but as soon as I apply it to another dataframe is highlights the index from the last dataframe for all of them.</p> <p>Here is my code:</p> <pre><code>import pandas as pd imp...
<p>You can use:</p> <pre><code>def highlight_yellow(df, cols): style = 'background-color: yellow' a = np.broadcast_to(np.where(df.duplicated(subset=cols), style, '')[:,None], df.shape) return pd.DataFrame(a, index=df.index, columns=df.columns) df.style.apply(highlight_yellow, cols=['colB'], axis=None) <...
python|pandas|dataframe|numpy
1
372,036
73,535,844
How to split a string column into two column by varying space delimiter on its last occurence
<p>I am trying a way to split a string column in python to two different columns by space delimiter. I have tried with below code:</p> <pre><code>df[['A', 'B']] = df['AB'].str.split(' ', 1, expand=True) </code></pre> <p>But this will work only if the space delimiter is having only single space. I would like to know if ...
<p>You can use this regex to split on:</p> <pre><code>\s+(?!.*\s) </code></pre> <p>This looks for a sequence of spaces which has no spaces after it in the string, so will only split into two values at most.</p> <p>Usage:</p> <pre class="lang-py prettyprint-override"><code>df = pd.DataFrame({'AB': ['aa bb cc', 'dd ee ...
python|python-3.x|pandas|dataframe
1
372,037
73,655,140
Converting a TF model to TFLite and then to EdgeTPU
<p>I am trying to take a simple keras model with an Add operation and convert to TFLite and then to EdgeTPU. Quantization for int8 needs to take place, but depending on the conversion parameters provided it results in either an unsupported operation FlexAddV2, or unsupported data type int32, or an error with AddV2 Erro...
<p>This was resolved here: <a href="https://github.com/google-coral/edgetpu/issues/655" rel="nofollow noreferrer">https://github.com/google-coral/edgetpu/issues/655</a></p> <p>Here is the python conversion code to accomplish this:</p> <pre><code>import tensorflow as tf from tensorflow import keras import numpy as np im...
tensorflow|keras|tflite|edge-tpu
0
372,038
73,787,467
Time since last trade on symbol
<p>I have a pandas DataFrame that contains rows. Each row represents a trade that took place on a particular symbol. The first column has the symbol name, and the second column has the time at which the trade took place. The other details about the trade (e.g., price or quantity) are not important to this question. The...
<p>My input:</p> <pre><code>df = pd.DataFrame( [&quot;AAPL&quot;, &quot;GOOG&quot;, &quot;AAPL&quot;, &quot;MSFT&quot;, &quot;GOOG&quot;], [1, 3, 6, 8, 10], columns=[&quot;trade&quot;] ) </code></pre> <p>My solution:</p> <pre><code>for stock in df.trade.unique(): #compute the distances between actual tim...
python|pandas
0
372,039
73,573,156
Change Numpy array values in-place
<p>Say when we have a randomly generated 2D 3x2 Numpy array <code>a = np.array(3,2)</code> and I want to change the value of the element on the first row &amp; column (i.e. a[0,0]) to 10. If I do</p> <p><code>a[0][0] = 10</code></p> <p>then it works and a[0,0] is changed to 10. But if I do</p> <p><code>a[np.arange(1)][...
<p><code>a[x][y]</code> is <em>wrong</em>. It <em>happens</em> to work in the first case, <code>a[0][0] = 10</code> because <code>a[0]</code> returns a <em>view</em>, hence doing <code>resul[y] = whatever</code> modifies the original array. However, in the second case, <code>a[np.arange(1)][0] = 10</code>, <code>a[np.a...
python|numpy|numpy-ndarray
4
372,040
73,560,570
How to solve Tensorflow 1 becomes unsupported in Google Colab
<p>As Tensorflow 1 becomes unsupported in Google Colab and StyleGAN2-ADA only works with Tensorflow 1. Can anyone help what I should do to solve this issue?</p>
<p>This <em>might</em> work, if you want to run TF1 code under a recent version of TF2</p> <pre class="lang-py prettyprint-override"><code>import tensorflow.compat.v1 as tf tf.disable_v2_behavior() </code></pre> <p>(Instead of <em>import tensorflow as tf</em>, of course)</p> <p>In TF 2.9.1 this is available but depreca...
python|tensorflow|google-colaboratory
0
372,041
73,691,780
unicode decode error while importing Medical Data on pandas
<p>I tried importing a medical data and I ran into this unicode error, here is my code:</p> <pre><code>output_path = r&quot;C:/Users/muham/Desktop/AI projects/cancer doc classification&quot; my_file = glob.glob(os.path.join(output_path, '*.csv')) for files in my_file: data = pd.read_csv(files) print(data) </code></...
<p>Try other encodings, default one is utf-8</p> <p>like</p> <pre><code>import pandas pandas.read_csv(path, encoding=&quot;cp1252&quot;) </code></pre> <p>or ascii, latin1, etc ...</p>
python|pandas|numpy|unicode
0
372,042
73,581,508
In pandas / python, ternary operator to replace NaN in string column with value from different column
<p>We are trying to replace NaN values that are appearing in a string column in our pandas dataframe:</p> <pre><code>d = {'col1': [np.nan, 'Team3'], 'col2': ['Team1', 'Team2']} dd = pd.DataFrame(data=d) dd </code></pre> <p><a href="https://i.stack.imgur.com/B0Lj0m.png" rel="nofollow noreferrer"><img src="https://i.stac...
<p>Try this</p> <pre><code>d = {'col1': [np.nan, 'Team3'], 'col2': ['Team1', 'Team2']} dd = pd.DataFrame(data=d) def check_missing_values(value): if value is np.nan: return True return False dd['col3'] = dd.apply(lambda row: row['col2'] if check_missing_values(row['col1']) else row['col1'], axis=1) <...
python|pandas|dataframe
0
372,043
73,612,885
Numbers in three different formats
<p>I am working with a dataset; below you can see a small example.</p> <pre><code>import pandas as pd import numpy as np data = { 'id':['9.','09', 9], } df = pd.DataFrame(data, columns = [ 'id',]) df['id'] = df['id'].replace(&quot;.&quot;,&quot;&quot;) df </cod...
<p>You could turn them all into doubles and then into integers</p> <pre><code>df['id'].astype('double').astype(int) </code></pre> <p>But this can be a problem for large numbers that exceed the 53 bit significand of the double. If the errant period is always at the end of the string, you could do</p> <pre><code>df['id']...
python|pandas
3
372,044
73,799,702
finding the element in a list closest to the mean of elements in python?
<p>This is my array <code>a= [5, 25, 50, 100, 250, 500] </code>. The mean value of a is 155 (i calculated using <code>sum(a)/len(a)</code>) but i have to store 100 in a variable instead of 155.</p> <p>Is there any easy way to solve this problem.</p>
<p>IIUC, use <a href="https://numpy.org/doc/stable/reference/generated/numpy.argmin.html" rel="nofollow noreferrer"><code>numpy.argmin</code></a> to find the the index of the value closest to the mean by computing the absolute difference to the mean:</p> <pre><code>a = np.array([5, 25, 50, 100, 250, 500]) out = a[np.a...
python|arrays|list|numpy|mean
1
372,045
73,620,291
Averaging values with if else statement of a Pandas DataFrame and creating a new resulting DataFrame
<p>I have a df which looks like this:</p> <pre><code>A B C 5.1 1.1 7.3 5.0 0.3 7.2 4.9 1.7 7.0 10.2 1.1 7.9 10.3 1.0 7.0 15.4 2.0 7.1 15.1 1.0 7.3 0.0 0.9 7.3 0.0 1.3 7.9 0.0 0.5 7.5 -5.1 1.0 7.3 -10.3 0.8 7.3 -10.1 1.0 7.1 </code></pre> <p>...
<p>Groups are defined by difference of values in <code>A</code> is greater like <code>5</code>, pass to <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.core.groupby.GroupBy.agg.html" rel="nofollow noreferrer"><code>GroupBy.agg</code></a> and aggregate <code>mean</code> with <code>std</code>:</...
python|pandas
2
372,046
73,650,263
Applying function after Conditional Group By in Python
<p>This is a simplified version of my problem: given this dataset</p> <pre><code>| link | category | | ---- | -------- | | 1 | 0 | | 1 | 0 | | 1 | 1 | | 2 | 0 | | 3 | 1 | | 3 | 1 | </code></pre> <p>I would like to obtain the following:</p> <pre><code>| link | ...
<p>You can use <a href="https://pandas.pydata.org/docs/reference/api/pandas.crosstab.html" rel="nofollow noreferrer"><code>crosstab</code></a> and <a href="https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.stack.html" rel="nofollow noreferrer"><code>stack</code></a>:</p> <pre><code>(pd.crosstab(df['link'], ...
python|pandas|group-by|conditional-statements
2
372,047
73,527,167
Calculate daily and monthly averages for categories in dataframe
<p>I have the following dataframe</p> <pre><code>date sales cat 29/4/2022 2 a 30/4/2022 5 a 30/4/2022 1 b 1/5/2022 1 a 1/5/2022 8 b 1/5/2022 4 c 1/6/2022 7 a 1/6/2022 9 b 1/6/2022 5 c </code></pre> <p>I...
<p>USE-</p> <pre><code>#Monthly Average df['monthly_avg'] = df.groupby(pd.PeriodIndex(df['date'], freq=&quot;M&quot;))['Value'].mean() #Daily Average df['daily_avg'] = df.groupby(pd.PeriodIndex(df['date'], freq=&quot;D&quot;))['Value'].mean() </code></pre> <p><code>This is untested code,always share reproducible code ...
python|pandas|dataframe|statistics
0
372,048
73,695,764
Add column from another dataframe if two column matches
<p>I am working with huge volume of data and trying to map values from two dataframe. Looking forward for better Time complexity.</p> <p>Here I am trying to match Code from df2 which are in df1 and take MLC Code from df1 if values match.</p> <p>df1</p> <div class="s-table-container"> <table class="s-table"> <thead> <tr...
<p>Try this</p> <pre class="lang-py prettyprint-override"><code>df2.merge(df1[['Code', 'MLC Code']], how='left', on='Code') </code></pre>
python|pandas|dataframe|concatenation|enumerate
0
372,049
73,805,285
Python Pandas Select columns contains different values
<p>I would like to select from DF columns contains only PF_20 and PF_70 values Original Table: <a href="https://i.stack.imgur.com/YFWF6.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/YFWF6.png" alt="enter image description here" /></a></p> <p>Expected Result:</p> <p><a href="https://i.stack.imgur.co...
<p>This might not be the fastest way to do it, but try this:</p> <pre><code>values = [&quot;PF_20&quot;, &quot;PF_70&quot;] columns = [] for col in df.columns: for value in values: if value in df[col].values: columns.append(col) columns = set(columns) #to get rid of duplicates. new_df = df[colum...
python|pandas|dataframe|indexing
0
372,050
73,746,451
How to extract the the key and values from list of dictionary?
<p>How to extract the the key and values from list of dictionary?</p> <p>Below is my data, i want to extract the key and values from list of dictionary.</p> <pre><code>data = [{'index': 0, 'MaterialCode': '67567412', 'DP_Category': 'HAIR CARE'}, {'index': 1, 'MaterialCode': '67567412', 'DP...
<pre class="lang-py prettyprint-override"><code>data = [ {&quot;index&quot;: 0, &quot;MaterialCode&quot;: &quot;67567412&quot;, &quot;DP_Category&quot;: &quot;HAIR CARE&quot;}, {&quot;index&quot;: 1, &quot;MaterialCode&quot;: &quot;67567412&quot;, &quot;DP_Category&quot;: &quot;HAIR CARE&quot;}, ] for idx, ele...
python|pandas
1
372,051
73,807,196
numpy: efficient sum of kronecker products
<ul> <li>I have three sets of matrices {A_i}, {B_i}, and {C_i} with n matrices in each set</li> <li>The A_i are of dimension l x m, the B_i are of dimension m x o and the C_i are of dimension p x q</li> <li>I would like to compute the following: <a href="https://i.stack.imgur.com/jEdHs.png" rel="nofollow noreferrer"><i...
<p>As suggested, I had a look into <a href="https://numpy.org/doc/stable/reference/generated/numpy.einsum.html" rel="nofollow noreferrer">numpy.einsum</a>. This turned out to be quite nice. A solution is:</p> <pre><code>np.einsum('ijk,imn-&gt;jmkn', np.einsum('ijk,ikm-&gt;ijm', A, B), C).reshape(A.shape[1] * C.shape[1]...
numpy|vectorization|kronecker-product
1
372,052
73,573,038
Recording the real time face expression detection
<p>I have coded the face expression detection using <code>Jupyter notebook</code>, detecting seven expressions of the face (Anger, Sad, Disgust, Happy, ...) and tried the real-time detection using the camera of my laptop. Now I want to record those expressions detected by the model in the real-time detection and create...
<p>You could do something like this:</p> <pre><code>from tensorflow import keras import cv2 all_labels = [&quot;Anger&quot;, &quot;Sad&quot;, &quot;Disgust&quot;, &quot;Happy&quot;] # load the trained model, or train a model model = keras.models.load_model('path/to/location') # Open the camera cap = cv2.VideoCaptur...
python|tensorflow|keras|jupyter-notebook|tf.keras
1
372,053
73,822,435
Pandas: How to convert list to data frame
<p>This is code that I have</p> <pre><code>import pandas as pd data = [[1,&quot;credit&quot;],[1,&quot;cash&quot;],[1,&quot;credit&quot;],[2,&quot;credit&quot;],[2,&quot;credit&quot;],[2,&quot;credit&quot;],[3,&quot;credit&quot;],[3,&quot;credit&quot;],[3,&quot;credit&quot;]] df = pd.DataFrame(data, columns=['account_...
<p>Before using groupby, you can do a filtering process to filter out rows of not credit.</p> <pre><code>result = df[df['type'] == 'credit'].groupby('account_id').value_counts() account_id type 1 credit 2 2 credit 3 3 credit 3 dtype: int64 </code></pre> <p>You can use <code>re...
python|pandas|dataframe
0
372,054
73,585,293
Switch between the heads of a model during inference
<p>I have 200 neural networks which I trained using transfer learning on text. They all share the same weights except for their heads which are trained on different tasks. Is it possible to merge those networks into a single model to use with Tensorflow such that when I call it with input (text, i) it returns me the pr...
<p>You probably have a model like the following:</p> <pre><code># Create the model inputs = Input(shape=(height, width, channels), name='data') x = layers.Conv2D(...)(inputs) # ... x = layers.GlobalAveragePooling2D(name='penultimate_layer')(x) x = layers.Dense(num_class, name='task0', ...)(x) model = models.Model(input...
tensorflow|deep-learning
1
372,055
73,742,257
Python (numpy?) - build transformation matrix from sets of source and destination points
<p>Let's imagine I have two sets points in a 2d Euclidean system:</p> <pre><code>src = [ [722.6, 1571.4], [832, 1466], [419, 1482], [1005, 2804], &lt;snip&gt; ] dst = [ [35839.65, 49808.55], [42771.08, 41488.07], [15764.26, 44065.95], [72760.36, 76645.15], &lt;snip&gt; ] </code><...
<p>You are probably looking for a <em>rigid registration</em>, which aligns the point sets allowing rotation and translation, or you may be looking for an <em>affine registration</em>, which also allows changes in scale, shear or reflection.</p> <p>You can use the function <a href="https://matthew-brett.github.io/trans...
python|numpy|matrix|2d|transformation
1
372,056
73,622,517
pandas group by custom column
<p>For example, I have a following df:</p> <div class="s-table-container"> <table class="s-table"> <thead> <tr> <th>score</th> <th>var1</th> </tr> </thead> <tbody> <tr> <td>0.465</td> <td>jack, jones, phil</td> </tr> <tr> <td>0.712</td> <td>don, sam, bob</td> </tr> <tr> <td>0.112</td> <td>jones, alex, sam</td> </tr> </...
<p>You can try</p> <pre class="lang-py prettyprint-override"><code>out_ = (df.assign(score=df['score'].round(1), var1=df['var1'].str.split(', ')) .explode('var1')) out = pd.crosstab(out_['var1'], out_['score']) </code></pre> <pre><code>print(out_) score var1 0 0.5 jack 0 0.5 jones...
python|pandas
1
372,057
71,242,456
Saving accuracy and loss with callback on colab
<p>So im trying to train a model on colab, and it is going to take me roughly 70-72 hr of continues running. I have a free account, so i get kicked due to over-use or inactivity pretty frequently, which means I cant just dump history in a pickle file.</p> <pre><code>history = model.fit_generator(custom_generator(train_...
<p>Think you want to write your callback as follows</p> <pre><code>class STOP(tf.keras.callbacks.Callback): def __init__ (self, model, csv_path, model_save_dir, epochs, acc_thld): # initialization of the callback # model is your compiled model # csv_path is path where csv file will be stored ...
python|tensorflow|google-colaboratory|tf.keras|custom-training
0
372,058
71,211,208
How can I obtain a cell value from the last line of a list?
<p>I'm struggling to figure out how, using Pandas, to obtain the last value in the second column of a list which is updating regularly (i.e. an every increasing list over time) and assign that to a value.</p> <pre><code> 1 2 15 21/02/2022 18:07:40 38055.3966 16 21/02/2022 18:07:49 ...
<p>One approach is to use <code>iloc</code>. For example, <code>df.iloc[-2,1]</code> gives the second to last entry of the second column.</p>
python|pandas|dataframe
1
372,059
71,100,624
Check if columns have a nan value if certain column has a specific value in Dataframe
<p>I'm trying to add column in Dataframe which has a result of checking if other columns have value in it.</p> <p>This is a test df I made:</p> <pre><code>df = pd.DataFrame({&quot;condition&quot;:[1,np.nan,np.nan,np.nan,1],&quot;a&quot;:[np.nan,4,5,6,np.nan],&quot;b&quot;:[np.nan,2,&quot;e&quot;,2,np.nan],&quot;c&quot;...
<p>so you have an if-elif-else situation. Then we can use <code>np.select</code> for it. It needs the conditions and what to do when they are satisfied:</p> <ul> <li>your if is:    &quot;condition is 1 and a,b,c has all nan&quot;</li> <li>your elif is: &quot;condition is nan&quot;</li> <li>what remains is else, as us...
python-3.x|pandas|dataframe
1
372,060
71,178,104
In pytorch, torch.unique is returning repititions
<p>I have this 2-D tensor:</p> <pre><code>tmp = torch.tensor([[ 0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6, 6, 6, 7, 7, 7, 8, 8, 8, 9, 9, 9, 10, 10, 10, 11, 11, 11, 12, 12, 12, 13, 13, 13, 14, 14, 14, 15, 15, 15, 15, 16, 16, 16, 17, 17, 17, 18, ...
<p>It is because you specified <code>dim=1</code>. PyTorch is thus checking for unique <em>pairs</em> (which it correctly does). Like (0, 0), (1, 1), (16, 0): these are the unique pairs that it generated. In general the pair <code>(temp[0,i], temp[1,i])</code> is unique for all <code>i</code>.</p> <p>If you want all th...
pytorch
0
372,061
71,321,630
How to convert each row of a dataframe to new column use concat in python
<p>If I have dataframes,</p> <pre><code>df1 = pd.DataFrame( { &quot;A&quot;: [&quot;A0&quot;, &quot;A1&quot;, &quot;A2&quot;, &quot;A3&quot;], &quot;B&quot;: [&quot;B0&quot;, &quot;B1&quot;, &quot;B2&quot;, &quot;B3&quot;], &quot;C&quot;: [&quot;C0&quot;, &quot;C1&quot;, &quot;C2&quot;, &quot;C3&quot;], ...
<p>IIUC, you could <code>stack</code> the individual dataframes, <code>concat</code> and reshape:</p> <pre><code>dfnew = pd.concat([df1.stack(), df2.stack()]).droplevel(0).to_frame().T </code></pre> <p>output:</p> <pre><code> A B C D A B C D A B C D A B C D A B C D A B C ...
python|pandas|concatenation
3
372,062
71,319,221
Merging data from a separate .csv file using Pandas
<p>I want to create two new columns in job_transitions_sample.csv and add the wage data from wage_data_sample.csv for both Title 1 and Title 2:</p> <p>job_transitions_sample.csv:</p> <pre><code> Title 1 Title 2 Count 0 administrative assistant office manager 20 ...
<p>Use <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.map.html" rel="nofollow noreferrer"><code>Series.map</code></a> by dictionary <code>d</code> - cannot use <code>dict</code> for varialbe name, because python code name:</p> <pre><code>df = pd.read_csv('job_transitions_sample.csv') w...
python|python-3.x|pandas|csv
1
372,063
71,211,177
How to export an Excel file from a dictionary?
<p>I want to make an excel file from a dictionary that I have. It simply is an dictionary with information about images, like the size, how many paragraphs, how many words, etc.</p> <p>Let's say the dictionary is:</p> <pre><code>{'Screenshot_1.jpg': {'SIZE': 214649, 'HEIGHT': 664, ...
<p><code>pandas</code> has <code>to_csv</code> function to export dataframe as a .csv file, as follows:</p> <pre class="lang-py prettyprint-override"><code>import pandas as pd dict1 = {'Screenshot_1.jpg': {'SIZE': 214649, 'HEIGHT': 664, 'WIDTH': 1351, 'PARAGRAPHS': 3, 'WORDS': 427, 'Parag...
python|excel|pandas|dataframe|dictionary
1
372,064
71,346,892
how to match a string from a list of strings and ignoring regex special characters?
<p>I have this string:</p> <pre><code>d = {'col1': ['Digital Forms - how to spousal information on DF 2,0']} </code></pre> <p>I turned it into a dataframe :</p> <pre><code>df = pd.DataFrame(d) </code></pre> <p>From this dataframe, I want to match this list of words:</p> <pre><code>wordlist = ['Digital Forms', 'how', 's...
<p>You may form a regex alternation from your word list using <code>re.escape</code> to escape the metacharacters:</p> <pre class="lang-py prettyprint-override"><code>wordlist = ['Digital Forms', 'how', 'spousal', 'DF 2.0'] regex = r'\b(' + '|'.join([re.escape(x) for x in wordlist]) + r')\b' words = df['col1'].str.find...
python|pandas|dataframe
3
372,065
71,119,338
How to find the intersection between two columns from two different dataframes
<p>I'm trying to compare two different columns from two different DataFrames.</p> <pre><code>test_website1 Domain 0 www.google.com 1 www.facebook.com 2 www.yahoo.com test_website2 Domain 0 www.bing.com 1 www.instagram.com 2 www.google.com </code></pre> <hr /> <pre><cod...
<p>The standard way to do this in pandas is an inner <a href="https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.merge.html" rel="nofollow noreferrer"><code>merge</code></a> (default is <code>how='inner'</code>):</p> <pre><code>pd.merge(df1['Domain'], df2['Domain']) # Domain # 0 www.google.com <...
python|pandas|dataframe
1
372,066
71,388,201
Checking a Column of Lists Against Another Column of Lists and Returning Multiple Column Values
<p>I am comparing the list values of <code>sb_list</code> against <code>psr_list</code>. If all list items from <code>sb_list['ASINs']</code> are found in any of the lists from <code>psr_list['Child ASIN']</code>, <code>sb_list['bucket']</code> is marked <code>'clean'</code>. This part of the code is running fine...</p...
<p>You could use <code>set.issubset</code> in a list comprehension to check if any list in <code>sb_list</code> is contained in any list in <code>psr_list</code>. If a list exists, then get &quot;Group&quot; value where it exists, if not fill in with <code>&quot;&quot;</code>. Note that this assumes only one list in <c...
python|pandas|dataframe|numpy
1
372,067
71,399,557
Unique combination of two columns in Pandas
<p>I would like to keep the unique combination of two columns. For example, A and B is the same as B and A.</p> <pre><code>df = pd.DataFrame({'col1': [1,2,4,3], 'col2': [2,1,3,4]}) col1 col2 0 1 2 1 2 1 2 4 3 3 3 4 </code></pre> <p>Desired outcome:</p> <pre><code> c...
<p>You can do <code>np.sort</code> then <code>drop_duplicates</code></p> <pre><code>df[:] = np.sort(df.values,1) out = df.drop_duplicates() Out[625]: col1 col2 0 1 2 2 3 4 </code></pre>
python|pandas
0
372,068
71,211,228
Problem with pytorch hooks? Activation maps allways positiv
<p>I was looking at the activation maps of vgg19 in pytorch. I found that all the values of the maps are positive even before I applied the ReLU.</p> <p>This seems very strange to me... If this would be correct (could be that I not used the register_forward_hook method correctly?) why would one then apply ReLu at all?<...
<p>You should <a href="https://pytorch.org/docs/stable/generated/torch.clone.html" rel="nofollow noreferrer"><code>clone</code></a> the output in</p> <pre><code>def get_activation(name): def hook(model, input, output): activation[name] = output.detach().clone() # return hook </code></pre> <hr /> <p>Note...
python-3.x|pytorch
1
372,069
71,416,560
How to get y_train in model
<p>For some reason I need to use y_train (the target) in my model (not only in loss function), but I didn't find a way to get it.</p> <p>I get my training dataset like this:</p> <pre><code> train_ds = DataGenerator(&quot;train&quot;, args).fetch() &lt;PrefetchDataset shapes: ((2, None), (2, 4000, 22)), types: (tf.f...
<p>Hi guys I just find some ways to solve this.</p> <ol> <li><p>Use tf.concat() to concatenate the data in y_train to x_train and send them together into the model, them separate them. But the dimension of both thing should be the same. (My input and target don't have the same size so I didn't try this.)</p> </li> <li>...
python|tensorflow|keras
0
372,070
71,169,941
Error with installation of pycocotools for Detr Tensorflow
<p>I'm trying to use Detr Tensorflow models and need to install pycocotools. On a Windows 10 PC, I'm executing this in a Visual Studio Code. I'm following the steps provided in this GitHub repo : <a href="https://github.com/Visual-Behavior/detr-tensorflow#install" rel="nofollow noreferrer">https://github.com/Visual-Be...
<p>The solution is in the error output.</p> <p>You need to install Microsoft C++ Build Tools.</p>
tensorflow|pycocotools
0
372,071
71,336,067
How to freeze some layers of BERT in fine tuning in tf2.keras
<p>I am trying to fine-tune 'bert-based-uncased' on a dataset for a text classification task. Here is the way I am downloading the model:</p> <pre><code>import tensorflow as tf from transformers import TFAutoModelForSequenceClassification, AutoTokenizer model = TFAutoModelForSequenceClassification.from_pretrained(&quo...
<p>I found the answer and I share it here. Hope it can help others. By the help of <a href="https://raphaelb.org/posts/freezing-bert/" rel="nofollow noreferrer">this article</a>, which is about fine tuning bert using pytorch, the equivalent in tensorflow2.keras is as below:</p> <pre><code>model.bert.encoder.layer[i].tr...
python-3.x|keras|tensorflow2.0|huggingface-transformers|bert-language-model
0
372,072
71,100,132
How to make a pivot table from a dataframe with multiple columns?
<p>Please help me.</p> <p>My dataframe looks like this:</p> <div class="s-table-container"> <table class="s-table"> <thead> <tr> <th><strong>date</strong></th> <th><strong>account</strong></th> <th><strong>action</strong></th> <th></th> <th></th> </tr> </thead> <tbody> <tr> <td>2021-01-11</td> <td>504</td> <td>login</t...
<p>Could this work?</p> <p>Get the piece of dataframe during a certain period of time:</p> <pre><code>new_df = df[start_date &lt; df['date'] &lt; end_date] </code></pre> <p>new_df now has all the rows during a certain period of time. Get all the unique account values:</p> <pre><code>accounts = new_df['account'].unique(...
python|pandas|dataframe|group-by|pivot-table
0
372,073
71,114,265
Why are my train/valid set loss curves dropping and plateau after X epochs?
<p>I am training a deep model for MRI segmentation. The models I am using are U-Net++ and UNet3+. However, when plotting the validation and training losses of these models over time, I find that they all end with a sudden drop in loss, and a permanent plateau. Any ideas for what could be causing this plateau? or any id...
<p>Due to the high number of parameters it is hard if not impossible to reason about the optimization landscape, so any speculations are really just that, speculations.</p> <p>If you assume that the model got stuck somewhere, that is, that the gradient is getting very small (it's sometimes worth plotting the distributi...
machine-learning|deep-learning|pytorch|statistics|evaluation
0
372,074
71,289,730
Pandas: How do I delete first two rows of headers?
<p>I'm using an excel file and would like to drop first two rows of headers that has 3 rows of headers.</p> <p>Current File Example:</p> <pre><code> Type1, Type2, Type3, Type4 SubType1, SubType2, SubType3, SubType4 SubSubType1-3,,,SubSubType4 0 Blah, Blah1, Blah2, Blah4 1 2 </code></pre> <p>After dropping ...
<p>Use header parameter with a value = 2.</p> <pre><code>data = pd.read_csv(&quot;file_name.csv&quot;, header=2) </code></pre>
python|pandas|data-science
2
372,075
71,228,810
Transposing a list of dicts
<p>Having some trouble trying to come up with the most pythonic way to rearrange this list of dictionaries...</p> <pre><code>[{1: ['s1e1.csv', 's1e2.csv']}, {2: ['s2e1.csv', 's2e2.csv']}] </code></pre> <p>So that it is arranged like this...</p> <pre><code>[{1: 's1e1.csv', 2: 's2e1.csv'}, {1: 's1e2.csv', 2: 's2e2.csv...
<p>Something like this should work:</p> <pre class="lang-py prettyprint-override"><code>vals = [{1: ['s1e1.csv', 's1e2.csv']}, {2: ['s2e1.csv', 's2e2.csv']}] newvals = [{1: vals [0][1][x], 2: vals [1][2][x]} for x in range(2)] # Output: [{1: 's1e1.csv', 2: 's2e1.csv'}, {1: 's1e2.csv', 2: 's2e2.csv'}] </code></pre>
python|pandas|numpy|csv|dictionary
0
372,076
71,386,143
Trying to save image from numpy array with PIL, getting errors
<p>trying to save an inverted image, saved inverted RGB colour data in array pixelArray, then converted this to a numpy array. Not sure what is wrong but any help is appreciated.</p> <pre><code>from PIL import Image import numpy as np img = Image.open('image.jpg') pixels = img.load() width, height = img.size pixelArr...
<p>Your np.array creates an array shape (4000000, 3) instead of (2000, 2000, 3).</p> <p>Also, you may find that directly mapping the subtraction to the NumPy array is faster and easier</p> <pre><code>from PIL import Image import numpy as np img = Image.open('image.jpg') pixelArray = np.array(img) pixelArray = 255 - p...
python|numpy|python-imaging-library
1
372,077
71,412,932
Pandas DataFrame - row comparision and isolation problem
<p>I have those DataFrame where I have fathers that are their own grandchild. I want to isolate the corresponding rows to treat them separately.</p> <pre><code>df = pd.DataFrame({ 'father' : ['a', 'b', 'e', 'f', 'j', 'k'], 'son' : ['b', 'a', 'f', 'g', 'k', 'j'] }) df df2 = pd.DataFrame({ 'father' : [1, 2, 4...
<p>You can use <code>frozenset</code> to group your rows:</p> <pre><code>df['group'] = df.apply(frozenset, axis=1) print(df) # Output father son group 0 a b (a, b) 1 b a (a, b) 2 e f (e, f) 3 f g (g, f) 4 j k (j, k) 5 k j (j, k) </code></pre> <p>After, you can use a ...
python|pandas|dataframe
1
372,078
71,274,535
How do I change my legend labels to other words?
<p><a href="https://i.stack.imgur.com/3GnQI.jpg" rel="nofollow noreferrer">enter image description here</a></p> <p>I have tried several different ways, but I have had no luck so far. I am trying to change my bar chart legend labels from (0,1,2,3,4,5,6) to (Monday, Tuesday, Wednesday Thursday, Friday, Saturday). Help pl...
<p>Try:</p> <pre><code>legend_labels = ['Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday'] plt.legend(labels = legend_labels, title = 'Weekdays', loc = 'upper left') plt.show() </code></pre>
python|pandas|matplotlib|legend
0
372,079
71,125,097
Pandas fillna() does not apply values from one column to an entire dataframe
<p>Please explain, why this won't work (hc is pandas dataframe on below example):</p> <pre><code>import pandas as pd import numpy as np hc = pd.DataFrame([['Adolf', np.nan], ['Hans', 'Johan']], columns=('First Name', 'Second Name')) hc.fillna(value=hc[&quot;First Name&quot;]) </cod...
<p>IIUC, you want to <a href="https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.fillna.html" rel="nofollow noreferrer"><code>fillna</code></a> with a Series as reference, per row.</p> <pre><code>hc = hc.fillna(hc['First Name']) </code></pre> <p>This is not currently supported on the columns for all pandas v...
python|pandas
0
372,080
71,298,402
Is there a better way to search a sorted list if the other list is sorted too?
<p>In the numpy library, one can pass a list into the <code>numpy.searchsorted</code> function, whereby it searched through a different list one element at a time and returns an array of the same sizes as the indices needed to preserve order. However, it seems to be wasting performance if both lists are sorted. For exa...
<p>You could use <a href="https://pypi.org/project/sortednp/" rel="nofollow noreferrer">sortednp</a>, unfortunately it does not give too much flexibility, In the code snippet below I used its <a href="https://gitlab.sauerburger.com/frank/sortednp#index-tracking-and-duplicates" rel="nofollow noreferrer">merge</a> tracki...
python|numpy
2
372,081
71,330,409
How can I calculate the F1-score and other classification metrics from a faster-RCNN? (object detection in PyTorch)
<p>I'm trying to wrap my head around this but struggling to understand how I can compute the f1-score in an object detection task.</p> <p>Ideally, I would like to know false positives, true positives, false negatives and true negatives for every target in the image (it's a binary problem with an object in the image as ...
<p>The use of the terms precision, recall, and F1 score in object detection are slightly confusing because these metrics were originally used for binary evaluation tasks (e.g. classifiation). In any case, in object detection they have slightly different meanings:</p> <p>let: TP - set of predicted objects that are succe...
python|deep-learning|pytorch|computer-vision|object-detection
1
372,082
71,385,637
What Loss function to use for binary classification in CNN using float labels?
<p>So I am building a CNN that gets images using labels that go from 0 to 1.</p> <p>What I mean is that I am trying to perform detection of <strong>one</strong> thing in the image and each image has a label between 0 and 1 that stands for the probability of said type of event being in that image.</p> <p>I want to outpu...
<p>This solution is for <code>logits</code> (output of last linear layer) not for output probabilities</p> <pre class="lang-py prettyprint-override"><code>def loss(logits, soft_labels): anti_soft_labels = 1 - soft_labels return soft_labels * tf.nn.softplus(-logits) + anti_soft_labels * tf.nn.softplus(logits) +...
tensorflow|deep-learning|conv-neural-network|classification|loss-function
1
372,083
71,160,692
Parallelization of for loop: pandas
<p>I have 1000 Tables and what I need to do:</p> <ol> <li>Extract data from table x</li> <li>Do some Calculation based on the data</li> <li>Save results in one big table</li> </ol> <p>Here is my code</p> <pre><code>table_names = pd.read_sql(&quot;SELECT table_name FROM information_schema.tables where table_schema = 'so...
<p>If you are looking to simply get one value (count) from each sql query then it likely makes more sense to let SQL DB do the hard work and just bring the result back into pandas.</p> <pre><code>table_names = pd.read_sql(&quot;SELECT table_name FROM information_schema.tables where table_schema = 'some_schema';&quot;,s...
python|pandas
0
372,084
71,127,140
Is there a faster or better way to segregate dataset into 80 20 ratio in python?
<pre><code>X.shape #output is =&gt; (2555904, 1024, 2) X[0] #Output is =&gt; array([[ 0.0420274 , 0.23476323], [-0.2728826 , 0.40513492], [-0.26707262, 0.22749889], ..., [-0.7055947 , -0.28693035], [-0.41157472, 0.66826206], [ 0.06487698, 0.6358149 ]], dtype=float32) total = len(X) n_train = int(0.8*total) #80% sample...
<p>You can use <strong>sklearn.model_selection.train_test_split</strong>. Check out the <a href="https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html" rel="nofollow noreferrer">official documentation</a> with an example. You have to split the data in target variable and explan...
python|pandas
0
372,085
71,423,135
Python Plotting Grouped Data
<p>The grouped data looks like</p> <p><a href="https://i.stack.imgur.com/4s4NJ.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/4s4NJ.png" alt="The Grouped Data looks like" /></a></p> <p>My approach yields to</p> <p><a href="https://i.stack.imgur.com/q7A05.png" rel="nofollow noreferrer"><img src="http...
<p>As the test DataFrame I used:</p> <pre><code> MAPPING CREATED_DTM counts 0 Beschaedigung 2020-04-30 22738 1 Beschaedigung 2020-05-31 21523 2 Beschaedigung 2020-06-30 18516 3 Beschaedigung 2020-07-31 21436 4 Beschaedigung 2020-08-31 22325 5 Verlust 2020-04-30 20000 6 Verl...
python|pandas|plot|pandas-groupby
1
372,086
71,364,802
How to overide NumPy ndarray attributes / properties?
<p>I have a class that has <code>real</code> and <code>imag</code> attributes,</p> <pre><code>class A(): def __init__(self, real, imag): self.real = real self.imag = imag def __repr__(self): return repr((self.real, self.imag)) def conjugate(self): return A(s...
<p>Your class, with repr tweaked:</p> <pre><code>In [181]: class A: ...: def __init__(self, real, imag): ...: self.real = real ...: self.imag = imag ...: ...: def __repr__(self): ...: return f&quot;A: {repr((self.real, self.imag))}&quot; ...: ......
python|arrays|numpy
0
372,087
71,277,555
Avoid for-loops when getting mean of every positive-value-interval in an array
<p>I want to get the mean of every interval with values above a threshold. Obviously, I could do a loop and just look if the next value is under the threshold etc., but I was hoping that there would be an easier way. Do you have ideas that are similar to something like masking, but include the &quot;interval&quot;-prob...
<p>The trick I am using here is to calculate where there are sudden differences in the mask which means we switch from a contiguous section to another. Then we get the indexes of where those sections start and end, and calculate the mean inside of them.</p> <pre><code># Imports. import matplotlib.pyplot as plt import n...
python|numpy|masking|threshold
2
372,088
71,139,246
groupby transform valueerror length of passed values
<p>Hello can you help me understand what is the issue here and how to solve it?</p> <pre><code>dft = pd.DataFrame({'B': [0, 1, 2, 5, 4, 7, 2, 2, 2, 5, 6, 7]}) dft['user'] = ['a','b','c','b','a','c','a','b','b','c','a', 'c'] dft.groupby('user')['B'].transform(lambda row: row.ewm(span=2)).mean() </code></pre> <p>gives <...
<p>I suppose that invocation of <code>mean()</code> should be a part of your lambda function.</p> <p>So maybe your code should be:</p> <pre><code>dft.groupby('user')['B'].transform(lambda row: row.ewm(span=2).mean()) </code></pre> <p>For your sample data I got:</p> <pre><code>0 0.000000 1 1.000000 2 2.00000...
python|pandas
1
372,089
71,109,317
Not being able to change the data type of a column in a dataframe
<p>I want to change the data type of the values in the column &quot;id&quot; from integer to string and then save the new dataframe to a CSV file. This is what I have tried:</p> <pre><code>import pandas as pd df = pd.read_csv ('DataSet.csv', header=[0], on_bad_lines='skip', sep = ';') df[&quot;id&quot;] = df[&quot;id&...
<p>In <code>csv</code> file are all values saved like strings, pandas by default converting types like <code>int, float</code> in <a href="http://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_csv.html" rel="nofollow noreferrer"><code>read_csv</code></a> metdod.</p> <p>So is necessary always converting ...
python|pandas|dataframe|csv
2
372,090
71,185,591
How to replace values in a matrix based on specific indices?
<p>I'm trying to index a matrix based on an array of coordinates, so that I can replace the values in those positions with 1, and leave the rest as is. Here's what I mean:</p> <p>I have the following 3x3 matrix</p> <pre><code>matrix = np.zeros((3,3), dtype=int) matrix &gt;&gt;&gt; array([[0, 0, 0], [0, 0, 0...
<p>You can run e.g.:</p> <pre><code>matrix[tuple(coords.T)] = 1 </code></pre>
python|arrays|numpy|matrix|indexing
1
372,091
71,267,401
Key-Error: KeyError: "None of [Float64Index([15.593, 15.577, 15.563], dtype='float64')] are in the [columns]"
<p>I'm trying to calculate a column, but I get the following error:</p> <pre><code>Key-Error: KeyError: &quot;None of [Float64Index([15.593, 15.577, 15.563], dtype='float64')] are in the [columns]&quot; </code></pre> <p>Where is my error coming from? and how can I fix it?</p> <pre><code>import pandas as pd import numpy...
<p>Try refactoring your function like this:</p> <pre class="lang-py prettyprint-override"><code>def absolute_humidity(temp, humidity): return ( ( (610.78 * np.exp((17.08085 * temp) / (234.175 + temp)) * humidity / 100) / (462 * (273.1 + temp)) * 1000 ) if ...
python-3.x|pandas|dataframe
0
372,092
71,300,849
How to check if a string value of one row is contained in the string value of another row in the same column in pandas dataframe
<p>I have a dataframe as follows :</p> <p>The &quot;docid&quot; is the exploded column of &quot;DocID&quot;. I want to check if a string in the &quot;Term&quot; column is contained in another row in the same column. For example, rows 3 and 4 have &quot;in the treatment&quot; and &quot;in the treatment of&quot;.</p> <p>...
<p>What you can do is substract the docfreq of &quot;in the treatment&quot; by &quot;in the treatment of&quot; which will return the number of only &quot;in the treatment&quot;, then for docID, remove any inctances of docID &quot;in the treatment of&quot; from &quot;in the treatment&quot;</p>
python|pandas|string|dataframe|pandas-groupby
0
372,093
71,232,030
how does array.shape[:] works?
<p>I think I am just confused with how <strong>array.shape[]</strong> works.<br /> The output of <strong>array.shape[:]</strong> is the shape of the array in tuple which is fine.<strong>output:(4,5)</strong><br /> The output of <strong>array.shape[1:]</strong> is <strong>(5,)</strong> and the output of <strong>array.sh...
<p>You got a problem with tuples here, not with <code>array.shape</code> in itself. As pointed out in the comments, <code>x[:1]</code> gives you the first element of a tuple as a singleton tuple, because you are indexing &quot;all&quot; elements of the tuple up to the first (included).</p> <p>Instead, <code>x[1:]</code...
python|numpy
1
372,094
71,331,781
RegEx negation to handle decimal values in Pandas dataframe using .replace()
<p>I have the following Pandas dataframe:</p> <pre><code>foo = { 'Sales' : [200, 'bar', 400, 500], 'Expenses' : [70, 90, 'baz', 170], 'Other' : [2.5, 'spam', 70, 101.25] } df = pd.DataFrame(foo) Sales Expenses Other 200 70 2.5 bar 90 spam 400 baz 70 500 170 ...
<p>Regex will only work on strings. You can cast all values to strings using .astype(str)</p> <pre><code>df['Other'].astype(str).replace('[^0-9]', np.NaN, regex=True) </code></pre>
python|regex|pandas|regex-negation
1
372,095
71,117,376
Python - Sum values for all dates prior to a specific date
<p>I currently have two dataFrames that look like this:</p> <pre><code>Df3 - which is the output dataFrame: | CompanyNm | CpID | Date | :--------|:-----------------------------------|:-------------|:----------| 0 | {Converting) | {C} ...
<p>EDIT: I misunderstood your problem definition earlier. Now corrected it:</p> <pre><code>def func(g): mask = (df4['CustID'] == g.name[0]) &amp; (df4['InvoiceDt'] &lt;= g.name[1]) return df4[mask]['SalesAmt'].sum() df3.groupby(['CpID','Date']).apply(func) </code></pre>
python|pandas
0
372,096
71,165,950
Stacked Bar Chart Y axis Values Missing
<p>I have a wide data frame and am using the following code to produce a stacked bar graph; however, the values on the Y axis are missing and I'm not sure why. I assume it has something to do with stacking the columns. You can see I tried to manually set the Y axis within my code but the graph did not capture that comm...
<p>This code provided the desired results:</p> <pre><code>ALL_BOARD_C=ALL_BOARD.drop(['Acute_Hours','ICU_Hours','PSYCH_Hours','index'],axis=1) ALL_BOARD_C.plot(ax=g9,x='Day' ,kind='bar', stacked=True, title='ED Total Boarders by Unit') loc = WeekdayLocator(byweekday=MO, interval=1) g9.xaxis.set_major_locator(loc) g9.se...
python|pandas|matplotlib
0
372,097
71,296,457
How to use np.where on 3D array with all()
<p>I've been stuck for quite some time on this problem.</p> <p>I have an array of shape (4,3,3). In my real case, the shape is much bigger.</p> <pre><code>a = np.array([ [[1,2,3], [3,4,2], [1,3,4]], [[1,2,3], [3,6,2], [1,4,4]], [[1,2,3], [3,6,2], [1,4,4]], [[1,2,3], [3,6,2], [1,2,4]] ]) </code></pre> <p...
<p>The <code>cond</code> array is central to the workings of <code>where</code>:</p> <pre><code>In [167]: a == [1, 2, 3] Out[167]: array([[[ True, True, True], [False, False, False], [ True, False, False]], [[ True, True, True], [False, False, False], [ True, False, False]],...
python|numpy
0
372,098
52,053,522
Summing up values from one column based on values in other column
<p>I have a dataframe something like below,</p> <pre><code>Timestamp count 20180702-06:26:20 50 20180702-06:27:11 10 20180702-07:05:10 20 20180702-07:10:10 30 20180702-08:27:11 40 </code></pre> <p>I want output something like below,</p> ...
<p>Use</p> <pre><code>In [252]: df.groupby(df.Timestamp.dt.strftime('%Y-%m-%d-%H'))['count'].sum() Out[252]: Timestamp 2018-07-02-06 60 2018-07-02-07 50 2018-07-02-08 40 Name: count, dtype: int64 In [254]: (df.groupby(df.Timestamp.dt.strftime('%Y-%m-%d-%H'))['count'].sum() .reset_index(name='Sum...
python|python-2.7|pandas
0
372,099
52,250,710
How do I covert an Excel Date in MMM-YYYY to datetime or strings?
<p>My dataframe is taken from a Excel file which formats their dates as e.g Jan 2018. </p> <p>I want to change to datetime such as 01-2018 or even as a string like 01/2018.</p> <p>I have two problems:</p> <ol> <li><p>When attempting to convert to datetime I have an out of bound error (nanosecond)</p> <pre><code>two...
<p>I have managed to successfully solve my own question. Thank you for your interest. If there any better solutions I am all ears</p> <pre><code>twoyear_df['Date'] = pd.to_datetime(twoyear_df['Date'], format='%b %y') </code></pre>
python|pandas|datetime|dataframe
2