Datasets:

---

suriyagunasekar

/

stackoverflow-python-with-meta-data

like 10

[ "You have two easy options:\nUp the DPI\nplt.savefig(\"graph.png\", dpi=1000)\n\n(larger image file size)\nSave as a PDF\nplt.savefig(\"graph.pdf\")\n\nThis is the best option, as the final graph is not rasterized. In theory, you should be able to zoom in indefinitely. \n", "While not in GTK, you might want to check out NetworkX Viewer.\n", "You may want to check this out:\nplt.savefig(\"name.svg\")\n\nthe quality is magnificent. although the dpi option is still alive.\n", "Use this plt.savefig('name.svg') and then just upload the file to Figma -https://www.figma.com\n" ]

[ 33, 2, 1, 0 ]

[]

[]

[ "graph", "matplotlib", "networkx", "python" ]

[ "It's not an error, it's only a warning. Pycharm is telling you that you have imported a library but you're not using this library (request). If you use the library the warning disappears\n" ]

[ 0 ]

[]

[]

[ "import", "pip", "python", "python_idle", "shared_libraries" ]

[ "Solved:\nan upper case letter in the bucket name causes the internal error...\nreplace\nstorage.create_bucket(\"Hello\")\n\nwith\nstorage.create_bucket(\"hello\")\n\n:)\nEDIT:\nthat's why:\nhttps://forum.storj.io/t/bucket-name-uppercase/20554/2?u=mike1\n" ]

[ 0 ]

[]

[]

[ "amazon_s3", "go", "python" ]

[ "I don't think that expr can do what you are trying to do. However, you don't have to concatenate all your expressions and use a single expr, instead you can do something like this\ndf_new = (\n df\n .select(\n *(col_lst + [expr(nc) for nc in new_col_list])\n ) \n\n\nThe above code is untested, but in general a technique of creating a list of columns is common in Spark.\n" ]

[ 0 ]

[]

[]

[ "apache_spark", "apache_spark_sql", "arithmetic_expressions", "python" ]

[ "You could do this using eval\nimg = eval('score' + str(correct))\n\nbut this is dangerous if correct is provided by the user. A better approach is to use a list\nimages = [ImageTk.PhotoImage(Image.open(\"scores/09.png\")),\n ImageTk.PhotoImage(Image.open(\"scores/19.png\")),\n ImageTk.PhotoImage(Image.open(\"scores/29.png\")),\n ImageTk.PhotoImage(Image.open(\"scores/39.png\")),\n ImageTk.PhotoImage(Image.open(\"scores/49.png\")),\n ImageTk.PhotoImage(Image.open(\"scores/59.png\"))]\n\nimg = images[correct]\n\n", "Firstly, you'll probably want to import Pathlib to work with the absolute paths to your image files\nfrom pathlib import Path\n\nThen I think it might make more sense to put your image filenames into a list...\nimage_dir = Path(r'C:\\<path>\\<to>\\scores') # the folder containing the images\nimages = [ # list of individual image file names\n \"09.png\",\n \"19.png\",\n \"29.png\",\n \"39.png\", \n \"49.png\", \n \"59.png\",\n ...\n] # etc.\n\nAnd then define a function that can handle fetching these images as needed\ndef set_image(correct): # I assume 'correct' is an integer\n img = ImageTk.PhotoImage(\n Image.open(\n # open the correct image (-1 to accommodate zero-indexing)\n image_dir.joinpath(images[correct - 1]) \n )\n )\n return img\n\nThen you can update your button's image like so, for example:\nself.rez.configure(image=img)\n\n" ]

[ 0, 0 ]

[]

[]

[ "python", "tkinter" ]

[ "The barcode field in res.partner is a property field and stored in ir.property model which name is Company Propeties in Odoo and you can access it with developer mode from Settings -> Technical -> Company Propeties.\nThe _name_search method for res.partner enable you to search in any Many2one partner relation field in any model by one of these fields display_name, email, reference and vat and you can override it to add barcode as below:\nfrom odoo import api, models\nfrom odoo.osv.expression import get_unaccent_wrapper\nimport re\n\n\nclass ResPartner(models.Model):\n _inherit = 'res.partner'\n\n @api.model\n def _name_search(self, name, args=None, operator='ilike', limit=100, name_get_uid=None):\n self = self.with_user(name_get_uid) if name_get_uid else self\n # as the implementation is in SQL, we force the recompute of fields if necessary\n self.recompute(['display_name'])\n self.flush()\n print(args)\n if args is None:\n args = []\n order_by_rank = self.env.context.get('res_partner_search_mode')\n if (name or order_by_rank) and operator in ('=', 'ilike', '=ilike', 'like', '=like'):\n self.check_access_rights('read')\n where_query = self._where_calc(args)\n self._apply_ir_rules(where_query, 'read')\n from_clause, where_clause, where_clause_params = where_query.get_sql()\n from_str = from_clause if from_clause else 'res_partner'\n where_str = where_clause and (\" WHERE %s AND \" % where_clause) or ' WHERE '\n print(where_clause_params)\n # search on the name of the contacts and of its company\n search_name = name\n if operator in ('ilike', 'like'):\n search_name = '%%%s%%' % name\n if operator in ('=ilike', '=like'):\n operator = operator[1:]\n\n unaccent = get_unaccent_wrapper(self.env.cr)\n\n fields = self._get_name_search_order_by_fields()\n\n query = \"\"\"SELECT res_partner.id\n FROM {from_str}\n LEFT JOIN ir_property trust_property ON (\n trust_property.res_id = 'res.partner,'|| {from_str}.\"id\"\n AND trust_property.name = 'barcode')\n {where} ({email} {operator} {percent}\n OR {display_name} {operator} {percent}\n OR {reference} {operator} {percent}\n OR {barcode} {operator} {percent}\n OR {vat} {operator} {percent})\n -- don't panic, trust postgres bitmap\n ORDER BY {fields} {display_name} {operator} {percent} desc,\n {display_name}\n \"\"\".format(from_str=from_str,\n fields=fields,\n where=where_str,\n operator=operator,\n email=unaccent('res_partner.email'),\n display_name=unaccent('res_partner.display_name'),\n reference=unaccent('res_partner.ref'),\n barcode=unaccent('trust_property.value_text'),\n percent=unaccent('%s'),\n vat=unaccent('res_partner.vat'), )\n\n where_clause_params += [search_name] * 4 # for email / display_name, reference\n where_clause_params += [re.sub('[^a-zA-Z0-9\\-\\.]+', '', search_name) or None] # for vat\n where_clause_params += [search_name] # for order by\n if limit:\n query += ' limit %s'\n where_clause_params.append(limit)\n print(query)\n print(where_clause_params)\n self.env.cr.execute(query, where_clause_params)\n return [row[0] for row in self.env.cr.fetchall()]\n\n return super(ResPartner, self)._name_search(name, args, operator=operator, limit=limit, name_get_uid=name_get_uid)\n\n\n" ]

[ 0 ]

[]

[]

[ "erp", "odoo", "odoo_14", "python" ]

[ "If you're open to another package, you could wrap it up with dask.\nfrom numpy.random import rand, randint\nfrom dask import array as da\n\nA = da.from_array(rand(1000,5,5))\nB = da.from_array(rand(40000000,5,1))\nI = da.from_array(randint(low=0, high=1000, size=40000000))\n\nfancy = A[I] @ B\n\n\nAfter finished manipulating, then bring it into memory using fancy.compute()\n" ]

[ 1 ]

[]

[]

[ "memory", "numpy", "python", "vectorization" ]

[ "A shell of Automator has different PATH, setopt, and other values from your shell in terminal. Therefore the script in your Automator won't work exactly the same in your shell.\nTo make it work, you need to use an absolute path of the python command.\nRun this command in your terminal to get the absolute path.\nwhich python\n\nAfter, fix python in your script with the previous result.\n#!/bin/bash\n\necho Running Script\n\n/Absolute/path/somewhere/python /Desktop/test.py\n\necho Script ended\n\nTo debug with showing a window of the script result\nAdd \"Run Applescript\" in your automator app with the code below.\nIt's not perfect but useful.\nAlso, in your picture, you can see the result by click the Results button.\non run {input, parameters}\n display dialog input as text\nend run\n\n" ]

[ 0 ]

[]

[]

[ "automator", "macos", "python" ]

[ "This is because ct = ColumnTransformer(transformers=[('encoder',OneHotEncoder(),[29])],remainder = 'passthrough') will one-hot encode the column of index 29.\nYou are fit-transforming y which only has 1 column. You can change the 29 to 0.\nct = ColumnTransformer(transformers=[('encoder',OneHotEncoder(),[0])],remainder = 'passthrough')\n\nEdit\nYou also need to change the iloc to keep the numpy array as column structure.\ny = dataset.iloc[:, [29]].values\n\n" ]

[ 0 ]

[]

[]

[ "artificial_intelligence", "data_preprocessing", "machine_learning", "python", "scikit_learn" ]

[ "You can implement your own logic, like this:\na = \"abbbbbbc\"\nb = \"bbb\"\n\ncount = 0\nfor i in range(len(a) - len(b)):\n if a[i:i+len(b)] == b:\n count += 1\nprint(count)\n\nOR\ncount = 0\nfor i in range(len(a)):\n if a[i:].startswith(b):\n count += 1\nprint(count)\n\nOR\ncount = sum([1 if a[i:].startswith(b) else 0 for i in range(len(a))])\nprint(count)\n\n" ]

[ 1 ]

[]

[]

[ "count", "python", "string" ]

[ "Found an answers in the documentation of sqlalchemy.\nfrom sqlalchemy.schema import CreateTable\nprint(CreateTable(my_mysql_table).compile(mysql_engine))\n\nCREATE TABLE my_table (\nid INTEGER(11) NOT NULL AUTO_INCREMENT,\n...\n)ENGINE=InnoDB DEFAULT CHARSET=utf8mb4\n\nSQLAlchemy Documentation!\n" ]

[ 1 ]

[]

[]

[ "python", "sqlalchemy" ]

[ "Maybe you forgot \" ; \" or/and \" mm \",\nit works:\n@page {\n size: A4; /* Change from the default size of A4 */\n margin: 0mm; /* Set margin on each page */\n }\n\n", "The weasyprint uses 3 sources of css, one of them is default user agent stylesheet\n(https://doc.courtbouillon.org/weasyprint/stable/api_reference.html#supported-features)\nThat defines:\nbody {\n display: block;\n margin: 8px;\n}\n\nmake sure to override that margin on tag and you will loose the margin.\n" ]

[ 13, 0 ]

[]

[]

[ "flask", "margin", "pdf", "python", "weasyprint" ]

[ "convert the integer to a string, and then use the len() method on the converted string. Unless you also consider taking floats as input too, and not integers exclusively.\n", "This is a Python3 behaviour. / returns float and not integer division.\nChange your code to:\ndef count_digit(num):\n if (num//10 == 0):\n return 1\n else:\n return 1 + count_digit(num // 10)\n\nprint(count_digit(23))\n\n", "Recursive\ndef count_digit(n):\n if n == 0:\n return 0\n return count_digit(n // 10) + 1\n\nEasy\ndef count_digit(n):\n return len(str(abs(n)))\n\n", "Assuming you always send integer to function and you are asking for a mathematical answer, this might be your solution:\nimport math\n\n\ndef count_digits(number):\n return int(math.log10(abs(number))) + 1 if number else 1\n\n\nif __name__ == '__main__':\n print(count_digits(15712))\n # prints: 5\n\n", "Here is an easy solution in python;\nnum=23\ntemp=num #make a copy of integer\ncount=0\n\n#This while loop will run unless temp is not zero.,so we need to do something to make this temp ==0\n\nwhile(temp): \n temp=temp//10 # \"//10\" this floor division by 10,remove last digit of the number.,means we are removing last digit and assigning back to the number;unless we make the number 0\n count+=1 #after removing last digit from the number;we are counting it;until the number becomes Zero and while loop becomes False\nprint(count)\n\n" ]

[ 1, 1, 0, 0, 0 ]

[]

[]

[ "count", "filter", "python" ]

[ "If you mean stacked horizontal bar chart, this can help.\ndf.plot.barh(x=\"type\", stacked=True, figsize=(10, 5))\nplt.show()\n\n\n" ]

[ 2 ]

[]

[]

[ "matplotlib", "pandas", "python" ]

[ "Well, you say that you first start the server and then type in the commands. That's also what the terminal feed you shared shows. \nDo not run the server if you want to run management commands using manage.py.\nHit Ctrl+C to exit the server and then run your migration commands, it will work.\n", "Try: \npython manage.py makemigrations\npython manage.py migrate\n\n", "@adam-karolczak n all\nIf there are multiple DJANGO Projects, it can happen that DJANGO_SETTINGS_MODULE is set to some other app in environment varibles, the current project manage.py will not point to current project settings thus the error.\nSo, confirm DJANGO_SETTINGS_MODULE in fact points to the settings.py of current project.\nClose the project if its running viz. ctrl+C. \nYou can also check the server is not running ( linux ) by\nps -ef | grep runserver\n\nThen kill the process ids if they exist.\nIf you confirmed settings.py in DJANGO_MODULE_SETTINGS is for the project you are having issue.\nRun the following it should resolve.\npython manage.py makemigrations\npython manage.py migrate\n\nHope it helps.\n", "I was getting the same error \nrunning this 2 command in terminal\n python manage.py makemigrations\n python manage.py migrate\n\nand then\n python manage.py runserver\n\nsolved my issues.\nThanks\n", "Have you tried with parameter?\npython manage.py makemigrations <app_name>\n", "I had the same issue and the problem was that there was a pg_dump script running at the same time I was trying to migrate. After the dump was completed, migrations ran successfully. \n", "\nCheck that INSTALL_APPS app exists, if not add it\n\nChecks the model for default attributes\n\nRunning this 2 command in terminal\npython manage.py makemigrations\npython manage.py migration\n\n\n", "\nFirst exit of the present web server by typing Ctrl + C\nThen run python manage.py migrate\n\nThe Warning is due to not configuring the initial database or migrating.\n" ]

[ 12, 10, 2, 2, 1, 0, 0, 0 ]

[]

[]

[ "django", "python", "python_3.5", "python_3.x" ]

[ "using the pip install pyqt5-tools method I found the designer on this path:\nC:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python39\\Lib\\site-packages\\qt5_applications\\Qt\\bin\n", "On my system QT Designer is saved under C:\\Users\\User\\AppData\\Local\\Qt Designer\nEDIT:\nIt seems like I installed QT Designer differently.\nYou can use pip install PyQt5Designer.\nThen it should be in the path I gave.\n", "pip install pyqt5-tools\nCheck the path your_python_installed\\Lib\\site-packages\\qt5_applications\\Qt\\bin\\designer.exe\n", "pip install pyqt5-tools\n\nI Found designer.exe in:\n%APPDATA%\\Roaming\\Python\\[Version]\\site-packages\\qt5_applications\\Qt\\bin\n\n", "I found it on path\nvenv\\Lib\\site-packages\\qt5_applications\\Qt\\bin\n", "If this helps someone, in my case there were two \"similar\" of pyqt5 folders inside site-packages, ones starts with pyqt5 and others with just qt5, I found the designer app in the folder qt5_aplications.\n", "I searched the site-packages folder and found out that the designer.exe now exists under the ./Lib/site-packages/qt5_applications/Qt/bin \nI am currently using Python 3.9.13\n" ]

[ 10, 4, 2, 2, 1, 0, 0 ]

[]

[]

[ "pip", "python" ]

[ "In my opinion, a number is expected, and the resulting value is a string. I tried removing the quotes in the conditions themselves. No errors occur. I also added one more line to the dataframe for verification. The flags are displayed as they should. The last line gives: False.\nimport pandas as pd\nimport numpy as np\n\ndata = [[1, 1980.0, 2000.0], [2, 1990, 3000.0]]\n\ndf = pd.DataFrame(data, columns=[\"Item\", \"year1\", \"start_year\"])\n\n\ndata = [[\"year1\", (df['year1'] < 1985), ],\n [\"year2\", ((df['year1'] < 1985) | ((df['start_year'] - df['year1']) < 10)), ], ]\n\ncondition_df = pd.DataFrame(data, columns=[\"column\", \"condition\"])\n\n\nfor idx, row in condition_df.iterrows():\n col = row[\"column\"]\n condition = row[\"condition\"]\n\n flag_col_name = f\"{col}_flag\"\n\n df = df.assign(flag_col_name=lambda df: np.where(condition, True, False))\n print(df)\n\n" ]

[ 0 ]

[]

[]

[ "conditional_statements", "dataframe", "loops", "pandas", "python" ]

[ "with open('new_student_data.txt', 'w') as f:\n for i in vars(item).values():\n f.write(f\"{i}\\n\")\n\nIf file.writelines only takes an iterable and doesn't support *args, you can always iterate over your list and write it with file.write.\n", "You can just print as follows:\nwith open('new_student_data.txt', 'w') as f:\n print(*vars(item).values(), file=f)\n\n", "Every class is capable of outputting a dictionary by calling __dict__ method. This way you are getting the attribute name and the value as a dictionary, then you could simply iterate through it and write it to a txt file:\nwith open('params.txt'), 'w') as f:\n for param, value in params.__dict__.items():\n if not param.startswith(\"__\"):\n f.write(f\"{param}:{value}\\n\")\n\nHere I am also getting rid of default class attributes that start with \"__\". Hope this helps\n" ]

[ 1, 1, 1 ]

[]

[]

[ "python" ]

[ "Create a list to store the group totals.\nRead the file a line at a time. Try to convert each line to int. If that fails then you're at a group separator so append zero to the group_totals list\nSort the list and print the last 3 items\nFILENAME = '/Users/dan/Desktop/day1 copy.txt'\n\ngroup_totals = [0]\n\nwith open(FILENAME) as data:\n for line in data:\n try:\n group_totals[-1] += int(line)\n except ValueError:\n group_totals.append(0)\n print(sorted(group_totals)[-3:])\n\nOutput:\n[740, 1350, 2000]\n\nNote:\nThis code implicitly assumes that there are at least 3 groups of values\n" ]

[ 0 ]

[]

[]

[ "file", "python" ]

[ "I found the answer out so I will be posting it here.\nI used a window function:\nm_window = Window.partitionBy([\"gender\",\"week\",\"type\"]).orderBy(\"share\")\nThen I create a column using the function first and over window like this:\ndf.withColumn(\"first\", first(\"units\").over(m_window))\n" ]

[ 1 ]

[]

[]

[ "conditional_statements", "dataframe", "pyspark", "python" ]

[ " icon = weather['icon']\n\nThis sets a variable icon to reference the string inside the dictionary.\n icon = 'https://dar-group-150-holborn.s3.eu-west-2.amazonaws.com/images/01d.svg'\n\nThis reassigns that variable to a URL string. It does NOT change the dictionary like you might think.\n context = {'weather_data' : weather_data, 'icon': icon}\n\nAfter the loop, you set a single value in the context which will be the last icon url.\nTwo suggestions:\n\nDon't reassign a variable to mean something else. Use two different variables instead. So instead of\n icon = 'https://dar-group-150-holborn.s3.eu-west-2.amazonaws.com/images/01d.svg'\n\ndo\n icon_url = 'https://dar-group-150-holborn.s3.eu-west-2.amazonaws.com/images/01d.svg'\n\n\nBetter yet, store the url in each dictionary:\n weather['icon_url'] = 'https://dar-group-150-holborn.s3.eu-west-2.amazonaws.com/images/01d.svg'\n\nNow you will have a url for the icon in each city.\n\nYou can build the URL directly from the name of the icon without all the if statements. Do this just once instead of 50 times:\n weather['icon_url'] = f'https://dar-group-150-holborn.s3.eu-west-2.amazonaws.com/images/{icon}.svg'\n\nAlternatively, you can do this directly in the template. Since you didn't share your template, I can't give any details beyond this vague hint.\n\n\n" ]

[ 1 ]

[]

[]

[ "django", "python" ]

[ "According to the manual, numpy.sum uses pairwise summation to get more precision. Another common algorithm is Kahan summation.\nAnyway, I wouldn't count too much on Numpy and MATLAB giving the same result up to the last bit, as there might me operation reordering if computations are made in parallel. See this for the kind of problem that can arise.\nHowever, we can cheat a little bit and force Python to do the sum without the extra precision:\nhexPDFSum = num2hex(np.sum(np.hstack((np.reshape(unnormalizedPDF / np.sum(unnormalizedPDF), (n, 1)), np.zeros((n, 1)))), 0)[0])\nhexPDFSum\n'0x3fefffffffffffff'\n\n" ]

[ 3 ]

[]

[]

[ "ieee_754", "matlab", "numpy", "python", "python_3.x" ]

[ "\nI want the list to save 5 different inputs sent from client\n\nThen you have to rearrange your code a bit:\n list1 = []\n while connected:\n msg_length = conn.recv(HEADER).decode(FORMAT)\n if msg_length:\n msg_length = int(msg_length)\n msg = conn.recv(msg_length).decode(FORMAT)\n if msg == DISCONNECT_MSG:\n connected = False\n if len(list1) < 5:\n values = msg\n list1.append(values)\n print(list1)\n\n" ]

[ 0 ]

[]

[]

[ "list", "python", "server", "sockets" ]

[ "Here you go, it's working without an error:\npage_number = 0\ndf_main = pd.DataFrame()\nwhile True:\n for address in wallet_addresses:\n url=f'https://api.etherscan.io/api?module=account&action=txlist&address={address}&startblock=0&endblock=99999999&page={page_number}&offset=10&sort=asc&apikey={ether_api}'\n output = requests.get(url).text\n df_temp = pd.DataFrame(json.loads(output)['result'])\n df_temp['wallet_address'] = address\n page_number += 1\n df_temp['timeStamp'] = pd.to_datetime(df_temp['timeStamp'], unit='s')\n df_main = df_main.append(df_temp)\n if min(pd.to_datetime(df_main['timeStamp']).dt.date) < datetime(2022, 1, 1).date():\n pass\n\n\nWrong append\nSo, what has happened here. As suggested in the first comment under question we acknowledged the type of first record in df_main with type(df_main['timeStamp'].iloc[0]). With IPython and Jupyter-Notebook one can look what is happening with df_main just after receiving an error with it being populated on the last for loop iteration that failed.\nOtherwise if one uses PyCharm or any other IDE with a possibility to debug, the contents of df_main can be revealed via debug.\nWhat we were missing, is that df_main = df_main.append(df_temp) is placed in a slightly wrong place. On first iteration it works well, pd.to_datetime(df_main['timeStamp'], unit='s') gets an str type with Linux epoch and gets converted to pandas._libs.tslibs.timestamps.Timestamp.\nBut on next iteration df_main['timeStamp'] already has the Timestamp type and it gets appended with str type, so we get a column with mixed type. E.g.:\ntype(df_main['timeStamp'].iloc[0]) == type(df_main['timeStamp'].iloc[-1])\n\nThis results with False. Hence when trying to convert Timestamp to Timestamp one gets an error featured in question.\nTo mitigate this we can place .append() below the conversion and do this conversion on df_temp instead of df_main, this way we will only append Timestamps to the resulting DataFrame and the code below with if clause will work fine.\nAs a side note\nAnother small change I've made was datetime.date(2022, 1, 1). This change was not needed, but the way one works with datetime depends on how this library was imported, so it's worth mentioning:\nimport datetime\ndatetime.date(2022, 1, 1)\ndatetime.datetime(2022, 1, 1).date()\n\nfrom datetime import datetime\ndatetime(2022, 1, 1).date()\n\nAll the above is legit and will produce the same. On the first import module gets imported, on the second one type gets imported.\n\nAlternative solution\nConversion to Timestamp takes time. If the API provides Linux epoch dates, why not use this date for comparison? Let's add this somewhere where you define wallet_addresses:\nreference_date = \"01/01/2021\"\nreference_date = int(time.mktime(datetime.datetime.strptime(reference_date, \"%d/%m/%Y\").timetuple()))\n\nThis will result in 1609448400. Other stack overflow question as reference.\nThis integer can now be compared with timestamps provided by the API. The only thing left is to cast str to int. We can have your code left intact with some minor changes at the end:\n<< Your code without changes >>\n df_main['timeStamp'] = df_main['timeStamp'].astype(int)\n if min(df_main['timeStamp']) < reference_date:\n pass\n\nTo make a benchmark I've changed while True: to for _ in range(0,4): to limit the infinite cycle, results are as follows:\n\nInitial solution took 11.6 s to complete\nAlternative solution took 8.85 s to complete\n\nIt's 30% faster. Casting str to int takes less time than conversion to TimeStamps, I would call this a preferable solution.\nFuture warning\nFutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.\n\nIt makes sense to comply with this warning. df_main = df_main.append(df_temp) has to be changed to df_main = pd.concat([df_main, df_temp]).\nAs for current 1.5.0 version it's already deprecated. Time to upgrade!\n" ]

[ 1 ]

[]

[]

[ "dataframe", "datetime", "etherscan", "pandas", "python" ]

[ "You assign an integer to guesses here:\nguesses = 0\n\nSo the interpreter is right saying you CANNOT append to int. Define it as a list:\nguesses = []\n\nBut there's more:\n\nYou ask for input BEFORE the loop, so it happens only once, later the loop is infinite, cause no new input is ever provided\nIf you need only current and previous value you don't need a list at all, rather 3 integers (current, previous, counter - to print number of guesses if you wish to do that)\nIf you want to stick to the list you'll get an IndexError next, since there is no guesses[-2] element during 1st iteration (and you don't check the length of the list before trying to access that)\nDo NOT call variables like \"cg\" it means nothing, abbreviations depend on a context (which you might have or might not have), now it's a simple program and you can instantly see that it's probably \"current_guess\", but that's not the case in general, the IDE should make your life easier and give you possibility to auto insert once defined name, so if somebody says it's time consuming they are plainly wrong\n\n", "\nwhile True:\n guess = int(input(\"Enter Guess Here: \\n\"))\n \n if guess < 1 or guess > 100:\n print('OOB, try again: ')\n continue\n \n # compare player guess to number\n if guess == win_num:\n print(f'you did it in {len(guess_list)} congrats')\n break\n \n #if guess is wrong add to guess to list\n guess_list.append(guess)\n \n #\n if guess_list[-2]:\n if abs(win_num - guess) < abs(win_num - guess_list[-2]):\n print('warmer')\n else:\n print('colder')\n \n \n else:\n if abs(win_num - guess) <= 10:\n print('warm')\n else:\n print('cold')\n \n \n\n\n" ]

[ 0, 0 ]

[]

[]

[ "error_handling", "list", "python" ]

[ "here is one way :\ndf['date'] = pd.to_datetime(df['date'])\n\ndf['delta_t'] = (df['date'] - df['date'].shift(1)).dt.total_seconds()\nprint(df)\n\noutput :\n>>\n date delta_t\n0 2022-11-09 00:00:00 NaN\n1 2022-11-09 00:00:30 30.0\n2 2022-11-09 00:01:00 30.0\n3 2022-11-09 00:01:30 30.0\n4 2022-11-09 00:02:00 30.0\n\n" ]

[ 0 ]

[]

[]

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

[ "You have a couple of options\neval\neval('[1, 2]')\n# [1, 2]\n\nast.literal_eval\nimport ast\nast.literal_eval('[1, 2]')\n# [1, 2]\n\nstring parsing\nlist(map(int, '[1, 2]'.strip('[]').split(', ')))\n# [1, 2]\n\n" ]

[ 2 ]

[]

[]

[ "list", "python" ]

[ "To save the \"comment\" metadata in the JPEG file, you can use the Image.save() method with the save_all=True and exif=img.app arguments. This will preserve the metadata in the JPEG file.\nHere is an example:\nfrom PIL import Image\n\n# open the image\nimg = Image.open(path)\n\n# save the image with the comment metadata preserved\nimg.save(new_path, save_all=True, exif=img.app)\nimg.close()\n\n# now open the new image to see if the metadata has been preserved\nnew_img = Image.open(new_path)\nprint(new_img.info['comment'])\n\nYou can also specify the comment metadata as a dictionary in the Image.save() method directly, instead of using the img.app property:\nfrom PIL import Image\n\n# open the image\nimg = Image.open(path)\n\n# create a dictionary with the comment metadata\ncomment_metadata = {'comment': \"this is my comment metadata\"}\n\n# save the image with the comment metadata preserved\nimg.save(new_path, save_all=True, exif=comment_metadata)\nimg.close()\n\n# now open the new image to see if the metadata has been preserved\nnew_img = Image.open(new_path)\nprint(new_img.info['comment'])\n\n", "Just been having a play with this and I couldn't see anything directly in Pillow to support this. I've found that the save() method supports a parameter called extra that can be used to pass arbitrary bytes to the output file.\nWe then just need a simple method to turn a comment into a valid JPEG segment, for example:\nimport struct\nfrom PIL import Image\n\ndef make_jpeg_variable_segment(marker: int, payload: bytes) -> bytes:\n \"make a JPEG segment from the given payload\"\n return struct.pack('>HH', marker, 2 + len(payload)) + payload\n\ndef make_jpeg_comment_segment(comment: bytes) -> bytes:\n \"make a JPEG comment/COM segment\"\n return make_jpeg_variable_segment(0xFFFE, comment)\n\n# open source image\nwith Image.open(\"foo.jpeg\") as im:\n # save out with new JPEG comment\n im.save('bar.jpeg', extra=make_jpeg_comment_segment(\"hello world\".encode()))\n\n# read file back in to ensure comment round-trips\nwith Image.open('bar.jpeg') as im:\n print(im.app['COM'])\n print(im.info['comment'])\n\nNote that in my initial attempts I tried appending the comment segment at the end of the file, but Pillow wouldn't load this comment even after calling the .load() method to force it to load the entire JPEG file.\nIt would be nice if this was supported natively, but it doesn't seem to do it yet!\n" ]

[ 1, 1 ]

[]

[]

[ "jpeg", "python", "python_imaging_library" ]

[ "As you can see by the error message, the issue is with ffprobe and not ffmpeg.\nMake sure that both ffprobe.exe and ffmpeg.exe are in the executable path.\n\nOne option is placing ffprobe.exe and ffmpeg.exe in the same folder as the Python script (D:\\Scripts\\captcha\\ in your case).\nOther option is adding the folder of ffprobe.exe and ffmpeg.exe to Windows system path.\n(Placing the EXE files in a folder that is already in the system path may also work).\n\n\nDebugging Pydub source code (pydub-0.25.1):\nThe code fails in the following code:\nhp, ht, pid, tid = _winapi.CreateProcess(executable, args,\n # no special security\n None, None,\n int(not close_fds),\n creationflags,\n env,\n os.fspath(cwd) if cwd is not None else None,\n startupinfo)\n\nWhere args = 'ffprobe -of json -v info -show_format -show_streams test.mp3'\nWe are getting there from:\ninfo = mediainfo_json(orig_file, read_ahead_limit=read_ahead_limit)\nFrom:\nprober = get_prober_name()\nHere is the source code of get_prober_name method:\ndef get_prober_name():\n \"\"\"\n Return probe application, either avconv or ffmpeg\n \"\"\"\n if which(\"avprobe\"):\n return \"avprobe\"\n elif which(\"ffprobe\"):\n return \"ffprobe\"\n else:\n # should raise exception\n warn(\"Couldn't find ffprobe or avprobe - defaulting to ffprobe, but may not work\", RuntimeWarning)\n return \"ffprobe\"\n\nThe which method looks for the command is in the execution path (returns None if it is not found).\nAs you can see from Pydub sources, ffprobe.exe should be in the execution path.\n\n\nNote\nFor setting FFmpeg path we may also apply something like:\n import pydub\n pydub.AudioSegment.converter = 'c:\\\\FFmpeg\\\\bin\\\\ffmpeg.exe'\n sound = pydub.AudioSegment.from_mp3(\"test.mp3\")\n\n\n\nBut there is no such option for FFprobe.\n", "If you haven't installed the ffmpeg/ffprobe as @Rotem answered, you can use my ffmpeg-downloader package:\npip install ffmpeg-downloader\nffdl install --add-path\n\nThe --add-path option adds the installed FFmpeg folder to the user's system path. Re-open the Python window and both ffmpeg and ffprobe will be available to your program.\n" ]

[ 0, 0 ]

[]

[]

[ "ffmpeg", "ffprobe", "pydub", "python", "selenium" ]

[]

[]

[ "1-) Classical solution: Rent a server, Deploy MySQL to server, Use bootstrap templates for web application design in which users add words, use PHP to connect MySQL database server (for adding and displaying words\n2-) Modern solution: Open 3 month free usage Google cloud account, create Firebase database, create program or web app and connect to firebase using credential keys with python\n", "TLDR: I recommend Python + SQLite\nReasoning for a relational DB\n\nYour data is very structured - therefore you can utilize the structure guarantees of a relational DB.\nJoins are very effective with relational DBs and can be done all in one query.\nThe size of your data should not exceed 1M rows, which is the perfect size for relational DBs.\nData is persisted on the hard drive, and you have all the nice ACID properties.\n\nReasoning for SQLite\n\nTakes about 5 Minutes to set it up.\n\nResoning for Python\nThis is entirely opinionated, in my experience the JS-database-connectors feel a lot less stable. Python is a perfect language for a beginner and is very rewarding when prototyping a project such as yours.\n\nAdditional: Redis will be overkill - that's for very high-speed applications with very small access times.\n" ]

[ -1, -1 ]

[ "database", "dictionary", "javascript", "python", "web" ]

[ "add a import for phi\nfrom sympy.abc import phi\n\n" ]

[ 2 ]

[]

[]

[ "jupyter_notebook", "math", "python", "python_3.x", "symbolic_math" ]

[ "I haven't found any better way than suppressing the warning using the warnings module:\nimport numpy as np\nimport pandas as pd\nimport warnings\n\ndf = pd.DataFrame({\"price\": [11.1, 12.2]}, index=[\"book1\", \"book2\"])\noriginal_prices = df[\"price\"]\nnew_prices = np.array([98, 99])\nwith warnings.catch_warnings():\n # Setting values in-place is fine, ignore the warning in Pandas >= 1.5.0\n # This can be removed, if Pandas 1.5.0 does not need to be supported any longer.\n # See also: https://stackoverflow.com/q/74057367/859591\n warnings.filterwarnings(\n \"ignore\",\n category=FutureWarning,\n message=(\n \".*will attempt to set the values inplace instead of always setting a new array. \"\n \"To retain the old behavior, use either.*\"\n ),\n )\n\n df.iloc[:, 0] = new_prices\n\ndf.iloc[:, 0]\n\n", "Post here because I can't comment yet.\nFor now I think I will also suppress the warnings, because I don't want the old behavior, never expected to use it that way. And the suggested syntax has the danger to trigger the SettingWithCopyWarning warning.\n", "As the changelog states, the warning is printed when setting an entire column from an array with different dtype, so adjusting the dtype is one way to silence it:\ndf = pd.DataFrame({'price': [11.1, 12.2]}, index=['book1', 'book2'])\noriginal_prices = df['price']\nnew_prices = np.array([98, 99]).astype(float)\ndf.iloc[:, 0] = new_prices\ndf.iloc[:, 0]\n\nNote the additional .astype(float). Not an ideal solution, but a solution.\n" ]

[ 4, 0, 0 ]

[ "I am just filtering all future warnings for now:\nimport warnings\nwarnings.simplefilter(\"ignore\", category=FutureWarning)\n\n" ]

[ -2 ]

[ "pandas", "python" ]

[ "You need to split the set_xticks() to have only the number of entries - (0,1,2) and use set_xticklables() to give the text you want to display - (25,80,95). Note that I have used numpy's linspace to get the list of numbers based on length of percentile_value. Also, I have removed the pad=-15 as you have indicated that you want the numbers below the ticks. Hope this is what you are looking for.\nimport matplotlib.pyplot as plt\nimport numpy as np\ndef box(ax_position, percentile_values, label, label_position):\n fig = plt.figure(figsize = (10, 2))\n ax1 = fig.add_axes(ax_position) \n ax1.set_xticks(np.linspace(0,len(percentile_values)-1,len(percentile_values)))\n ax1.set_xticklabels(percentile_values)\n ax1.tick_params(axis=\"x\",direction=\"out\", colors='b')\n ax1.set_yticks([])\n ax1.text(*label_position, label, size=8)\n plt.show()\n\nbox((0.4, 0.5, 0.4, 0.1), (25,80,95), \"CA\",(0.01, 1.3))\n\n\n" ]

[ 0 ]

[]

[]

[ "matplotlib", "python" ]

[ "It looks like you have tab delimiter, not comma, so replace\nreader = csv.reader(csvfile, delimiter=',')\n\nwith\nreader = csv.reader(csvfile, delimiter='\\t')\n\nI would note that you could replace this whole block with\nimport numpy as np\nworld_data = np.genfromtext(f'level{level}_data.csv', delimiter='\\t')\n\n" ]

[ 0 ]

[]

[]

[ "python" ]

[ "Looks like your \"y\" argument accepts only list [ele1, ele2], not a tuple(ele1, ele2). I changed the brackets to squares and I ran your code just fine:\n import plotly.express as px\n figure = px.line(df, x=\"Time Period\", \n y=[\"Fin\",\"Legal\",\"Leadership\",\"Overall\"],\n title='Quarterly scores')\n\nfigure.show()\n\nproduces:\nthis\n" ]

[ 0 ]

[]

[]

[ "matplotlib", "pandas", "python" ]

[ "It's a common problem due to {barcodes}{sample} pattern.\nSnakemake won't know where {barcodes} ends and where {sample} starts without a wildcard_constraint. Right now, snakemake is thinking that your sample wildcard is just a 0.\n" ]

[ 1 ]

[]

[]

[ "bash", "bioinformatics", "python", "snakemake" ]

[ "I think you're going to have to use some sort of machine learning based text analysis to figure out if it's the context you're looking for.\n" ]

[ 0 ]

[]

[]

[ "python", "selenium", "web_scraping" ]

[ "I'm not sure what I was doing wrong before but I was able to get the following variations working.\nEdge(color=\"black\", label=\"texthere\", fontsize=\"20\")\n\nfont = \"20\"\nEdge(color=\"black\", label=\"texthere\", fontsize=font)\n\nThere is also a head and tail label. See the following examples:\nEdge(color=\"red\", headlabel=\"port info\", labelfontsize=\"20\")\n\n" ]

[ 0 ]

[]

[]

[ "diagram", "python" ]

[ "student_profile = Profile.objects.create( # Profile\n user=user, #user\n surname=student_surname,\n initials=student_initials,\n entrance_number=student_entrance,\n email=student_email,\n father=student_father,\n skills=student_other_skills,\n sports=student_sports,\n birthday=student_birthday,\n contact=student_contact,\n address=student_residence,\n )\n\nNot User model, must be Profile model, your model field is user, but you have used User\n" ]

[ 1 ]

[]

[]

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

[ "\nhow to get the method parameters type stubs?\n\nPyCharm does generate the docstring stub with the type placeholders, but the placeholders aren't currently (using PyCharm 2022.1) populated from the __annotations__ with the types. This has been marked with the state \"To be discussed\" in the JetBrains bugtracker, see issues PY-23400 and PY-54930.\nInclusion of the type placeholders is configured by checking File > Settings > Editor > General > Smart Keys > Python > Insert type placeholders in the documentation comment stub.\n\n\nIt's worth noting the insert docstring stub intention is selected on the function name itself\n\nIn this example with Napoleon style docstring format selected in File > Settings > Tools > Python Integrated Tools > Docstrings, the result:\n\n" ]

[ 1 ]

[]

[]

[ "docstring", "pycharm", "python" ]

[ "Are you using python 3.10? Looks like beautifulsoup library is using removed deprecated aliases to Collections Abstract Base Classes. More info here: https://docs.python.org/3/whatsnew/3.10.html#removed\nA quick fix is to paste these 2 lines just below your imports:\nimport collections\ncollections.Callable = collections.abc.Callable\n\n", "Andrey, i cannot comment yet. But i tried your fix and Im using Thonny and using 3.10 in terminal. But after adding the two import collections and callable lines. i get another error in Thonny that isnt shown in terminal. when i run the program in terminal it simply seems to do nothing. In Thonny it suggests that \"Module has no attribute \"Callable\"\n" ]

[ 2, 0 ]

[]

[]

[ "beautifulsoup", "python" ]

[ "As for the answer to your specific question, why the attrib list doesn't contain the namespace prefix decls, sorry for the unquenching answer: because they're not attributes.\nhttp://www.w3.org/XML/1998/namespace is a special schema that doesn't act like the other schemas in your userspace. In that representation, xmlns:prefix=\"uri\" is an attribute. In all other subordinate (by parsing sequence) schemas, xmlns:prefix=\"uri\" is a special thing, a namespace prefix declaration, which is different than an attribute on a node or element. I don't have a reference for this but it holds true perfectly in at least a half dozen (correct) implementations of XML parsers that I've used, including those from IBM, Microsoft and Oracle.\nAs for the ugliness of reparsing the file, I feel your pain but it's necessary. As tdelaney so well pointed out, you may not assume that all of your namespace decls or prefixes must be on your root element.\nBe prepared for the possibility of the same prefix being redefined with a different namespace on every node in your document. This may hold true and the library must correctly work with it, even if it is never the case your document (or worse, if it's never been the case so far).\nConsider if perhaps you are shoehorning some text processing to parse or query XML when there may be a better solution, like XPath or XQuery. There are some good recent changes to and Python wrappers for Saxon, even though their pricing model has changed.\n" ]

[ 1 ]

[]

[]

[ "elementtree", "python", "xml", "xml_namespaces" ]

[ "Alternative 1\nResample to a daily resolution, selecting the nearest day to fill in missing values:\ndf2 = df.resample('D').nearest()\ndf2 = df2.loc[df2.index.day == 8]\n\nAlternative 2\nA more general method (and a tiny bit faster) is to generate dates/times of your choice, then use reindex() and method 'nearest'. It is more general because you can use any series of timestamps you could come up with (not necessarily aligned with any frequency).\ndates = pd.date_range(\n start=df.first_valid_index().normalize(), end=df.last_valid_index(),\n freq='D')\ndates = dates[dates.day == 8]\ndf2 = df.reindex(dates, method='nearest')\n\nExample\nLet's start with a reproducible example:\nimport yfinance as yf\n\ndf = yf.download(['AAPL', 'AMZN'], start='2022-01-01', end='2022-12-31', freq='D')\n>>> df.iloc[:10, :5]\n Adj Close Close High\n AAPL AMZN AAPL AMZN AAPL\nDate \n2022-01-03 180.959747 170.404495 182.009995 170.404495 182.880005\n2022-01-04 178.663086 167.522003 179.699997 167.522003 182.940002\n2022-01-05 173.910645 164.356995 174.919998 164.356995 180.169998\n2022-01-06 171.007523 163.253998 172.000000 163.253998 175.300003\n2022-01-07 171.176529 162.554001 172.169998 162.554001 174.139999\n2022-01-10 171.196426 161.485992 172.190002 161.485992 172.500000\n2022-01-11 174.069748 165.362000 175.080002 165.362000 175.179993\n2022-01-12 174.517136 165.207001 175.529999 165.207001 177.179993\n2022-01-13 171.196426 161.214005 172.190002 161.214005 176.619995\n2022-01-14 172.071335 162.138000 173.070007 162.138000 173.779999\n\nNow:\ndf2 = df.resample('D').nearest()\ndf2 = df2.loc[df2.index.day == 8]\n\n>>> df2.iloc[:5, :5]\n Adj Close Close High\n AAPL AMZN AAPL AMZN AAPL\n2022-01-08 171.176529 162.554001 172.169998 162.554001 174.139999\n2022-02-08 174.042633 161.413498 174.830002 161.413498 175.350006\n2022-03-08 156.730942 136.014496 157.440002 136.014496 162.880005\n2022-04-08 169.323975 154.460495 170.089996 154.460495 171.779999\n2022-05-08 151.597595 108.789001 152.059998 108.789001 155.830002\n\nWarning\nReplacing a missing day with data from the future (which is what happens when the nearest day is after the missing one) is called peak-ahead and can cause peak-ahead bias in quant research that would use that data. It is usually considered dangerous. You'd be safer using method='ffill'.\n" ]

[ 1 ]

[]

[]

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

[ "Try pip install wget, maybe you’re using python 2\n", "With pip3 you are installing module for python 3,\nIt can b that you have both versions of python 2 and 3 and you your environment is pointing default to python 2 \nCheck python version or install wget for python 2\npython -V \npip install wget\n\n", "this should not be the case, but check if site-packages is in the path for accessing modules\n>>> import sys\n>>> sys.path\n[..., '...\\\\python3.6\\\\lib\\\\site-packages', ...] ## if this is here I cannot help you\n\nif not, try repairing python\nyou can do that by clicking setup file (one with which you installed in the first place),\nand among 3 options click repair\n", "If you process the python script by command:\npython import.py\n\nor \npython3 import.py\n\nit should work.\nBut if you process the executable python script by command:\n./import.py ENTER\n\nthen incldue as the first line of the script import.py:\n#!/usr/bin/env python\n\nor\n#!/usr/bin/env python3\n\n", "sudo apt-get install --reinstall python3-wget\n\n", "The following command worked for me in Jupyter Lab\n!pip install wget\n\nHope this does help!\n", "In Jupyter Lab, although my Python was 3.9 but it was using 3.7 paths (I have multiple Pythons installed):\nimport sys\n\nsys.path\n\n['D:\\\\Projects',\n 'C:\\\\Program Files\\\\Python37\\\\python37.zip',\n 'C:\\\\Program Files\\\\Python37\\\\DLLs',\n 'C:\\\\Program Files\\\\Python37\\\\lib',\n 'C:\\\\Program Files\\\\Python37',\n '',\n 'C:\\\\Users\\\\John\\\\AppData\\\\Roaming\\\\Python\\\\Python37\\\\site-packages',\n 'C:\\\\Program Files\\\\Python37\\\\lib\\\\site-packages',\n 'C:\\\\Program Files\\\\Python37\\\\lib\\\\site-packages\\\\win32',\n 'C:\\\\Program Files\\\\Python37\\\\lib\\\\site-packages\\\\win32\\\\lib',\n 'C:\\\\Program Files\\\\Python37\\\\lib\\\\site-packages\\\\Pythonwin',\n 'C:\\\\Program Files\\\\Python37\\\\lib\\\\site-packages\\\\IPython\\\\extensions',\n 'C:\\\\Users\\\\John\\\\.ipython']\n\nSo I did !pip3.7 install --user wget, and then it worked.\n", "pip install wget\n\nif in colab use:\n!pip install wget\n\n", "I had the same problem recently and using python3 instead of py worked for me.\n" ]

[ 14, 4, 1, 1, 1, 0, 0, 0, 0 ]

[]

[]

[ "macos", "python", "wget" ]

[ "Based on your comment.\nrequest_id = models.UUIDField(primary_key=False, default=uuid.uuid4().hex, editable=False, unique=True)\n\nYou assigned an instance of the UUID for the default value. In fact, you didn't set a function to generate a new UUID for each record. If you check the related migration file, you can see a line like this:\n('request_id', models.UUIDField(default='97c8a76eefe8445081fcfec3af4f1df2', editable=False, unique=True))\n\nyou set an instance of the UUID as default but the unique property is set, because of this, the first time you save a record is ok but the next record, you face with error.\nActually, you have to set a function for the default to run for each record. You should set the properties like the below:\nrequest_id = models.UUIDField(primary_key=False, default=uuid.uuid4, editable=False, unique=True)\n\n" ]

[ 0 ]

[]

[]

[ "django", "django_rest_framework", "django_views", "mysql", "python" ]

[ "Try this:\nviews.py\n# other views ...\n\ndef company_page(request, slug):\n product_list = Product.objects.filter(company=company).order_by('-pk')\n\n page = request.GET.get('page', 1)\n paginator = Paginator(product_list, 4)\n\n try:\n Products = paginator.page(page)\n except PageNotAnInteger:\n Products = paginator.page(1)\n except EmptyPage:\n Products = paginator.page(paginator.num_pages)\n\n context = {'Products': Products}\n return render(request, 'product/product_list.html', context)\n\nproduct_list.html\n<nav aria-label=\"Pagination\">\n{% if Products.has_other_pages %}\n <ul class=\"pagination justify-content-center my-5\">\n {% if Products.has_previous %}\n <li class=\"page-item mx-auto lead\">\n <a class=\"page-link\" href=\"?page={{ Products.previous_page_number }}\" tabindex=\"-1\" aria-disabled=\"true\">\n Previous</a>\n </li>\n {% else %}\n <li class=\"page-item disabled\">\n <a class=\"page-link\" href=\"#\" tabindex=\"-1\" aria-disabled=\"true\">\n Previous</a>\n </li>\n {% endif %}\n {% for i in Products.paginator.page_range %}\n {% if Products.number == i %}\n <li class=\"page-item mx-auto lead\">\n <a class=\"page-link\" href=\"?page={{ i }}\">\n {{ i }}</a>\n </li>\n {% endif %}\n {% endfor %}\n {% if Products.has_next %}\n\n <li class=\"page-item mx-auto lead\">\n <a class=\"page-link\" href=\"?page={{ Products.next_page_number }}\">\n Next</a>\n </li>\n {% else %}\n <li class=\"page-item disabled mx-auto lead\">\n <a class=\"page-link\" href=\"#!\">\n Next</a>\n </li>\n {% endif %}\n </ul>\n {% endif %}\n\n </nav>\n\n" ]

[ 0 ]

[]

[]

[ "django", "python" ]

[ "When a web element is present in the HTML-DOM but it is not in the state that can be interacted. Other words, when the element is found but we can’t interact with it, it throws ElementNotInteractableException.\nThe element not interactable exception may occur due to various reasons.\n\nElement is not visible\nElement is present off-screen (After scrolling down it will display)\nElement is present behind any other element\nElement is disabled\n\nIf the element is not visible then wait until element is visible. For this we will use wait command in selenium\nwait = WebDriverWait(driver, 10)\nelement = wait.until(EC.element_to_be_clickable((By.ID, 'someid')))\nelement.click()\n\nIf the element is off-screen then we need to scroll down the browser and interact with the element.\nUse the execute_script() interface that helps to execute JavaScript methods through Selenium Webdriver.\nbrowser = webdriver.Firefox()\nbrowser.get(\"https://en.wikipedia.org\")\nbrowser.execute_script(\"window.scrollTo(0,1000)\") //scroll 1000 pixel vertical\n\nReference this solution in to the regarding problem.\n" ]

[ 0 ]

[]

[]

[ "python", "selenium", "web_scraping" ]

[ "Your matrix has 1000 rows of 1000 elements each and you are summing each row 100 times. By my calculation, that is 100,000 tasks you are submitting to the pool passing a one-million element matrix each time. Ouch!\nNow I know you say that the worker function addMatrixRow must have access to the complete matrix. Fine. But instead of passing it a 100,000 times, you can reduce that to 4 times by initializing each process in the pool with a global variable set to the matrix using the initializer and initargs arguments when you construct the pool. You are able to get away with this because the matrix is read-only.\nAnd instead of creating poolArgs as a large list you can instead create a generator function that when iterated returns the next argument to be submitted to the pool. But to take advantage of this you cannot use the map method, which will convert the generator to a list and not save you any memory. Instead use imap_unordered (rather than imap since you do not care now in what order your worker function is returning its results because of the commutative law of addition). But with such a large input, you should be using the chunksize argument with imap_unordered. So that the number of reads and writes to the pool's task queue is greatly reduced(albeit the size of the data being written is larger for each queue operation).\nIf all of this is somewhat vague to you, I suggest reading the docs thoroughly for class multiprocessing.pool.Pool and its imap and imap_unordered methods.\nI have made a few other optimizations replacing for loops with list comprehensions and using the built-in sum function.\nimport random\nfrom multiprocessing import Pool\nimport time\n\n\ndef init_pool_processes(m):\n global matrix\n matrix = m \n\ndef addMatrixRow(rowNum):\n return sum(matrix[rowNum])\n\ndef genMatrix(row, col):\n return [[random.randint(0, 1) for _ in range(col)] for _ in range(row)]\n \ndef compute_chunksize(pool_size, iterable_size):\n chunksize, remainder = divmod(iterable_size, 4 * pool_size)\n if remainder:\n chunksize += 1\n return chunksize\n\ndef main():\n matrix = genMatrix(1000, 1000)\n print(\"generated matrix\")\n MAX_PROCESSES = 4\n\n processPool = Pool(processes=MAX_PROCESSES, initializer=init_pool_processes, initargs=(matrix,))\n start = time.time()\n # Use a generator function:\n poolData = (rowNum for _ in range(100) for rowNum in range(len(matrix)))\n # Compute efficient chunksize\n chunksize = compute_chunksize(MAX_PROCESSES, len(matrix) * 100)\n finalSum = sum(processPool.imap_unordered(addMatrixRow, poolData, chunksize=chunksize))\n end = time.time()\n print(end-start)\n print(f'final sum {finalSum}')\n processPool.close()\n processPool.join()\n\n\nif __name__ == '__main__':\n main()\n\nPrints:\ngenerated matrix\n0.35799622535705566\nfinal sum 49945400\n\nNote the running time of .36 seconds.\nAssuming you have more CPU cores (than 4), use them all for an even greater reduction in time.\n", "you are serializing the entire matrix on each function call, you should only send the data that you are processing to the function, nothing more ... and python has a built-in sum function that has a very optimized C code.\nimport random\nfrom multiprocessing import Pool\nimport time\n\n\ndef addMatrixRow(row_data):\n rowSum = sum(row_data)\n return rowSum\n\n\ndef genMatrix(row, col):\n matrix = list()\n for i in range(row):\n matrix.append(list())\n for j in range(col):\n matrix[i].append(random.randint(0, 1))\n return matrix\n\ndef main():\n matrix = genMatrix(1000, 1000)\n print(\"generated matrix\")\n MAX_PROCESSES = 4\n finalSum = 0\n\n processPool = Pool(processes=MAX_PROCESSES)\n poolData = list()\n\n start = time.time()\n for i in range(100):\n for rowNum in range(len(matrix)):\n matrixData = matrix[rowNum]\n poolData.append(matrixData)\n\n finalData = processPool.map(addMatrixRow, poolData)\n poolData = list()\n finalSum += sum(finalData)\n end = time.time()\n print(end-start)\n print(f'final sum {finalSum}')\n\n\nif __name__ == '__main__':\n main()\n\ngenerated matrix\n3.5028157234191895\nfinal sum 49963400\n\njust not using process pool and running the code serially using list(map(sum,poolData))\ngenerated matrix\n1.2143816947937012\nfinal sum 50020800\n\nso yeh python can do it in a second.\n" ]

[ 3, 1 ]

[]

[]

[ "matrix", "multiprocessing", "process_pool", "python" ]

[ "Instead of using /dev/ttyUSB0 I recommend using the symlinks provided by the kernel in /dev/serial/by-id. They contain a lot of info about the USB device, including the vendor ID and product ID, so you can be sure you are opening the right device. They also should be pretty stable, not depending on the USB port you use or the order the devices are plugged in. Run ls -l /dev/serial/by-id to explore the options.\n" ]

[ 0 ]

[]

[]

[ "python", "raspberry_pi4", "usb" ]

[ "Use the slice notation\ndef swap(value):\n middle = len(value) // 2\n return value[middle:] + value[:middle]\n\nprint(swap(\"abcd\")) # cdab\nprint(swap(\"abcde\")) # cdeab\nprint(swap(\"1234567890\")) # 6789012345\n\n", "s = \"abcdefghijklmn\"\n\nans = []\nn = 2\nflip = True\nfor i in range(0, len(s), n):\n if flip:\n ind = i + n\n else:\n ind = i - n\n flip = not flip\n \n if len(s[ind:ind + n]) > 0:\n ans.append(s[ind:ind + n])\n else:\n ans.append(s[i:])\n\n\"\".join(ans)\n# 'cdabghefklijmn'\n\nOR\ns = \"abcdefghijklmn\"\n\nans = list(s)\nn = 2\n\nfor i in range(0, len(s), 2 * n):\n ans[i + n : i + 2 * n], ans[i : i + n] = ans[i : i + n], ans[i + n : i + 2 * n]\n\"\".join(ans)\n# 'cdabghefklijmn'\n\n" ]

[ 0, 0 ]

[]

[]

[ "algorithm", "python", "string" ]

[ "Looks like you are looking in the file path that ends with \\Lab 5 .py, which will just contain your python script. So you need to look one layer \"above\" your python script, which is the directory containing the script and the iris.csv-file.\nTry: iris_df = pd.read_csv('../iris.csv')\n" ]

[ 1 ]

[]

[]

[ "python" ]

[ "define validation type and validatecommand. validate = key makes with every key input it runs validatecommand. It only types if that function returns true which is 'validate' function in this case.\nvcmd = (root.register(validate), '%P')\ntk.Entry(textvariable = e1,validate=\"key\", validatecommand=vcmd).grid(row=1, column=1)\n\nthis is the validation function\ndef validate(input):\n if not input:\n return True\n elif re.fullmatch(r'[0-9]*',input):\n return True\n myText.set(\"Please check selections, only Integers are allowed\")\n return False\n\nit return true only when its full of numbers([0-9]* is an regular expression which defines all numbers) or empty. If it contains any letter it return False any it denied this way.\nAlso do not forget to imports\nimport re\n\n" ]

[ 0 ]

[]

[]

[ "interface", "python", "tkinter" ]

[ "So your roll are doing:\nIn [118]: x=np.arange(10)\nIn [119]: np.roll(x,-1)\nOut[119]: array([1, 2, 3, 4, 5, 6, 7, 8, 9, 0])\n\nYou can look at the np.roll code; it's probably more general, it has to, in one way or other, copy all the values of x to a new array. This might be a bit faster, since it doesn't try to be as general:\nIn [120]: y=np.zeros_like(x)\n ...: y[:-1] = x[1:]; y[-1] = x[0]\nIn [121]: y\nOut[121]: array([1, 2, 3, 4, 5, 6, 7, 8, 9, 0])\n\ntimes\nNope, it isn't faster:\nIn [122]: x=np.arange(100000)\n\nIn [123]: timeit np.roll(x,-1)\n82.1 µs ± 102 ns per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n\nIn [124]: %%timeit \n ...: y=np.zeros_like(x)\n ...: y[:-1] = x[1:]; y[-1] = x[0]\n ...: \n ...: \n93.4 µs ± 4.84 µs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n\nother timings:\nIn [128]: timeit y=x[1:].copy()\n52.4 µs ± 164 ns per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n\nIn [129]: timeit np.concatenate((x[1:],x[0:1]))\n58.6 µs ± 289 ns per loop (mean ± std. dev. of 7 runs, 10,000 loops each)\n\n" ]

[ 0 ]

[]

[]

[ "numpy", "python", "signal_processing" ]

[ "Try:\npip install --upgrade --force-reinstall coursera-dl\n\nor\npip install --ignore-installed coursera-dl\n\n" ]

[ 0 ]

[]

[]

[ "cmd", "coursera_api", "python" ]

[ "The error stems from the invalid regular expression - specifically, you've omitted the right-bracket.\nHowever, even if you fix that, based on the code shown in the question, this isn't going to work for a couple of reasons.\n\nThe return value from re.search will always be unequal to 'None'\nThe final if test in the code is outside the while loop which is almost certainly not what's wanted.\n\nTry this instead:\nimport string\n\nVALIDCHARS = set(string.ascii_letters+string.digits+'$#@')\n\nfor word in input().split(','):\n if 6 <= len(word) <= 12 and all(c in VALIDCHARS for c in word):\n print(f'{word} is valid')\n else:\n print(f'{word} is invalid')\n\n", "Your regular expression is missing a closing bracket. It should be:\na=re.search(\"[0-9a-zA-Z\\$#@]\",x)\n\nAlso, replace all instances of \"None\" the string with None the keyword. This is because .search returns None as can be seen here.\nhttps://docs.python.org/3/library/re.html#re.search\n" ]

[ 0, 0 ]

[]

[]

[ "python", "python_re", "regex", "search", "tuples" ]

[ "Not sure I understand 100%, but let me try to help.\nThe focus points are:\n# import the pandas library and alias as pd\nimport pandas as pd\n\n# read a csv with the example data\ndf = pd.read_csv(\"data.csv\", sep=\",\", quoting=False, header = None)\n\n# replace any values that match the pattern \"something=value\" with \"value\"\ndf.replace(to_replace=r\"^(.*)=\", value=\"\", regex=True, inplace=True)\n# save to a new csv file:\ndf.to_csv(\"new_data.csv\", sep=\",\", header = None, index = False)\n\n" ]

[ 0 ]

[]

[]

[ "csv", "pandas", "python" ]

[ "XGBoost has also since version 1.3.0 added experimental support for categorical encoding.\nCopying my answer from another question.\nNov 23, 2020\nXGBoost has since version 1.3.0 added experimental support for categorical features. From the docs:\n\n1.8.7 Categorical Data\nOther than users performing encoding, XGBoost has experimental support\nfor categorical data using gpu_hist and gpu_predictor. No special\noperation needs to be done on input test data since the information\nabout categories is encoded into the model during training.\n\nhttps://buildmedia.readthedocs.org/media/pdf/xgboost/latest/xgboost.pdf\nIn the DMatrix section the docs also say:\n\nenable_categorical (boolean, optional) – New in version 1.3.0.\nExperimental support of specializing for categorical features. Do not\nset to True unless you are interested in development. Currently it’s\nonly available for gpu_hist tree method with 1 vs rest (one hot)\ncategorical split. Also, JSON serialization format, gpu_predictor and\npandas input are required.\n\nOther models option:\nIf you don't need to use XGBoost, you can use a model like LightGBM or or CatBoost which support categorical encoding without one-hot-encoding out of the box.\n", "You could use some kind of embeddings that reflect better those cities (and compress the number of total features by direct OHE), maybe using some features to describe the continet where each city belongs, then some other features to describe the country/region, etc.\nNote that since you didn't provide any specific detail about this task, I've used only geographical data on my example, but you could use some other variables related to each city, like the mean temprature, the population, the area, etc, depending on the task you are trying to address here.\nAnother approach could be replacing the city name with its coordinates (latitude and longitude). Again, this may be helpful depending on the task for your model.\nHope this helps\n", "Beside the models, you could also decrease the number of the features (cities) by grouping them in geographical regions. Another option is grouping them by population size.\nAnother option is grouping them by their frequency by using quantile bins. Target encoding might be another option for you.\nFeature engineering in many cases involves a lot of manual work, unfortunately you cannot always have everything sorted out automatically.\n", "There are already great responses here.\nOther technique I would use is cluster those cities into groups using K-means clustering with some of the features specific to cities in your dataset.\nBy this way you could use the cluster number in place of the actual city. This could reduce the number of levels quite a bit.\n" ]

[ 2, 1, 1, 0 ]

[]

[]

[ "forecasting", "python", "xgboost" ]

[ "Initially I downloaded two en packages using following statements in anaconda prompt.\npython -m spacy download en_core_web_lg\npython -m spacy download en_core_web_sm\n\nBut, I kept on getting linkage error and finally running below command helped me to establish link and solved error.\npython -m spacy download en\n\nAlso make sure you to restart your runtime if working with Jupyter.\n-PS : If you get linkage error try giving admin previlages.\n", "The answer to your misunderstanding is a Unix concept, softlinks which we could say that in Windows are similar to shortcuts. Let's explain this.\nWhen you spacy download en, spaCy tries to find the best small model that matches your spaCy distribution. The small model that I am talking about defaults to en_core_web_sm which can be found in different variations which correspond to the different spaCy versions (for example spacy, spacy-nightly have en_core_web_sm of different sizes). \nWhen spaCy finds the best model for you, it downloads it and then links the name en to the package it downloaded, e.g. en_core_web_sm. That basically means that whenever you refer to en you will be referring to en_core_web_sm. In other words, en after linking is not a \"real\" package, is just a name for en_core_web_sm.\nHowever, it doesn't work the other way. You can't refer directly to en_core_web_sm because your system doesn't know you have it installed. When you did spacy download en you basically did a pip install. So pip knows that you have a package named en installed for your python distribution, but knows nothing about the package en_core_web_sm. This package is just replacing package en when you import it, which means that package en is just a softlink to en_core_web_sm.\nOf course, you can directly download en_core_web_sm, using the command: python -m spacy download en_core_web_sm, or you can even link the name en to other models as well. For example, you could do python -m spacy download en_core_web_lg and then python -m spacy link en_core_web_lg en. That would make \nen a name for en_core_web_lg, which is a large spaCy model for the English language.\nHope it is clear now :) \n", "The below worked for me :\nimport en_core_web_sm\n\nnlp = en_core_web_sm.load()\n\n", "For those who are still facing problems even after installing it as administrator from Anaconda prompt, here's a quick fix:\n\nGot to the path where it is downloaded. For e.g.\nC:\\Users\\name\\AppData\\Local\\Continuum\\anaconda3\\Lib\\site-packages\\en_core_web_sm\\en_core_web_sm-2.2.0\n\n\nCopy the path.\n\nPaste it in:\nnlp = spacy.load(r'C:\\Users\\name\\AppData\\Local\\Continuum\\anaconda3\\Lib\\site-packages\\en_core_web_sm\\en_core_web_sm-2.2.0')\n\n\nWorks like a charm :)\n\n\nPS: Check for spacy version\n", "Using the Spacy language model in Colab requires only the following two steps:\n\nDownload the model (change the name according to the size of the model)\n\n!python -m spacy download en_core_web_lg \n\n\nRestart the colab runtime!\nPerform shortcut key： Ctrl + M + .\n\nTest\nimport spacy\nnlp = spacy.load(\"en_core_web_lg\")\n\nsuccessful!!!\n", "Try this method as this worked like a charm to me:\nIn your Anaconda Prompt, run the command:\n!python -m spacy download en\n\nAfter running the above command, you should be able to execute the below in your jupyter notebook:\nspacy.load('en_core_web_sm')\n\n", "First of all, install spacy using the following command for jupyter notebook\npip install -U spacy\nThen write the following code:\nimport en_core_web_sm\nnlp = en_core_web_sm.load()\n\n", "I am running Jupyter Notebook on Windows.\nFinally, its a version issue, Need to execute below commands in conda cmd prompt( open as admin)\n\npip install spacy==2.3.5\n\npython -m spacy download en_core_web_sm\n\npython -m spacy download en\n\n\nfrom chatterbot import ChatBot\nimport spacy\nimport en_core_web_sm\nnlp = en_core_web_sm.load()\nChatBot(\"hello\")\n\nOutput -\n\n", "Don't run !python -m spacy download en_core_web_lg from inside jupyter.\nDo this instead:\nimport spacy.cli\nspacy.cli.download(\"en_core_web_lg\")\n\nYou may need to restart the kernel before running the above two commands for it to work.\n", "import spacy\n\nnlp = spacy.load('/opt/anaconda3/envs/NLPENV/lib/python3.7/site-packages/en_core_web_sm/en_core_web_sm-2.3.1')\n\nTry giving the absolute path of the package with the version as shown in the image.\nIt works perfectly fine.\n", "a simple solution for this which I saw on spacy.io\nfrom spacy.lang.en import English\nnlp=English()\n\nhttps://course.spacy.io/en/chapter1\n", "As for Windows based Anaconda,\n\nOpen Anaconda Prompt\n\nActivate your environment. Ex: active myspacyenv\n\npip install https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-2.2.0/en_core_web_sm-2.2.0.tar.gz\n\npython -m spacy download en_core_web_sm\n\nOpen Jupyter Notebook ex: active myspacyenv and then jupyter notebook on Anaconda Promt\n\n\n\nimport spacy spacy.load('en_core_web_sm')\n\nand it will run peacefully!\n", "Steps to load up modules based on different versions of spacy\ndownload the best-matching version of a specific model for your spaCy installation\npython -m spacy download en_core_web_sm\npip install .tar.gz archive from path or URL\npip install /Users/you/en_core_web_sm-2.2.0.tar.gz\n\nor\npip install https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-2.2.0/en_core_web_sm-2.2.0.tar.gz\n\nAdd to your requirements file or environment yaml file. Theres range of version that one spacy version is comptable with you can view more under https://github.com/explosion/spacy-models/releases\nif your not sure running below code\nnlp = spacy.load('en_core_web_sm') \n\nwill give off a warning telling what version model will be compatible with your installed spacy verion\nenironment.yml example\nname: root\nchannels:\n - defaults\n - conda-forge\n - anaconda\ndependencies:\n - python=3.8.3\n - pip\n - spacy=2.3.2\n - scikit-learn=0.23.2\n - pip:\n - https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-2.3.1/en_core_web_sm-2.3.1.tar.gz#egg=en_core_web_sm\n\n", "Open Anaconda Navigator. Click on any IDE. Run the code: \n!pip install -U spacy download en_core_web_sm\n!pip install -U spacy download en_core_web_sm\n\nIt will work. If you are open IDE directly close it and follow this procedure once.\n", "Loading the module using the different syntax worked for me.\nimport en_core_web_sm\nnlp = en_core_web_sm.load()\n\n", "Anaconda Users\n\nIf you're using a conda virtual environment, be sure that its the same version of Python as that in your base environment. To verify this, run python --version in each environment. If not the same, create a new virtual environment with that version of Python (Ex. conda create --name myenv python=x.x.x).\nActivate the virtual environment (conda activate myenv)\nconda install -c conda-forge spacy\npython -m spacy download en_core_web_sm\n\nI just ran into this issue, and the above worked for me. This addresses the issue of the download occurring in an area that is not accessible to your current virtual environment.\nYou should then be able to run the following:\nimport spacy\nnlp = spacy.load(\"en_core_web_sm\")\n\n", "Open command prompt or terminal and execute the below code:\npip3 install https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-2.2.0/en_core_web_sm-2.2.0.tar.gz\n\nExecute the below chunk in your Jupiter notebook.\nimport spacy\nnlp = spacy.load('en_core_web_sm')\nHope the above code works for all:)\n", "I had also same issue as I couldnt load module using '''spacy.load()'''\nYou can follow below steps to solve this on windows:\n\ndownload using !python -m spacy download en_core_web_sm\nimport en_core_web_sm as import en_core_web_sm\nload using en_core_web_sm.load() to some variable\n\nComplete code will be:\npython -m spacy download en_core_web_sm\n\nimport en_core_web_sm\n\nnlp = en_core_web_sm.load()\n\n", "This works with colab:\n!python -m spacy download en\nimport en_core_web_sm\nnlp = en_core_web_sm.load()\n\nOr for the medium:\nimport en_core_web_md\nnlp = en_core_web_md.load()\n\n", "Instead of any of the above, this solved my error.\nconda install -c conda-forge spacy-model-en_core_web_sm\nIf you are an anaconda user, this is the solution.\n", "I'm running PyCharm on MacOS and while none of the above answers completely worked for me, they did provide enough clues and I was finally able to everything working. I am connecting to an ec2 instance and have configured PyCharm such that I can edit on my Mac and it automatically updates the files on my ec2 instance. Thus, the problem was on the ec2 side where it was not finding Spacy even though I installed it several different times and ways. If I ran my python script from the command line, everything worked fine. However, from within PyCharm, it was initially not finding Spacy and the models. I eventually fixed the \"finding\" spacy issue using the above recommendation of adding a \"requirements.txt\" file. But the models were still not recognized.\nMy solution: download the models manually and place them in the file system on the ec2 instance and explicitly point to them when loaded. I downloaded the files from here:\nhttps://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.0.0/en_core_web_sm-3.0.0.tar.gz\nhttps://github.com/explosion/spacy-models/releases/download/en_core_web_lg-3.0.0/en_core_web_lg-3.0.0.tar.gz\nAfter downloading, I dropped moved them to my ec2 instance, decompressed and untared them in my filesystem, e.g. /path_to_models/en_core_web_lg-3.0.0/\nI then load a model using the explicit path and it worked from within PyCharm (note the path used goes all the way to en_core_web_lg-3.0.0; you will get an error if you do not use the folder with the config.cfg file):\nnlpObject = spacy.load('/path_to_models/en_core_web_lg-3.0.0/en_core_web_lg/en_core_web_lg-3.0.0')\n\n", "Check installed version of spacy\npip show spacy\nYou will get something like this:\nName: spacy\nVersion: 3.1.3\nSummary: Industrial-strength Natural Language Processing (NLP) in Python\nInstall the relevant version of the model using:\n!pip install -U https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.0.0/en_core_web_sm-3.0.0.tar.gz\n", "I tried all the above answers but could not succeed. Below worked for me :\n(Specific to WINDOWS os)\n\nRun anaconda command prompt with admin privilege(Important)\nThen run below commands:\n\n pip install -U --user spacy \n python -m spacy download en\n\n\nTry below command for verification:\n\nimport spacy\nspacy.load('en')\n\n\nIt might work for others versions as well:\n\n\n", "If you have already downloaded spacy and the language model (E.g., en_core_web_sm or en_core_web_md), then you can follow these steps:\n\nOpen Anaconda prompt as admin\n\nThen type : python -m spacy link [package name or path] [shortcut]\nFor E.g., python -m spacy link /Users/you/model en\n\n\nThis will create a symlink to the your language model. Now you can load the model using spacy.load(\"en\") in your notebooks or scripts\n", "This is what I did:\n\nWent to the virtual environment where I was working on Anaconda Prompt / Command Line\n\nRan this: python -m spacy download en_core_web_sm\n\n\nAnd was done\n", "TRY THIS :-\n!python -m spacy download en_core_web_md\n", "Even I faced similar issue. How I resolved it\n\nstart anaconda prompt in admin mode.\ninstalled both\npython -m spacy download en\nand\npython -m spacy download en_core_web_sm\nafter above steps only I started jupyter notebook where I am accessing this package.\nNow I can access both\nimport spacy\nnlp = spacy.load('en_core_web_sm')\nor\nnlp = spacy.load('en')\nBoth are working for me.\n\n", "I faced a similar issue. I installed spacy and en_core_web_sm from a specific conda environment. However, I got two(02) differents issues as following:\n[Errno 2] No such file or directory: '....\\en_core_web_sm\\en_core_web_sm-2.3.1\\vocab\\lexemes.bin'\nor\nOSError: [E050] Can't find model 'en_core_web_sm'.... It doesn't seem to be a shortcut link, a Python package or a valid path to a data directory.\nI did the following:\n\nOpen Command Prompt as Administrator\nGo to c:>\nActivate my Conda environment (If you work in a specific conda environment):\n\nc:\\>activate <conda environment name>\n\n\n(conda environment name)c:\\>python -m spacy download en\nReturn to Jupyter Notebook and you can load the language library:\n\nnlp = en_core_web_sm.load()\n\nFor me, it works :)\n", "Download en_core_web_sm tar file\nOpen terminal from anaconda or open anaconda evn.\nRun this:\npip3 install /Users/yourpath/Downloads/en_core_web_sm-3.1.0.tar.gz;\n\nor\npip install /Users/yourpath/Downloads/en_core_web_sm-3.1.0.tar.gz;\n\nRestart jupyter, it will work.\n", "Run this in os console:\npython -m spacy download en\npython -m spacy link en_core_web_sm en_core_web_sm\n\nThen run this in python console or on your python IDE:\nimport spacy\nspacy.load('en_core_web_sm')\n\n", "This worked for me:\nconda install -c conda-forge spacy-model-en_core_web_sm\n", "Best is to follow the official spacy docs for installation (https://spacy.io/usage):\nFirst uninstall your current spacy version\npip uninstall spacy\n\nThen install pacy correctly\npip install -U pip setuptools wheel\npip install -U spacy\npython -m spacy download en_core_web_sm\n\n" ]

[ 156, 83, 40, 17, 15, 11, 5, 5, 5, 4, 4, 3, 3, 2, 2, 2, 2, 2, 2, 2, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[]

[]

[ "nlp", "python", "python_3.x", "spacy" ]

[ "It's quite unfortunate that you couldn't setup AIOHTTP properly because this is one of the most efficient way to do asynchronous requests in Python.\nSetup is not that hard:\nimport asyncio\nimport aiohttp\nfrom time import perf_counter\n\n\ndef urls(n_reqs: int):\n for _ in range(n_reqs):\n yield \"https://python.org\"\n\nasync def get(session: aiohttp.ClientSession, url: str):\n async with session.get(url) as response:\n _ = await response.text()\n \nasync def main(n_reqs: int):\n async with aiohttp.ClientSession() as session:\n await asyncio.gather(\n *[get(session, url) for url in urls(n_reqs)]\n )\n\n\nif __name__ == \"__main__\":\n n_reqs = 10_000\n \n start = perf_counter()\n asyncio.run(main(n_reqs))\n end = perf_counter()\n \n print(f\"{n_reqs / (end - start)} req/s\")\n\nYou basically need to create a single ClientSession which you then reuse to send the get requests. The requests are made concurrently with to asyncio.gather(). You could also use the newer asyncio.TaskGroup:\nasync def main(n_reqs: int):\n async with aiohttp.ClientSession() as session:\n async with asyncio.TaskGroup() as group:\n for url in urls(n_reqs):\n group.create_task(get(session, url))\n\nThis easily achieves 500+ requests per seconds on my 7+ years old bi-core computer. Contrary to what other answers suggested, this solution does not require to spawn thousands of threads, which are expensive.\nYou may improve the speed even more my using a custom connector in order to allow more concurrent connections (default is 100) in a single session:\nasync def main(n_reqs: int):\n let connector = aiohttp.TCPConnector(limit=0)\n async with aiohttp.ClientSession(connector=connector) as session:\n ...\n\n\n", "Hope this helps, this question asked What is the fastest way to send 10000 http requests\nI observed 15000 requests in 10s, using wireshark to trap on localhost and saved packets to CSV, only counted packets that had GET in them.\nFILE: a.py\nfrom treq import get\nfrom twisted.internet import reactor\n\ndef done(response):\n if response.code == 200:\n get(\"http://localhost:3000\").addCallback(done)\n\nget(\"http://localhost:3000\").addCallback(done)\n\nreactor.callLater(10, reactor.stop)\nreactor.run()\n\nRun test like this:\npip3 install treq\npython3 a.py # code from above\n\nSetup test website like this, mine was on port 3000\nmkdir myapp\ncd myapp\nnpm init\nnpm install express\nnode app.js\n\nFILE: app.js\nconst express = require('express')\nconst app = express()\nconst port = 3000\n\napp.get('/', (req, res) => {\n res.send('Hello World!')\n})\n\napp.listen(port, () => {\n console.log(`Example app listening on port ${port}`)\n})\n\nOUTPUT\ngrep GET wireshark.csv | head\n\"5\",\"0.000418\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"13\",\"0.002334\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"17\",\"0.003236\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"21\",\"0.004018\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"25\",\"0.004803\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\ngrep GET wireshark.csv | tail\n\"62145\",\"9.994184\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"62149\",\"9.995102\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"62153\",\"9.995860\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"62157\",\"9.996616\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\"62161\",\"9.997307\",\"::1\",\"::1\",\"HTTP\",\"139\",\"GET / HTTP/1.1 \"\n\n\n", "This works, getting around 250+ requests a second.\nThis solution does work on Windows 10. You may have to pip install for concurrent and requests.\nimport time\nimport requests\nimport concurrent.futures\n\nstart = int(time.time()) # get time before the requests are sent\n\nurls = [] # input URLs/IPs array\nresponses = [] # output content of each request as string in an array\n\n# create an list of 5000 sites to test with\nfor y in range(5000):urls.append(\"https://example.com\")\n\ndef send(url):responses.append(requests.get(url).content)\n\nwith concurrent.futures.ThreadPoolExecutor(max_workers=10000) as executor:\n futures = []\n for url in urls:futures.append(executor.submit(send, url))\n \nend = int(time.time()) # get time after stuff finishes\nprint(str(round(len(urls)/(end - start),0))+\"/sec\") # get average requests per second\n\nOutput:\n286.0/sec\nNote: If your code requires something extremely time dependent, replace the middle part with this:\nwith concurrent.futures.ThreadPoolExecutor(max_workers=10000) as executor:\n futures = []\n for url in urls:\n futures.append(executor.submit(send, url))\n for future in concurrent.futures.as_completed(futures):\n responses.append(future.result())\n\nThis is a modified version of what this site showed in an example.\nThe secret sauce is the max_workers=10000. Otherwise, it would average about 80/sec. Although, when setting it to beyond 1000, there wasn't any boost in speed.\n" ]

[ 1, 0, 0 ]

[]

[]

[ "concurrency", "http", "https", "python", "python_3.x" ]

[ "@DavidRanieri's comment resolved my problem\n" ]

[ 0 ]

[]

[]

[ "c", "ctypes", "python", "scipy", "scipy.ndimage" ]

[ "Your error means that lower + upper is an odd number, which when you divide by 2 results in something like 3.5, 8.5, etc., which is an invalid index for a list.\nTo solve this, use floored division (rounding down) with the double slash // operator\nif target == nums[(lower+upper)//2]:\n\n", "Once you've fixed the integer division you'll have a problem when you try to make the recursive call because you're not providing enough parameters.\nYou may find this helpful:\nrecursions = 0\ncomparisons = 0\n\ndef binary_search(lst, t):\n def _binary_search(lst, lo, hi, t):\n global recursions, comparisons\n recursions += 1\n if hi >= lo:\n mid = (hi + lo) // 2\n comparisons += 1\n if lst[mid] == t:\n return mid\n\n comparisons += 1\n\n if lst[mid] > t:\n return _binary_search(lst, lo, mid - 1, t)\n else:\n return _binary_search(lst, mid + 1, hi, t)\n else:\n return -1\n return _binary_search(lst, 0, len(lst)-1, t)\n \nindex = binary_search([1, 2, 3, 4, 5, 6, 7, 8, 9], 2)\n\nprint(f'{index=} {recursions=} {comparisons=}')\n\nOutput:\nindex=1 recursions=2 comparisons=3\n\n" ]

[ 1, 0 ]

[]

[]

[ "binary_search", "python", "recursion" ]

[ "You need to reference the database with the correct path: https://help.pythonanywhere.com/pages/NoSuchFileOrDirectory/\n", "You should give the absolute path with one extra '/'\ndb = SQL(\"sqlite:////home/routsiddharth/mysite/finance.db\")\n\n" ]

[ 2, 0 ]

[]

[]

[ "flask", "python", "pythonanywhere" ]

[ "thanks for helping me out\nits done!\ndef treeview_click(self, event):\n iid = self.cuadro_blanco_facturas.focus() # id \n name = self.cuadro_blanco_facturas.item(iid)[\"values\"][2] #column name\n espacio = \" \"\n file = os.startfile(f\"facturas\\\\{iid}{espacio}{nombre}.pdf\")\n\n" ]

[ 0 ]

[]

[]

[ "python", "tkinter", "treeview" ]

[ "The main purpose of virtual environments or venv is to manage settings and dependencies of a particular project regardless of other Python projects. virtualenv tool comes bundled with PyCharm, so the user doesn't need to install it. It is always found in the project directory named venv which should be a unique folder design to fulfil a specific purpose.\nNote: No external file(s) should be added to the venv folder.\nThis clearly indicates that your structure is not appropriate. I will recommend you visit pycharm project structure to read more about configuration of virtual environments. You should restructure your project properly. It might feel like a pain on the neck but I bet it worth it.\nAttention:\nAll the external files you added to venv should rather be in your samik folder which is your project main folder.\n" ]

[ 1 ]

[]

[]

[ "contextmanager", "filenotfounderror", "pycharm", "python", "streamlit" ]

[ "The program will exit after executing code. Add below statements to keep program running:\ntime.sleep(300) #300 seconds i.e. 5 minutes\n\n# close the browser window\ndriver.quit()\n\n", "This will fix your problem\nfrom selenium.webdriver.chrome.options import Options\n\n# Stop Selenium from closing browser automatically\nchrome_options = Options()\nchrome_options.add_experimental_option(\"detach\", True)\n\n# Chrome driver to run chrome\ndriver = webdriver.Chrome(options=chrome_options)\n\n" ]

[ 1, 0 ]

[]

[]

[ "python", "selenium", "selenium_webdriver" ]

[ "most applications such as tablular text extractors simply scrape the visible surface as plain text and actually hyperlinks are often stored elsewhere in the pdf which is NOT a WTSIWYG word processor file.\nSo, if you're lucky you can extract the co-ordinates (without their page allocation like this)\nC:\\Users\\lz02\\Downloads>type \"7 - 20 November 2022 (003).pdf\" |findstr /i \"(http\"\n<</Subtype/Link/Rect[ 69.75 299.75 280.63 313.18] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(http://www.bbc.co.uk/complaints/complaint/) >>/StructParent 5>>\n<</Subtype/Link/Rect[ 219.37 120.85 402.47 133.06] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(http://www.bbc.co.uk/complaints/handle-complaint/) >>/StructParent 1>>\n<</Subtype/Link/Rect[ 146.23 108.64 329.33 120.85] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(http://www.bbc.co.uk/complaints/handle-complaint/) >>/StructParent 2>>\n<</Subtype/Link/Rect[ 412.48 108.64 525.55 120.85] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(https://www.ofcom.org.uk/tv-radio-and-on-demand/broadcast-codes/broadcast-code) >>/StructParent 3>>\n<</Subtype/Link/Rect[ 69.75 96.434 95.085 108.64] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(https://www.ofcom.org.uk/tv-radio-and-on-demand/broadcast-codes/broadcast-code) >>/StructParent 4>>\n<</Subtype/Link/Rect[ 69.75 683.75 317.08 697.18] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(http://www.bbc.co.uk/complaints/comp-reports/ecu/) >>/StructParent 7>>\n<</Subtype/Link/Rect[ 463.35 604.46 500.24 617.89] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(https://www.bbc.co.uk/contact/ecu/reporting-scotland-bbc-one-scotland-20-december-2021) >>/StructParent 8>>\n<</Subtype/Link/Rect[ 463.35 577.11 500.24 590.54] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(https://www.bbc.co.uk/contact/ecu/book-of-the-week-preventable-radio-4-19-april-2022) >>/StructParent 9>>\n<</Subtype/Link/Rect[ 463.35 522.4 521.41 535.83] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(https://www.bbc.co.uk/contact/ecu/the-one-show-bbc-one-6-october-2022) >>/StructParent 10>>\n<</Subtype/Link/Rect[ 463.35 495.04 518.04 508.47] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(https://www.bbc.co.uk/contact/ecu/news-6pm-bbc-one-22-september-2022) >>/StructParent 11>>\n<</Subtype/Link/Rect[ 463.35 469.04 518.04 482.47] /BS<</W 0>>/F 4/A<</Type/Action/S/URI/URI(https://www.bbc.co.uk/contact/ecu/news-1030am-bbc-news-channel-20-september-2022) >>/StructParent 12>>\n\nNOTE, the random order, to find which page they belong to you need to traceback to their /StructParent ##\n", "Yes, it's possible. Camelot, by default, only extracts the text from PDF files, but it also provides options to extract additional information, such as the position and size of text blocks, as well as the coordinates of the lines and curves that define the table cells. With this information, it is possible to identify the table cells that contain hyperlinks, and to extract the text and the hyperlink destination for each of these cells.\nHere is an example of how this can be done using Camelot:\nimport camelot\n\n# Load the PDF file\npdf = camelot.read_pdf(\"example.pdf\")\n\n# Extract the tables, including their coordinates and text blocks\ntables = pdf.extract(flavor=\"lattice\", tables=None, spreadsheets=None,\n str_columns_map=None, columns=None, suppress_stdout=False)\n\n# Iterate over the tables\nfor table in tables:\n # Iterate over the rows in the table\n for row in table.data:\n # Iterate over the cells in the row\n for cell in row:\n # If the cell contains a hyperlink, extract the text and the hyperlink destination\n if cell.text.startswith(\"http\"):\n text = cell.text\n hyperlink = cell.bbox[0]\n print(text, hyperlink)\n\n" ]

[ 0, 0 ]

[]

[]

[ "pdf", "python", "python_camelot" ]

[ "It sounds like your model is overfitting to your training data. This means that it is performing well on the data it was trained on, but not generalizing well to new data.\nOne common cause of overfitting is using a model that is too complex for the amount of training data you have. In your case, using an LSTM with 50 units may be too complex for your data, especially if you don't have a lot of training examples.\nTo combat overfitting, you can try using regularization techniques such as adding dropout layers to your model. You can also try using a simpler model with fewer parameters, or using more training data.\nAdditionally, it's a good idea to monitor the performance of your model on a validation set during training, to ensure that it is not overfitting. You can do this by passing a validation set to the fit method of your model, and setting the validation_split argument to a value between 0 and 1. This will cause the model to evaluate its performance on the validation set after each epoch of training.\n" ]

[ 0 ]

[]

[]

[ "categorical", "categories", "keras", "lstm", "python" ]

[ "PostgresToGCSOperator inherits from BaseSQLToGCSOperator(https://airflow.apache.org/docs/apache-airflow-providers-google/stable/_api/airflow/providers/google/cloud/transfers/sql_to_gcs/index.html)\nAccording to source code, approx_max_file_size_bytes=1900000000. So if you split your table into 10 parts (or workers lets say) the maximum size of this chunk should be maximum 1.9 gigabyte. In case this chunk is bigger, the previous chunk will be replaced with the new one as you did not specify to create \"chunks of your chunk\" by PostgresToGCSOperator.\nYou can to it by adding placeholder {} in the filename and the Operator will handle it.\ndef create_operator_write_append_init(self, worker=10):\n worker_var = dict()\n with TaskGroup(group_id=self.task_id_init) as tg1:\n for i in range(worker):\n worker_var[f'worker_{i}'] = PostgresToGCSOperator(\n task_id = f'worker_{i}',\n postgres_conn_id = self.conn_id,\n sql = 'extract_init.sql',\n bucket = self.bucket,\n filename = f'{self.filename_init}_{i}_part_{{}}.{self.export_format}', \n export_format = self.export_format, # the export format is json\n gzip = True,\n params = {\n 'worker': i\n }\n )\n return tg1\n\n" ]

[ 0 ]

[]

[]

[ "airflow", "airflow_2.x", "google_cloud_composer", "python" ]

[]

[]

[ "The time module is part of Python's standard library. It's installed along with the rest of Python, and you don't need to (nor can you!) install it with pip.\n\nI can import time in the Python Console\n\nYes, because it's already installed.\n\nbut not in my actual code\n\nI don't believe you. Show us the exact error message you get when you try.\n" ]

[ -1 ]

[ "module", "pip", "python", "time" ]

[ "Per https://github.com/tensorflow/tensorflow/issues/30409#issuecomment-508962873 you have to:\n\nReplace arr.write(j, t) with arr = arr.write(j, t)\nThe issue is that tf.function executes as a graph. In eager mode the array will be updated (as a convenience), but you're really meant to use the return value to chain operations: https://www.tensorflow.org/api_docs/python/tf/TensorArray#returns_6\n\n", "instead of array.gather(), try using array.stack(), it'll return a tensor from the TensorArray\n" ]

[ 1, 0 ]

[]

[]

[ "python", "tensorflow", "tensorflow2.0" ]

[ "if asOfDate column has a date type or of it is a string with date in the format yyyy-mm-dd you can filter the dateframe for the rows you want to write\ndf[df.asOfDate == df.asOfDate.max()].to_csv('output.csv', mode='a', index=True, header=False, columns=header)\n\n" ]

[ 1 ]

[]

[]

[ "csv", "dataframe", "date", "pandas", "python" ]

[ "In order to plot the time lag on the y-axis, you will need to convert the time lag from a timedelta object to a numerical value that can be used in the plot. One way to do this is to convert the time lag to seconds using the total_seconds method, and then plot the resulting values on the y-axis.\nHere is an example of how you can do this:\nimport pandas as pd\nimport matplotlib.pyplot as plt\n\n# Create a dataframe with the time and time lag data\ndata = [ ['2020-04-18 14:00:00', '0 days 03:00:00'],\n ['2020-04-19 02:00:00', '1 days 13:00:00'],\n ['2020-04-28 14:00:00', '1 days 17:00:00'],\n ['2020-04-29 20:00:00', '2 days 09:00:00'],\n ['2020-04-30 19:00:00', '2 days 11:00:00'],\n]\ndf = pd.DataFrame(data, columns=['time', 'time_lag'])\n\n# Convert the time and time lag columns to datetime and timedelta objects\ndf['time'] = pd.to_datetime(df['time'])\ndf['time_lag'] = pd.to_timedelta(df['time_lag'])\n\n# Convert the time lag to seconds\ndf['time_lag_seconds'] = df['time_lag'].dt.total_seconds()\n\n# Create a scatter plot with the time on the x-axis and the time lag in seconds on the y-axis\nplt.scatter(df['time'], df['time_lag_seconds'])\nplt.show()\n\n" ]

[ 1 ]

[]

[]

[ "matplotlib", "python", "timedelta" ]

[ "You can use the scatter_ function to calculate the product of the tensors in each group.\nsamples = torch.Tensor([\n [0.1, 0.1], #-> group / class 1\n [0.2, 0.2], #-> group / class 2\n [0.4, 0.4], #-> group / class 2\n [0.0, 0.0] #-> group / class 0\n])\n\nlabels = torch.LongTensor([1,2,2,0])\n\nlabel_size = 3\nsample_dim = samples.size(1)\n\nindex = labels.unsqueeze(1).repeat((1, sample_dim))\n\nres = torch.ones(label_size, sample_dim, dtype=samples.dtype)\nres.scatter_(0, index, samples, reduce='multiply')\n\nres:\ntensor([[0.0000, 0.0000],\n [0.1000, 0.1000],\n [0.0800, 0.0800]])\n\n" ]

[ 0 ]

[]

[]

[ "python", "pytorch", "tensor" ]

[ "From the example input and output, I see that you want to do an element wise comparison, and store the values per columns. From your code you understand the 1D variant of this problem, so the question seems to be how to do it in 2D.\nSolution 1\nIn order to achieve this, we have to make the 2D problem, a 1D problem, so you can do what you already did. If for example the columns would become rows, then you can redo your zip strategy for every row.\nIn otherwords, if we can turn:\na = np.array(\n [[2, 5, 6, 3, 7],\n [6, 2, 9, 2, 3],\n [2, 1, 4, 5, 7]]\n)\n\ninto:\narray([[2, 6, 2],\n [5, 2, 1],\n [6, 9, 4],\n [3, 2, 5],\n [7, 3, 7]])\n\nwe can iterate over a and b, at the same time, and get our 1D version of the problem. Swapping the x and y axis of the matrix like this, is called transposing, and is very common, the operation for numpy is a.T, (docs ndarry.T).\nNow we use your code onces for the outer loop of iterating over all the rows (after transposing, all the rows actually hold the column values). After which we use the code on those values, because every row is a 1D numpy array.\nresult = []\n\n# Outer loop, to go over the columns of `a` and `b` at the same time.\nfor row_a, row_b in zip(a.T, b.T):\n\n result_col = []\n # Inner loop to compare a whole column element wise.\n for col_a, col_b in zip(row_a, row_b):\n result_col.append('bigger' if col_a > col_b else 'smaller')\n result.append(result_col)\n\nNote: I use a ternary operator to assign smaller and bigger.\nSolution 2\nAs indicated before you are only looking at 2 values that are in the same position for both arrays, this is called an elementwise comparison. Since we are only interested in the values that are at the exact same position, and we know the output shape of our result array (input 1000x2000, output will be 2000x1000), we can also iterate over all the elements using their index.\nNow some quick handy shortcuts,\n\na.shape holds the dimensions of the array, therefore a.shape will be (1000, 2000).\nusing [::-1] will reverse the order, similar to reverse()\nCombining a.shape[::-1] will hold (2000, 1000), our expected output shape.\nnp.ndindex provides indexing, based on the number of dimensions provided.\nAn *, performs tuple unpacking, so using it like np.ndindex(*a.shape), is equivalent to np.ndindex(1000, 2000).\n\nTherefore we can use their index (from np.ndindex) and turn the x and y around to write the result to the correct location in the output array:\na = np.random.randint(0, 255, (1000, 2000))\nb = np.random.randint(0, 255, (1000, 2000))\nresult = np.zeros(a.shape[::-1], dtype=object)\n\nfor rows, columns in np.ndindex(*a.shape):\n result[columns, rows] = 'bigger' if a[rows, columns] > b[rows, columns] else 'smaller'\n\nprint(result)\n\nThis will lead to the same result. Similarly we could also first transpose the a and b array, drop the [::-1] in the result array, and swap the assignment result[columns, rows] back to result[rows, columns].\nEdit\n\nThinking about it a bit longer, you are only interested in doing a comparison between two array of the same shape (dimension). For this numpy already has a good solution, np.where(cond, <true>, <false>).\nSo the entire problem can be reduced to:\nanswer = np.where(a > b, 'bigger', 'smaller').T\n\nNote the .T to transpose the solution, such that the answer has the columns in the rows.\n" ]

[ 0 ]

[]

[]

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

[ "Assuming a dataframe like the following:\ndf = pd.DataFrame({\"messages\": [\n {\"id\": 21263, \"from\": \"user3984\", \"text\": \"jajajajaja\"},\n {\"id\": 21264, \"from\": \"user837\", \"text\": ['Not sure', {'type': 'hashtag', 'text': '#confused'}]}, \n {\"id\": 21265, \"from\": \"user3984\", \"text\": ['O ', {'type': 'mention', 'text': '@user87324'}, ' really?']}\n]})\n\nFirst, expand the messages dictionaries into separate id, from and text columns.\n expanded = pd.concat([df.drop(\"messages\", axis=1), pd.json_normalize(df[\"messages\"])], axis=1)\n\nThen explode the dataframe to have a row for each entry in text:\nexploded = expanded.explode(\"text\")\n\nThen expand the dictionaries that are in some of the entries, converting them to lists of text:\ndef convert_dict(entry):\n if type(entry) is dict:\n return [f\"{k}: {v}\" for k, v in entry.items()]\n else:\n return entry\n\nexploded[\"text\"] = exploded[\"text\"].apply(convert_dict)\n\nFinally, explode again to separate the converted dicts to separate rows.\nfinal = exploded.explode(\"text\")\n\nThe resulting output should look like this\n id from text\n0 21263 user3984 jajajajaja\n1 21264 user837 Not sure\n1 21264 user837 type: hashtag\n1 21264 user837 text: #confused\n2 21265 user3984 O \n2 21265 user3984 type: mention\n2 21265 user3984 text: @user87324\n2 21265 user3984 really?\n\n", "Just to share some ideas to flatten your list,\ndef flatlist(srclist):\n flatlist=[]\n if srclist: #check if srclist is not None\n for item in srclist:\n if(type(item) == str): #check if item is type of string\n flatlist.append(item)\n if(type(item) == dict): #check if item is type of dict\n for x in item:\n flatlist.append(x + ' ' + item[x]) #combine key and value\n return flatlist\n\nfor item in df[\"messages\"]:\n tg_text = item.get(\"text\", \"None\")\n flat_list = flatlist(tg_text) # get the flattened list\n for tg in flat_list: # loop through the list and get the data you want\n tg_id = item.get(\"id\", \"None\")\n tg_from = item.get(\"from\", \"None\")\n \n print(tg_id, tg_from, tg)\n\n" ]

[ 1, 0 ]

[]

[]

[ "for_loop", "json", "list", "nested", "python" ]

[ "Alternative is to use a list comprehension:\nTF_notinrange = [tf for tf in TFlist \n if not any(istart <= tf[0] <= tf[1] <= iend \n for istart, iend in enhancerlist)]\nprint(TF_notinrange)\n>>> TF_notinrange\n\nExplanation\nTake ranges of TFlist which are not contained in any ranges of enhancerlist\n", "You can use chained comparisons along with the less-common for-else block where the else clause triggers only if the for loop was not broken out of prematurely to achieve this:\nnon_overlapping = []\n\nfor tf_a, tf_b in TFlist:\n for enhancer_a, enhancer_b in enhancerlist:\n if enhancer_a <= tf_a < tf_b <= enhancer_b:\n break\n else:\n non_overlapping.append([tf_a, tf_b])\n\nNote that this assumes that all pairs are already sorted and that no pair comprises a range of length zero (e.g., (2, 2)).\n" ]

[ 1, 0 ]

[]

[]

[ "python" ]

[ "Using Sniffer class from csv.\nFrom docs:\n\nhas_header(sample)\n\n\nAnalyze the sample text (presumed to be in CSV format) and return True if the first row appears to be a series of column headers. Inspecting each column, one of two key criteria will be considered to estimate if the sample contains a header:\n\n\n the second through n-th rows contain numeric values\n\n\n\n the second through n-th rows contain strings where at >least one value’s length differs from that of the putative header of that column.\n\n\n\nTwenty rows after the first row are sampled; if more than half of columns + rows meet the criteria, True is returned.\n\n\nNote\n\n\nThis method is a rough heuristic and may produce both false positives and negatives.\n\nimport csv \n\nwith open('mailing_list.csv') as csv_file:\n hdr = csv.Sniffer().has_header(csv_file.read())\n csv_file.seek(0)\n r = csv.reader(csv_file)\n mailing_list = []\n if hdr:\n next(r)\n for row in r:\n mailing_list.append(row)\n\nmailing_list \nOut[11]: \n[['307919e9-d6f0-4ecf-9bef-c1320db8941a',\n 'afarrimond0',\n '[email protected]',\n 'opt-out'],\n ['8743d75d-c62a-4bae-8990-3390fefbe5c7',\n 'tdelicate1',\n '[email protected]',\n 'opt-out'],\n ['68a32cae-847a-47c5-a77c-0d14ccf11e70',\n 'edelahuntyk',\n '[email protected]',\n 'OPT-OUT'],\n ['a50bd76f-bc4d-4141-9b5d-3bfb9cb4c65d',\n 'tdelicate10',\n '[email protected]',\n 'active'],\n\n ...\n\n ['3fd55224-dbff-4c89-baec-629a3442d8f7',\n 'smcgonnelli',\n '[email protected]',\n 'opt-out'],\n ['2ac17a63-a64b-42fc-8780-02c5549f23a7',\n 'mmayoralj',\n '[email protected]',\n 'unsubscribed'],\n []]\n\n\n" ]

[ 1 ]

[]

[]

[ "append", "csv", "python", "readlines" ]

[ "It looks like you are running the pip install command with sudo privileges, which is generally not recommended. The warning message is suggesting that you try running the command with the --user flag instead, which will install the packages locally for the current user instead of system-wide with root privileges.\nYou can try running the following command instead:\npython3 -m pip install --user Cython==0.29.4 boto==2.49.0 boto3==1.18.50 numpy==1.19.5 pandas==1.3.2 pyarrow==5.0.0\n\nThis should install the packages locally for the current user, without requiring sudo privileges. You may also need to add the ~/.local/bin directory to your PATH environment variable so that you can run the installed packages from the command line.\n" ]

[ 0 ]

[]

[]

[ "amazon_emr", "python" ]

[ "IDLE stands for Integrated Development and Learning Environment. It is a built-in development environment for writing and running Python code.\n" ]

[ 0 ]

[]

[]

[ "python", "python_idle" ]

[ "If need filter DataFrame by year, brand and months you can avoid groupby and use DataFrame.loc with mask - if scalar compare by Series.eq, if multiple values use Series.isin:\ndef sumMonthValues (year, brand):\n \n months = 10 if year == 2022 else 12\n \n mask = (df['ID_MES'].isin(range(1, months+1)) &\n df['ANIO'].eq(year) & \n df['MARCA'].isin(list(brand)))\n \n return df.loc[mask, \"UNI_VEH\"].sum()\n\nyear = 2020\n#one element tuple - added ,\nbrand = ('Acura', )\n\nchuck = sumMonthValues (year, brand)\nprint (chuck)\n110\n\n", "So i got arround it: Storing the values given from the sum of sales given year and brand per month.\nimport pandas as pd\nimport matplotlib.pyplot as plt\n\ndf = pd.read_excel(\"ventasHondaMexico2020-2019.xlsx\")\n\ndef sumMonthValues (year, brand):\n \n sMonthSum = []\n months = 10 if year == 2022 else 12\n nmes = 1\n mes = [nmes]\n while nmes <= months:\n mask = (df['ID_MES'].isin(mes) &\n df['ANIO'].eq(year) & \n df['MARCA'].isin(list(brand)))\n nmes = nmes +1\n mes = [nmes]\n sumMes = df.loc[mask, \"UNI_VEH\"].sum()\n sMonthSum += [sumMes]\n return sMonthSum\n \n\nyear = 2020\n#one element tuple - added ,\nbrand = ('Acura', )\nconteo = 1 \nchuck = sumMonthValues (year, brand)\nprint (chuck)\n\n" ]

[ 0, 0 ]

[]

[]

[ "pandas", "python" ]

[ "To store the value of the trip_id column based on the first most popular trip_id of each unique route_id, direction_id combination, you can use the idxmax method on the groupby object to get the index of the first most popular trip_id, and then use this index to access the value of the trip_id column.\nHere is an example of how you can do this:\nimport pandas as pd\n\n# Create the dataframe\ndata = [[1000, 1, 1], [1000, 1, 1], [1000, 1, 1], [1000, 1, 2], [1000, 1, 2], [1000, 1, 2], [2000, 0, 1], [2000, 0, 1], [2000, 1, 2], [2000, 0, 2], [2000, 1, 2]]\ndf = pd.DataFrame(data, columns=['route_id', 'direction_id', 'trip_id'])\n\n# Group the dataframe by route_id and direction_id\nt_groups = df.groupby(['route_id','direction_id'])\n\n# Get the index of the first most popular trip_id for each group\nidx = t_groups['trip_id'].apply(lambda x: x.value_counts().index[0])\n\n# Access the value of the trip_id column at the index for each group\ntrip_ids = t_groups['trip_id'].apply(lambda x: x.loc[idx])\n\n# Print the values of the trip_id column for each group\nprint(trip_ids)\n\n", "This is what you are looking for.\ndf = df.groupby(['route_id', 'direction_id']).first().reset_index()\n\nThe reset_index() just moves your indices into columns looking exactly like the output you want.\n" ]

[ 0, 0 ]

[]

[]

[ "group_by", "pandas", "python" ]

[ "You could use '\\r' in some way. Carriage return is a control character or mechanism used to reset a device's position to the beginning of a line of text. e.g.\nimport time\n\nfor i in range(100):\n time.sleep(1)\n print(i, end=\"\\r\");\n\nExample including controls\nimport time\nimport keyboard\n\n# params\nsize = 20\nposition = 0\n\n# game loop\nwhile True:\n # draw\n print(\"\", end=\"\\r\")\n for i in range(size):\n print(\"-\" if (position != i) else \"o\", end=\"\")\n # move\n if keyboard.is_pressed(\"left\"): \n position = position - 1\n if position < 0:\n position = 0\n elif keyboard.is_pressed(\"right\"): \n position = position + 1\n if position >= size:\n position = size\n # delay \n time.sleep(0.1)\n \n\n", "when printing the output use the end attribute for print statements and set it as \"\\r\" to get back the output position to starting of the line.\n" ]

[ 0, 0 ]

[]

[]

[ "input", "output", "python", "string" ]

[ "You don't have to directly operate with the time. Just transform the string and do the calculations you need, for example, like this:\nn = '3:51:52' # As an example. Substitute with your actual strings.\nhours, minutes, seconds = map(int, n.split(':'))\n# Effectively, the above line does something akin to this: \n# hours, minutes, seconds = n.split(':')\n# hours = int(hours)\n# minutes = int(minutes)\n# seconds = int(seconds)\n\ntime_in_seconds = hours * 360 + minutes * 60 + seconds\ntime_in_minutes = hours * 60 + minutes + seconds / 60\n\n" ]

[ 0 ]

[]

[]

[ "pandas", "python", "time" ]

[ "Setting the crs like:\ngdf.crs = {'init' :'epsg:32618'}\n\ndoes not transform your data, it only sets the CRS (it basically says: \"my data is represented in this CRS\"). In most cases, the CRS is already set while reading the data with geopandas.read_file (if your file has CRS information). So you only need the above when your data has no CRS information yet.\nIf you actually want to convert the coordinates to a different CRS, you can use the to_crs method:\ngdf_new = gdf.to_crs(epsg=32618)\n\nSee https://geopandas.readthedocs.io/en/latest/projections.html\n", "super late, but it's:\ntmp.set_crs(...)\n\n" ]

[ 4, 0 ]

[]

[]

[ "geopandas", "python" ]

[ "Maybe you could just make a string representation of your class that uses numpy.np.array2string()\nimport numpy as np\n\nclass Map:\n def __init__(self, n):\n self.n = n\n self.create_grid(n)\n\n def create_grid(self, n):\n self.arr = np.random.randint(0, 2, size=(n, n))\n \n def __str__(self):\n return np.array2string(self.arr)\n\nprint(Map(5))\n\nThis will print something like:\n[[1 1 1 0 1]\n [1 0 1 0 1]\n [1 0 0 1 1]\n [0 1 1 0 0]\n [1 0 0 0 1]]\n\nOf course, you can wrap this in another method if you like:\n ...\n def print_grid(self):\n print(self)\n ...\nMap(5).print_grid()\n\n" ]

[ 1 ]

[]

[]

[ "2d", "arrays", "list", "numpy", "python" ]

[ "You have to filter the imported object to select only ones contained in the wanted schema with:\ndef include_name(name, type_, parent_names):\n if type_ == \"schema\":\n return name == SCHEMA_WANTED\n return True\n\nand then:\ncontext.configure(\n connection=connection,\n target_metadata=get_metadata(),\n process_revision_directives=process_revision_directives,\n include_name=include_name,\n **current_app.extensions['migrate'].configure_args,\n)\n\nMore info: https://alembic.sqlalchemy.org/en/latest/api/runtime.html#alembic.runtime.environment.EnvironmentContext.configure.params.include_name\nPersonally I don't like to change the alembic env.py so I give those parameters as the Flask-Migrate initialization :\nalembic_ctx_kwargs = {\n 'include_name': include_name,\n 'include_schemas': True,\n 'version_table_schema': SCHEMA_WANTED,\n}\nMigrate(app, db, **alembic_ctx_kwargs)\n\n" ]

[ 0 ]

[]

[]

[ "alembic", "flask_migrate", "flask_script", "python", "sqlalchemy" ]

[ "This time since it was needed to make a depth of 3. I simply concatenated same image 3 times with the help of np.concatenate function and it worked\n" ]

[ 0 ]

[]

[]

[ "audio_processing", "deep_learning", "python", "signal_processing" ]

[ "As indicated in the comments, your input data is very obscure. This provides a lot of trouble for us, because we don't know what we can assume or not. For my solution I will assume at least the following, based on the example you provide:\n\nIn the dictionary there is an entry containing the iid and Invnum as keys in the same level.\nThe Invnum key is the only key, which has multiple values, or in otherwords is iterable (besides df), and on iteration it must hold the last dictionary. In otherwords, after the Invnum value (e.g. B11), you can only get the last dict with the other fields as keys (@type, execId, CId, AId, df, type), if they exists.\nIf there is a df value, it will hold a list.\n\n# This is a place holder dictionary, so we can create entries that have the same pattern.\nentry = {'@type': '', 'execId': '', 'CId': '', 'AId': '', 'df': '', 'type': ''}\n\n# This will hold all the (properly) format entries for the df.\nitems = []\n\ndef flatten(data):\n if isinstance(data, dict):\n match data:\n # We are searching for a level that contains both an `iid` and `Invnum` key.\n case {'iid': id, 'Invnum': remainder}:\n for each in remainder:\n entry_row = dict(**entry, iid=id, Invnum=each)\n entry_row.update(remainder[each])\n items.append(entry_row)\n case _:\n for key, value in data.items():\n flatten(value)\n\n# We flatten the data, such that the `items` variable will hold consistent entries\nflatten(data)\n\n# Transfer to pandas dataframe, and reorder the values for easy comparison.\ndf = pd.DataFrame(items)\ndf = df[['iid', 'Invnum', '@type', 'execId', 'CId', 'AId', 'df', 'type']]\nprint(df.to_string(index=False))\n\nOutput:\niid Invnum @type execId CId AId df type\n B1 B11 /test_data 42 42 BAZ \n B1 B12 /test_data 8 123 \n B2 B21 /test_data 215 253 [] F\n B2 B22 /test_data 10 [] F\n\nNote:\n\nAll entries have been turned into strings, since I am using '' for empty values.\nI heavily rely on the above made assumptions, in case they are incorrect, the answer will not match your expectation.\nI am using Structural pattern matching, which is introduced in python 3.10.\n\n" ]

[ 1 ]

[]

[]

[ "dataframe", "dictionary", "pandas", "python", "python_3.x" ]

[ "Use pd.concat([sr1, sr2], axis=1) if neither one of join and merge work.\n" ]

[ 3 ]

[]

[]

[ "dataframe", "pandas", "python", "yahoo_api", "yahoo_finance" ]

[ "This should work because it seems that e is a list[dict] in your case:\ntry:\n pass # Your code...\nexcept api.error.PermissionError as e:\n print(e.args[0][0]['detail'])\n\nIf you added manually the [] then maybe you will have to remove one of the [0].\n" ]

[ 0 ]

[]

[]

[ "exception", "python" ]

[ "There is some discussion regarding this here. I had the same issue but using cuda 11.1 resolved it for me.\nThis is the exact pip command\npip install torch==1.8.0+cu111 torchvision==0.9.0+cu111 torchaudio==0.8.0 -f https://download.pytorch.org/whl/torch_stable.html\n\n", "In my case it actually had nothing do with the PyTorch/CUDA/cuDNN version. PyTorch initializes cuDNN lazily whenever a convolution is executed for the first time. However, in my case there was not enough GPU memory left to initialize cuDNN because PyTorch itself already held the entire memory in its internal cache. One can release the cache manually with \"torch.cuda.empty_cache()\" right before the first convolution that is executed. A cleaner solution is to force cuDNN initialization at the beginning by doing a mock convolution:\ndef force_cudnn_initialization():\n s = 32\n dev = torch.device('cuda')\n torch.nn.functional.conv2d(torch.zeros(s, s, s, s, device=dev), torch.zeros(s, s, s, s, device=dev))\n\nCalling the above function at the very beginning of the program solved the problem for me.\n", "I am also using Cuda 10.2. I had the exact same error when upgrading torch and torchvision to the latest version (torch-1.8.0 and torchvision-0.9.0). Which version are you using?\nI guess this is not the best solution but by downgrading to torch-1.7.1 and torchvision-0.8.2 it works just fine.\n", "In my cases this error occurred when trying to estimate loss.\nI used a mixed bce-dice loss.\nIt turned out that my output was linear instead of sigmoid.\nI then used the sigmoid predictions as of bellow and worked fine.\noutput = torch.nn.Sigmoid()(output)\nloss = criterion1(output, target)\n\n", "I had the same issue when I was training yolov7 with a chess dataset. By reducing batch size from 8 to 4, the issue was solved.\n", "In my case, I had an array indexing operation but the index was out of bounds. CUDA did not tell me that. I was using inference on a neural network. So I moved to CPU instead of the GPU. The logs were much more informative after that. For debugging if you see this error, switch to CPU first and you will know what to do.\n" ]

[ 19, 12, 7, 1, 1, 0 ]

[]

[]

[ "gpu", "python", "pytorch" ]

[ "You can also use assign like this:\ndf.assign(Pop=df.loc[[0], 'Pop'])\n\nOutput:\n Date Pop\n0 2020-01-01 90.0\n1 2021-01-01 NaN\n2 2022-01-01 NaN\n\nNote: assign works with nice column headers, if your headers have spaces or special characters you will need to use a different method.\n", "To replace the values from the second row onwards in your DataFrame using the pandas method pipe, you can do the following:\nimport pandas as pd\nimport numpy as np\n\ndf = pd.DataFrame(\n {\n \"Date\": [\"2020-01-01\", \"2021-01-01\", \"2022-01-01\"],\n \"Pop\": [90, 70, 60],\n }\n)\n\n# Use the `pipe` method to apply a function to your DataFrame\ndf = df.pipe(lambda x: x.iloc[1:]).replace(np.nan)\n\nprint(df)\n\nThis will replace the values from the second row onwards in your DataFrame with NaN. The resulting DataFrame will look like this:\n Date Pop\n0 2021-01-01 NaN\n1 2022-01-01 NaN\n\nNote that this will modify your original DataFrame in place, so if you want to keep the original DataFrame, you should make a copy of it before applying the pipe method. You can do this by using the copy method, like this:\nimport pandas as pd\nimport numpy as np\n\ndf = pd.DataFrame(\n {\n \"Date\": [\"2020-01-01\", \"2021-01-01\", \"2022-01-01\"],\n \"Pop\": [90, 70, 60],\n }\n)\n\n# Make a copy of the original DataFrame\ndf_copy = df.copy()\n\n# Use the `pipe` method to apply a function to your DataFrame\ndf_copy = df_copy.pipe(lambda x: x.iloc[1:]).replace(np.nan)\n\nprint(df_copy)\n\nThis will create a copy of your DataFrame and then modify the copy, so the original DataFrame will remain unchanged. The resulting DataFrame will look the same as in the previous example\n" ]

[ 3, 2 ]

[]

[]

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

[ "From Itertools recipes:\ndef nth(iterable, n, default=None):\n \"Returns the nth item or a default value\"\n return next(islice(iterable, n, None), default)\n\nApplied to your example:\nimport itertools\n\ndef imparesgen():\n n = 0\n while n<200:\n n=n+2\n yield n\n\ngen = imparesgen()\n\nprint(next(itertools.islice(gen, 3, None)))\n\n", "You could iterate over the generator three times. Since zip stops at the shortest input, you'll only get three iterations provided gen yields at least three elements.\ngen = imparesgen()\nfor _, item in zip(range(3), gen):\n pass\n\n# Now, item is the third element\nprint(item)\n\nAlternatively, use the next() function to get the next element of the generator:\ngen = imparesgen()\nnext(gen)\nnext(gen)\nitem = next(gen)\nprint(item)\n\nOr, if you don't want to write out those next lines multiple times:\ngen = imparesgen()\nfor _ in range(3):\n item = next(gen)\n\nprint(item)\n\n", "Is this what you're looking for?:\nfrom itertools import islice\n\ndef imparesgen():\n n = 0\n while n<200: \n n=n+2\n yield n\n\ngen = imparesgen()\n\nthird = list(islice(gen, 2, 3))[0] # -> 6\n\n" ]

[ 2, 0, 0 ]

[ "Since a generator acts like an iterator, you can call next on it :\nthird = next(next(next(gen)))\n\nI don't think you can go much faster than that in pure-Python. But I think that without benchmarking the code, the speedup won't be noticed.\n" ]

[ -2 ]

[ "generator", "python" ]

[ "The thing about many to many field is not an actual field. If you take a look at the generated schema you wouldnt find the field as a column in either of the table. What happens in the back is django creates a ItemCatagory table.\nclass ItemCatagory(models.Model):\n item = modes.ForegnKeyField(Item, related_name=\"catagories\", on_delete... )\n catagory = models.ForegnKeyField(Item, related_name=\"items\", on_delete... )\n\ncatagory.items will give u the ItemCatagory RelatedObject and if u do catagoty.items.all() it will give u QuerySet[ItemCatagory]. so the default model.save() method woont add those rship. so u would have to overide save method on either of the models.\n", "When you call Item.objects.create(), you need to omit the categories from the args. Then afterwards you can call set() to set the categories.\nitem = Item.objects.create()\nitem.categories.set(categories)\n\nIf you only have one category to add, you can call add() instead:\nitem = Item.objects.create()\nitem.categories.add(category)\n\nNote: both add() and set() save the update to the database, so you don’t need to call item.save() afterwards.\nSource: Many-to-Many relations\n" ]

[ 1, 0 ]

[ "See, ManyToManyField can't make reverse relationship with related model as python is interpreted language, so it can't read model class of previous one. Instead, you can do one thing ...\n# models.py\n\nclass Item(models.Model):\n item_name = models.CharField(max_length=255, default=\"\")\n\n def __str__(self):\n return self.item_name\n\n\nclass Category(models.Model):\n category_name = models.CharField(max_length=255, default=\"\")\n items = models.ManyToManyField(Item)\n\n def __str__(self):\n return self.category_name\n\nAfter that, you can list down your requirements in views.py file.\n# views.py\n\ndef get_items_by_categories(request):\n \n # Here, you will receive a set of items ...\n \n get_categories = Category.objects.all()\n\n # Filter out items with respect to categories ...\n\n get_items_list = [{\"category\": each.category_name, \"items\": each.items} for each in get_categories]\n\n return render(request, \"categories.html\", {\"data\": get_items_list})\n\nIterate your items with categories in categories.html file.\n{% for each in data %}\n {% for content in each %}\n {{content.category}}\n {% for item in content.items.all %}\n {{item.item_name}}\n {% endfor %}\n {% endfor %}\n{% endfor %}\n\nI hope this solution will help you ..\nThank You !\n" ]

[ -1 ]

[ "django", "django_models", "many_to_many", "python" ]

[ "bro.\ni read your code but i can't get some points as below.\n\n\nif dice1 == '1' and dice2 == '1': #both die = 1 (banked & running = 0)\n player1Score = 0\n\n\n\nthe code above makes player1Score as zero.\nand there's no adding score code for player2Score.\nFor this reason, the while loop doesn't stop.\nplz, check it first.\n", "Boolean logic works like this.\nYou want both scores to be less then 100 while the game runs which also means one of the scores should not be 100 or higher.\nboth scores less then 100\nint(player1Score) < 100 and int(player2Score) < 100\n\none of the socres is not 100 or higher\n!int(player1Score) >= 100 or !int(player1Score) >= 100\n\nThe reason why the game didn't stop when you changed the \"or\" to \"and\" is because player1Score and player2Score are being evaluated after both player1() and player2() are called. Even if player1 scored over 100, the game wouldn't stop and the turn would go over to player2, goes into the next loop evaluation and finally stop the loop.\nTo fix this, change the while loop of both player1() and player() to this.\nwhile player1Score < 100:\n\nAnd also add evaluation after calculating scores.\nelse: #normal dice roll - gamble or bank\n choice = input(\"Gamble (G) or Bank (B): \")\n choice = choice.upper() #incase lowercase letter given\n \n # rTotal += int(dice1) + int(dice2)\n # rewriting the same code is a bad practice. \n # In this case, you can just write the same code outside the if-else logic since if-else doesn't effect the code.\n\n rTotal += int(dice1) + int(dice2)\n\n if choice == 'G': #if Gamble chosen - reroll dice & add to running\n # added evaluation to stop the game\n if player1Score+rTotal >= 100:\n break\n if choice == 'B': #used to save score.\n player1Score += rTotal\n print(f\"\\nPlayer 1 banked! Total: {player2Score}\")\n break\n\nHope this helps.\n", "What I did is inside the function put an if statement (outside of the while) to return a value from the function using return like this:\ndef exampleFunction1():\n #scoring logic here\n if bank > finalScore: #Score to reach to win game\n print(f\"Score is over {finalScore}!\")\n return 0\n\nThen inside the while loop, attach the functions to a variable so that the return value could be identified. Use the return value to break the while loop:\nwhile True: #Breaks as soon as either player gets above 100 score\n x = exampleFunction1()\n if x == 0:\n break\n x = exampleFunction2()\n if x == 0:\n break\n\n(Sidenote: I rewrote my whole program to take one gameLogic() function and apply this to the different players, but I still used the while loop stated second)\n" ]

[ 0, 0, 0 ]

[]

[]

[ "boolean_logic", "logic", "python" ]

[ "You can use matplotlib to set the axis scaling, as the difference is really subtle in your data:\nimport matplotlib.pyplot as plt\n\nax = plt.gca()\nax.set_ylim([df.yoy.min(numeric_only=True), df.yoy.max(numeric_only=True)])\n\nsns.lineplot(\n x = 'date',\n y = 'yoy',\n data = df,\n ax = ax\n)\n\nWith this the result should be more of a stepping function.\n\nYou can use something like the max difference to the mean times 1.01 to set the limits a little better, but this is the idea. You can set the axis ticks using ax.set_yticks(ticks=<list of ticks>) (documentation).\n" ]

[ 0 ]

[]

[]

[ "python", "seaborn" ]

[ "Uninstall your current spyder installation and reinstall it with Anaconda. Your terminal environment is fundamentally different from your overall environment. As such installing sci-kit learn on your Linux distro on terminal still doesn't give you access to the library elsewhere. Anaconda is the easiest way to install all the scientific computing packages you need.\n" ]

[ 0 ]

[]

[]

[ "installation", "python", "python_3.x", "scikit_learn" ]

[ "I found this very weird too, and couldn't find any direct way to pass the environment into the replaced process. But I didn't do a full system debugging either.\nWhat I found to work as a workaround is this:\npypath = os.environ.get('PYTHONPATH', \"\")\nargs = ['sudo', f\"PYTHONPATH={pypath}\", sys.executable] + sys.argv\nos.execvpe('sudo', args, os.environ)\n\nI.e. explicitly pass PYTHONPATH= to the new process. Note that I prefer to use os.execvpe(), but it works the same with the other exec*(), given the correct call. See this answer for a good overview of the schema.\nHowever, PATH and the rest of the environment is still it's own environment, as an initial print(os.environ) shows. But PYTHONPATH will be passed on this way.\n" ]

[ 0 ]

[ "You're passing the environment as arguments to your script instead of arguments to execlpe. Try this instead:\nargs = ['sudo', sys,executable] + sys.argv + [os.environ]\nos.execvpe('sudo', args, os.environ)\n\nIf you just want to inherit the environment you can even\nos.execvp('sudo', args)\n\n" ]

[ -1 ]

[ "python" ]

[ "You are trying to reference the product variable before it has been assigned when serializer.is_valid() is False.\nYou should move the response line inside the if statement, so that it is only returned if serializer.is_valid() is True and handle the response for invalid serializer with an http error for example.\n", "Check if the serializer.is_valid() returns True. If it doesn't, your code attempts to return a Response object, which uses the product variable. However, if your serializer.is_valid() does returns False, the Response is still created with the product variable, which has not been created yet.\nIt might be wise to write functionality that handles the situation when your serializer.is_valid() returns False. In that case, probably you don't need the product variable.\n" ]

[ 0, 0 ]

[]

[]

[ "django", "python" ]

[ "To iterate through each row in a DataFrame and count the number of NaN values in the original columns and the number of non-NaN values in the extra columns, you can do the following:\nimport pandas as pd\n\n# Define the dataframe\ndf = pd.DataFrame(\n {\n \"Heading 1\": [np.nan, np.nan, 5, 5, np.nan, np.nan],\n \"Heading 2\": [34, np.nan, np.nan, 7, np.nan, np.nan],\n \"Unnamed: 1\": [24, 44, np.nan, np.nan, 13, np.nan],\n \"Unnamed: 2\": [np.nan, np.nan, np.nan, np.nan, 77, 18]\n }\n)\n\n# Define the original columns and the extra columns\noriginal_cols = [\"Heading 1\", \"Heading 2\"]\nextra_cols = [\"Unnamed: 1\", \"Unnamed: 2\"]\n\n# Create a dictionary to store the counts\ncounts = {}\n\n# Iterate through each row in the DataFrame\nfor index, row in df.iterrows():\n # Count the number of NaN values in the original columns\n original_nan_count = sum(row[col].isna() for col in original_cols)\n \n # Count the number of non-NaN values in the extra columns\n extra_non_nan_count = sum(not row[col].isna() for col in extra_cols)\n \n # Add the counts to the dictionary\n counts[index] = [original_nan_count, extra_non_nan_count]\n\n# Print the dictionary of counts\nprint(counts)\n\nThis will iterate through each row in the DataFrame, count the number of NaN values in the original columns and the number of non-NaN values in the extra columns, and store the counts in a dictionary where the keys are the row indexes and the values are lists containing the counts. The resulting dictionary will look like this:\n{0: [1, 1],\n 1: [2, 1],\n 2: [1, 0],\n 3: [0, 0],\n 4: [2, 2],\n 5: [2, 1]}\n\n", "As an alternative:\ndf['Count'] = df[['Heading 1', 'Heading 2']].apply(lambda x: sum(x.isnull()), axis=1)\ndf['Count2'] = df[['Unnamed: 1', 'Unnamed: 2']].apply(lambda x: sum(x.notnull()), axis=1)\ndf['total']=df[['Count','Count2']].values.tolist()\n\noutput=dict(zip(df.index, df.total))\n'''\n{0: [1, 1], 1: [2, 1], 2: [1, 0], 3: [0, 0], 4: [2, 2], 5: [2, 1]}\n'''\n\nor\nmask=list(map(list, zip(df[['Heading 1', 'Heading 2']].isnull().sum(axis=1), df[['Unnamed: 1', 'Unnamed: 2']].notnull().sum(axis=1))))\noutput=dict(zip(df.index,mask))\n#{0: [1, 1], 1: [2, 1], 2: [1, 0], 3: [0, 0], 4: [2, 2], 5: [2, 1]}\n\n", "The .isna() (in Cyzanfar's answer) raises an exception for me:\nAttributeError: 'numpy.float64' object has no attribute 'isna'\n\nYou could instead try the following:\ncounts = {}\n\nfor index, row in df.iterrows():\n# Count the number of NaN values in the original columns\nnum_nan_orig = np.sum(np.isnan(row[['Heading 1', 'Heading 2']]))\n\n# Count the number of non-NaN values in the extra columns\nnum_non_nan_extra = np.sum(~np.isnan(row[['Unnamed: 1', 'Unnamed: 2']]))\n\ncounts[index] = [num_nan_orig, num_non_nan_extra]\nprint(counts) \n\nOutputs the following:\n# {0: [1, 1], 1: [2, 1], 2: [1, 0], 3: [0, 0], 4: [2, 2], 5: [2, 1]}\n\n\nThe ~ operator (third last line in the code) is the bitwise negation operator in Python, which inverts the boolean value of its operand. In this case, it is inverts the boolean values produced by np.isnan() method, so that the non-NaN values can be counted.\n" ]

[ 0, 0, 0 ]

[]

[]

[ "data_cleaning", "data_preprocessing", "dataframe", "pandas", "python" ]