KaneAI - World’s First E2E Software Testing Agent.Book a Demo Join TestMu Conference

How to use log_input method in grail

Best Python code snippet using grail_python

app.py

Source:app.py

...24        query=APP.local_stack.query25        self.finder = FIND_TICKET(query)26        time.sleep(1)27        while not APP.local_stack.mission:28            log_input(APP.local_stack.user["user_name"],"è·åä»»å¡ï¼å¼å§æ¥è¯¢")29            #æ¥è¯¢åæ£æ¥æ¶é´30            APP.myclock.check_clock(APP.local_stack.user["user_name"])31            missions = self.finder()32            if missions:33                for mission in missions:34                    APP.local_stack.mission.append(mission)35                log_input(APP.local_stack.user["user_name"],"å®æä¸æ¬¡æ¥è¯¢ï¼åç°ç¬¦åè¦æ±çè½¦ç¥¨")36            else:37                log_input(APP.local_stack.user["user_name"],"å®æä¸æ¬¡æ¥è¯¢ï¼æªåç°ç¬¦åè¦æ±çè½¦ç¥¨")38                time.sleep(random.randrange(5, 10))39    def execute_mission(self):40        #å½æ¥è¯¢å¨è·åç¬¦åä»»å¡éæ±çç¥¨æ®ä¿¡æ¯æ¶ï¼è¿è¡ä¹°ç¥¨æä½41        mission=APP.local_stack.mission.pop()42        try:43            APP.myclock.check_clock(APP.local_stack.user["user_name"])44            res =self.buyer.buy_ticket(mission)45        except:46            self.buyer.driver.close()47            raise Exception("ä¹°ç¥¨å¤±è´¥ï¼æªç¥éè¯¯ï¼è¯·èç³»ä½è")48        return res49    @classmethod50    def start_mission(cls,user,query):51        app=cls(user,query)52        APP.local_stack.user=user53        APP.local_stack.query=query54        log_input(APP.local_stack.user["user_name"], "åå§åæµè§å¨")55        try:56            app.buyer=BUY_TICKET(user,query)57        except:58            raise Exception("æµè§å¨åå§åå¤±è´¥")59        log_input(APP.local_stack.user["user_name"], "æµè§å¨åå§åæåï¼ä¹°æå°±ä½")60        APP.local_stack.mission = []61        while True:62            try:63                app.get_mission()64            except:65                log_input(APP.local_stack.user["user_name"], "æ¥è¯¢å¤±è´¥")66                raise Exception("æ¥è¯¢å¤±è´¥")67            log_input(APP.local_stack.user["user_name"],"å¼å§ä¹°ç¥¨")68            res = app.execute_mission()69            if res == "succeeded":70                log_input(APP.local_stack.user["user_name"],"ä¹°ç¥¨æåï¼è¯·å¨30åéåè¿å¥12306appä»æ¬¾")71                log_input(APP.local_stack.user["user_name"], "ä»»å¡ç»æï¼éåºçº¿ç¨")72                break73            else:74                app.buyer.driver.close()75                log_input(APP.local_stack.user["user_name"],"ä¹°ç¥¨å¤±è´¥ï¼åæ¬¡å°è¯ä¹°ç¥¨")76                log_input(APP.local_stack.user["user_name"], "éå¯æµè§å¨")77                try:78                    app.buyer = BUY_TICKET(user, query)79                except:80                    raise Exception("æµè§å¨åå§åå¤±è´¥")81                log_input(APP.local_stack.user["user_name"], "æµè§å¨åå§åæåï¼ä¹°æå°±ä½")82                APP.local_stack.mission = []83    @classmethod84    def start(cls,users,querys):85        APP.checker.check_before_start(users,querys)86        APP.users=users87        APP.querys=querys88        if len(users)==1:89            user = users[0]90            query=querys.get(user['user_name'])91            log_input(user["user_name"], "*" * 120)92            log_input(user["user_name"], "ç¨åºå¯å¨")93            log_input(user["user_name"], "è½½å¥ç¨æ·")94            cls.start_mission(user,query)95            return  None96        else:97            thread_list=[]98            for user in APP.users:99                log_input(user["user_name"], "*"*120)100                log_input(user["user_name"], "ç¨åºå¯å¨")101                log_input(user["user_name"], "è½½å¥ç¨æ·ï¼å¼å¯çº¿ç¨")102                query=querys.get(user["user_name"])103                t=threading.Thread(target=cls.start_mission,args=(user,query))104                t.start()105                thread_list.append(t)106                time.sleep(1)107            for t in thread_list:108                t.join()...

loss_functions.py

Source:loss_functions.py

1import torch2import math3def log_poisson_loss(targets, log_input, compute_full_loss=False):4    """Computes log Poisson loss given `log_input`.5    FOLLOWS TF IMPLEMENTATION6    Gives the log-likelihood loss between the prediction and the target under the7    assumption that the target has a Poisson distribution.8    Caveat: By default, this is not the exact loss, but the loss minus a9    constant term [log(z!)]. That has no effect for optimization, but10    does not play well with relative loss comparisons. To compute an11    approximation of the log factorial term, specify12    compute_full_loss=True to enable Stirling's Approximation.13    For brevity, let `c = log(x) = log_input`, `z = targets`.  The log Poisson14    loss is15        -log(exp(-x) * (x^z) / z!)16      = -log(exp(-x) * (x^z)) + log(z!)17      ~ -log(exp(-x)) - log(x^z) [+ z * log(z) - z + 0.5 * log(2 * pi * z)]18          [ Note the second term is the Stirling's Approximation for log(z!).19            It is invariant to x and does not affect optimization, though20            important for correct relative loss comparisons. It is only21            computed when compute_full_loss == True. ]22      = x - z * log(x) [+ z * log(z) - z + 0.5 * log(2 * pi * z)]23      = exp(c) - z * c [+ z * log(z) - z + 0.5 * log(2 * pi * z)]24    Args:25    targets: A `Tensor` of the same type and shape as `log_input`.26    log_input: A `Tensor` of type `float32` or `float64`.27    compute_full_loss: whether to compute the full loss. If false, a constant28      term is dropped in favor of more efficient optimization.29    name: A name for the operation (optional).30    Returns:31    A `Tensor` of the same shape as `log_input` with the componentwise32    logistic losses.33    Raises:34    ValueError: If `log_input` and `targets` do not have the same shape.35    """36    try:37        assert (targets.size() == log_input.size())38    except ValueError:39        raise ValueError(40            "log_input and targets must have the same shape (%s vs %s)" %41            (log_input.size(), targets.size()))42    result = torch.exp(log_input) - log_input * targets43    if compute_full_loss:44        point_five = 0.545        two_pi = 2 * math.pi46        stirling_approx = (targets * torch.log(targets)) - targets + \47                          (point_five * torch.log(two_pi * targets))48        zeros = torch.zeros_like(targets, dtype=targets.dtype)49        ones = torch.ones_like(targets, dtype=targets.dtype)50        cond = (targets >= zeros) & (targets <= ones)51        result += torch.where(cond, zeros, stirling_approx)...

module.py

Source:module.py

1import cv2 as cv2import random3import tensorflow as tf4import numpy as np5from cv2.ximgproc import guidedFilter678def decomposition(input):9    # 1. Logarithm10    log_input = np.log(input + 1e-6)1112    # 2. Scale to [0,1]13    log_input = (log_input - log_input.min() + 1e-6) / (log_input.max() - log_input.min() + 1e-6)1415    # 3. Guided filter16    base = guidedFilter(log_input.astype('single'), log_input.astype('single'), radius=3, eps=0.1)1718    # 4. Decomposition19    detail = log_input - base2021    return base, detail222324def decomposition_NotLog(input):25    # 1. Logarithm26    # log_input = np.log(input + 1e-6)2728    # 2. Scale to [0,1]29    # log_input = (log_input - log_input.min() + 1e-6) / (log_input.max() - log_input.min() + 1e-6)3031    # 3. Guided filter32    base = guidedFilter(input.astype('single'), input.astype('single'), radius=3, eps=0.1)3334    # 4. Decomposition35    detail = input - base3637    return base, detail3839def randomCropHL(img, img2, width, height):40    if img.shape[0] < height or img.shape[1] < width:41        img = cv.resize(img, dsize=(width, height))42        img2 = cv.resize(img2, dsize=(width, height))43    else:44        x = random.randint(np.round((img.shape[1] - width)/2.1), np.round((img.shape[1] - width)*1.1/2.1))45        y = random.randint(np.round((img.shape[0] - height)/2.1), np.round((img.shape[0] - height)*1.1/2.1))46        img = img[y:y + height, x:x + width]47        img2 = img2[y:y + height, x:x + width]48    return img, img2495051def randomCrop(img, width, height):52    if img.shape[0] < height or img.shape[1] < width:53        img = cv.resize(img, dsize=(width, height))54    else:55        x = random.randint(0, img.shape[1] - width)56        y = random.randint(0, img.shape[0] - height)57        img = img[y:y + height, x:x + width]58    return img596061def ref_padding(tensor, pad_size=1):62    output = tf.pad(tensor, [[0, 0], [pad_size, pad_size], [pad_size, pad_size], [0, 0]], "REFLECT")
...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.