How to use seq2str method in Robotframework

Best Python code snippet using robotframework

generate.py

Source:generate.py

...81    '''seq_len` indicates the length of the input number.'''82    r = [random.randint(0, 10**seq_len - 1) for _ in xrange(batch_size)]83    t = [2 * a for a in r]84    return mapstr(r, seq_len), mapstr(t, seq_len+1)85def seq2str(seq, sep=','):86    return sep.join(map(str, seq))87def getprioritySort(batch_size, seq_len, low=0, high=10, tostr=False):88    out = np.random.randint(low, high, (batch_size, seq_len))89    inp = [np.random.permutation(list(enumerate(l))).reshape(-1) for l in out]90    if tostr:91        return map(seq2str, inp), map(seq2str, out)92    else:93        return inp, out94def getinput(batch_size, seq_len, low=0, high=10, tostr=False, prepend_seqlen=False):95    '''samples a tensor of integers from `low` to `high`,96    with size (`batch_size`, `seq_len`).97    Return this tensor if not `tostr` else return a list of strings,98    with str_i being the comma-separated version of each row of tensor.'''99    tensor = np.random.randint(low, high, (batch_size, seq_len))100    if prepend_seqlen:101        tensor = np.append([[seq_len]]*batch_size, tensor, axis=1)102    if not tostr:103        return tensor104    else:105        return map(seq2str, tensor)106def getcopy(batched_in):107    return batched_in108def getrepeatCopy(batched_in, n):109    return np.tile(batched_in, n)110def getreverse(batch_in):111    return [s[::-1] for s in batch_in]112def getoddFirst(batch_in):113    return [list(s[::2]) + list(s[1::2]) for s in batch_in]114def _ngramflip(seq, n):115    return sum([list(seq[i+n-1:(i-1 if i > 0 else None):-1])116            for i in range(0, len(seq), n)],117               [])118def getbigramFlip(batch_in):119    '''`batch_in` should be a tensor fo size (batch_size, seq_len)'''120    return [_ngramflip(row, n=2) for row in batch_in]121def getbatch(batch_size, seq_len, task, start=START, end=END, sep=SEP, alphabet=10,122            nrepeat=None):123    gettask = globals()['get' + task]124    def _format(batch_in, batch_out):125        return ','.join([126            ','.join([start] + [seq2str(in_)] + [sep] + [seq2str(out)] + [end])127            for in_, out in izip(batch_in, batch_out)])128    if task in PERM:129        batch_in = getinput(batch_size, seq_len, high=alphabet)130        batch_out = gettask(batch_in)131        return _format(batch_in, batch_out)132    elif task == 'prioritySort':133        batch_in, batch_out = gettask(batch_size, seq_len, high=alphabet)134        samples = []135        for b in range(batch_size):136            sample = []137            for i in range(seq_len):138                pri = batch_in[b][2*i]139                val = batch_in[b][2*i + 1]140                sample += [start] * pri + [str(val)]141            samples.append(','.join([start] + sample + [sep, seq2str(batch_out[b]), end]))142        return ','.join(samples)143    elif task == 'repeatCopy':144        batch_in = getinput(batch_size, seq_len, high=alphabet)145        assert nrepeat is not None, 'nrepeat should not be None'146        batch_out = getrepeatCopy(batch_in, nrepeat)147        return ','.join([148            ','.join([start] * (nrepeat+1) + [seq2str(in_)] + [sep] + [seq2str(out)] + [end])149            for in_, out in izip(batch_in, batch_out)])150    elif task in ARITH:151        batch_in, batch_out = gettask(batch_size, seq_len)152        return ''.join([start + sep.join([in_, out]) + end153            for in_, out in izip(batch_in, batch_out)])154    else:155        raise Exception('unknown task')156def generate_set(batch_size, nbatch, min_len, max_len, task, alphabet=10, tofile=None,157                nrepeat_low=None, nrepeat_high=None):158    '''`min_len` and `max_len` are inclusive bounds on the input size.159    `nrepeat_low` and `nrepeat_high` are inclusive bounds on repeatCopy repeat size.'''160    batches = []161    for _ in xrange(nbatch):162        seq_len = np.random.randint(min_len, max_len+1)...

arguments.py

Source:arguments.py

1import argparse2import data_loader3import os4import subprocess5base_dir = "/projects/ml/TrRosetta/PDB30-20FEB17"6base_torch_dir = base_dir7if not os.path.exists(base_dir):8    # training on blue9    base_dir = "/gscratch/TrRosetta/"10    if os.path.exists("/scratch/torch/hhr"):11        base_torch_dir = "/scratch"12    else:13        base_torch_dir = base_dir14TRUNK_PARAMS = ['n_module', 'n_module_str', 'n_layer', 'd_msa', 'd_msa_full', 'd_pair', 'd_templ',\15                'n_head_msa', 'n_head_pair', 'n_head_templ', 'd_hidden',\16                'r_ff', 'n_resblock', 'p_drop']17SE3_PARAMS = ['num_layers', 'num_channels', 'num_degrees', 'n_heads', 'div', 18              'l0_in_features', 'l0_out_features', 'l1_in_features', 'l1_out_features',19              'num_edge_features']20# absolute path to folder containing this file21script_dir = os.path.dirname(os.path.realpath(__file__))22def get_args():23    parser = argparse.ArgumentParser()24    # i/o parameters 25    io_group = parser.add_argument_group("i/o parameters")26    io_group.add_argument('-indir', default='models/',27            help='input directory for loading model checkpoint')28    io_group.add_argument('-outdir',default='models/',29            help='output directory for dumping model checkpoints')30    io_group.add_argument('-dont_load_ckpt', default=False, action='store_true',31            help='If True, DO NOT Load pretrained checkpoint params into model?')32    io_group.add_argument('--verbose', '-v', default=False, action='store_true',33            help='If True, will print lots of stuff for debugging')34    # training parameters35    train_group = parser.add_argument_group("training parameters")36    train_group.add_argument("-model_name", default=None,37            help="model name for saving")38    train_group.add_argument('-batch_size', type=int, default=1,39            help="Batch size [1]")40    train_group.add_argument('-lr', type=float, default=2.0e-4, 41            help="Learning rate [5.0e-4]")42    train_group.add_argument('-num_epochs', type=int, default=300,43            help="Number of epochs [300]")44    train_group.add_argument("-step_lr", type=int, default=300,45            help="Parameter for Step LR scheduler [300]")46    train_group.add_argument("-port", type=int, default=12319,47            help="PORT for ddp training, should be randomized [12319]")48    train_group.add_argument("-accum", type=int, default=1,49            help="Gradient accumulation when it's > 1 [1]")50    train_group.add_argument("-f_seq2str", type=float, default=0.5, 51            help="Frequency of doing the seq-->str task vs the str-->seq task")52    train_group.add_argument("-fixbb_des_frac", type=float, default=1.01, 53            help="Of the training examples that are fixed BB examples, what fraction\54                    will be purely sequence design, rather than 'hal' task.")55    train_group.add_argument("-hal_mask_low", type=int, default=3,56            help='Smallest number of residues to mask out for a hal example')57    train_group.add_argument("-hal_mask_high", type=int, default=20, 58            help='Largest number of residues to mask out for a hal example')59    train_group.add_argument("-mode", type=str, default='str2seq', choices=['str2seq', 'seq2str', 'mixed'],60            help="training mode for model")61    train_group.add_argument('-mask_low', type=float, default=0.,62            help='Lower bound for amino acid masking fraction')63    train_group.add_argument('-mask_high', type=float, default=1.,64            help='Upper bound for amino acid masking fraction')65    train_group.add_argument('-mask_structure', default=False, action='store_true',66            help='If True, mask out portions of structure during training/val')67    train_group.add_argument('-get_grad_norm', default=False, action='store_true', 68            help='If True, spit out report of norms of gradients with respect to different loss terms')69    # data-loading parameters70    data_group = parser.add_argument_group("data loading parameters")71    data_group.add_argument('-val', default="%s/val_lists/xaa"%base_dir,72            help="Validation list [%s/val_lists/xaa]"%base_dir)73    data_group.add_argument('-maxseq', type=int, default=1000,74            help="Maximum depth of subsampled MSA [1000]")75    data_group.add_argument('-maxtoken', type=int, default=2**16,76            help="Maximum depth of subsampled MSA [2**16]")77    data_group.add_argument('-maxlat', type=int, default=50,78            help="Maximum depth of subsampled MSA [50]")79    data_group.add_argument("-lmin", type=int, default=100,80            help="Lower limit of crop size [100]")81    data_group.add_argument("-lmax", type=int, default=260,82            help="Upper limit of crop size [260]")83    data_group.add_argument("-rescut", type=float, default=3.5,84            help="Resolution cutoff [3.5]")85    data_group.add_argument("-slice", type=str, default="CONT",86            help="How to make crops [CONT (default) / DISCONT]")87    data_group.add_argument("-subsmp", type=str, default="LOG",88            help="How to subsample MSAs [UNI / LOG (default) / CONST]")89    data_group.add_argument('-mintplt', type=int, default=0,90            help="Minimum number of templates to select [0]")91    data_group.add_argument('-maxtplt', type=int, default=10,92            help="maximum number of templates to select [10]")93    data_group.add_argument('-seqid', type=float, default=150.0,94            help="maximum sequence identity cutoff for template selection [150.0]")95    # Trunk module properties96    trunk_group = parser.add_argument_group("Trunk module parameters")97    trunk_group.add_argument('-n_module', type=int, default=4,98            help="Number of iteration blocks without structure [4]")99    trunk_group.add_argument('-n_module_str', type=int, default=4,100            help="Number of iteration blocks with structure [4]")101    trunk_group.add_argument('-n_layer', type=int, default=1,102            help="Number of attention layer for each transformer encoder [1]")103    trunk_group.add_argument('-d_msa', type=int, default=384,104            help="Number of MSA features [384]")105    trunk_group.add_argument('-d_msa_full', type=int, default=64,106            help="Number of MSA features [64]")107    trunk_group.add_argument('-d_pair', type=int, default=288,108            help="Number of pair features [288]")109    trunk_group.add_argument('-d_templ', type=int, default=64,110            help="Number of templ features [64]")111    trunk_group.add_argument('-n_head_msa', type=int, default=12,112            help="Number of attention heads for MSA2MSA [12]")113    trunk_group.add_argument('-n_head_pair', type=int, default=8,114            help="Number of attention heads for Pair2Pair [8]")115    trunk_group.add_argument('-n_head_templ', type=int, default=4,116            help="Number of attention heads for template [4]")117    trunk_group.add_argument("-d_hidden", type=int, default=64,118            help="Number of hidden features for initial structure builder [64]")119    trunk_group.add_argument("-r_ff", type=int, default=4,120            help="ratio for feed-forward network in transformer encoder [4]")121    trunk_group.add_argument("-n_resblock", type=int, default=1,122            help="Number of residual blocks for MSA2Pair [1]")123    trunk_group.add_argument("-p_drop", type=float, default=0.1,124            help="Dropout ratio [0.1]")125    trunk_group.add_argument("-not_use_perf", action="store_true", default=False,126            help="Use performer or not [False]")127    # Structure module properties128    str_group = parser.add_argument_group("structure module parameters")129    str_group.add_argument('-num_layers', type=int, default=3,130            help="Number of equivariant layers in structure module block [3]")131    str_group.add_argument('-num_channels', type=int, default=32,132            help="Number of channels [32]")133    str_group.add_argument('-num_degrees', type=int, default=2,134            help="Number of degrees for SE(3) network [2]")135    str_group.add_argument('-l0_in_features', type=int, default=32,136            help="Number of type 0 input features [32]")137    str_group.add_argument('-l0_out_features', type=int, default=8,138            help="Number of type 0 output features [8]")139    str_group.add_argument('-l1_in_features', type=int, default=3,140            help="Number of type 1 input features [3]")141    str_group.add_argument('-l1_out_features', type=int, default=2,142            help="Number of type 1 output features [2]")143    str_group.add_argument('-num_edge_features', type=int, default=32,144            help="Number of edge features [32]")145    str_group.add_argument('-n_heads', type=int, default=4,146            help="Number of attention heads for SE3-Transformer [4]")147    str_group.add_argument("-div", type=int, default=4,148            help="Div parameter for SE3-Transformer [4]")149    # Loss function parameters150    loss_group = parser.add_argument_group("loss parameters")151    loss_group.add_argument('-w_dist', type=float, default=1.0,152            help="Weight on distd in loss function [1.0]")153    loss_group.add_argument('-w_str', type=float, default=1.0,154            help="Weight on strd in loss function [1.0]")155    loss_group.add_argument('-w_rms', type=float, default=1.0,156            help="Weight on rmsd in loss function [1.0]")157    loss_group.add_argument('-w_lddt', type=float, default=1.0,158            help="Weight on predicted lddt loss [1.0]")159    loss_group.add_argument('-w_blen', type=float, default=0.1,160            help="Weight on predicted blen loss [0.1]")161    loss_group.add_argument('-w_bang', type=float, default=0.1,162            help="Weight on predicted bang loss [0.1]")163    loss_group.add_argument('-w_aa', type=float, default=1.0,164            help="Weight on masked language task")165    # parse arguments166    args = parser.parse_args()167    # Setup dataloader parameters:168    loader_param = data_loader.set_data_loader_params(args)169    # make dictionary for each parameters170    trunk_param = {}171    for param in TRUNK_PARAMS:172        trunk_param[param] = getattr(args, param)173    if not args.not_use_perf:174        trunk_param["performer_N_opts"] = {"nb_features": 64, "feature_redraw_interval": 10000}175        trunk_param["performer_L_opts"] = {"nb_features": 64, "feature_redraw_interval": 10000}176    SE3_param = {}177    for param in SE3_PARAMS:178        if hasattr(args, param):179            SE3_param[param] = getattr(args, param)180    trunk_param['SE3_param'] = SE3_param 181    182    # loss function parameters 183    loss_param = {}184    seq2str_param = {}185    str2seq_param = {}186    187    # take in seq2str loss params from cmd line 188    for param in ['w_dist', 'w_str', 'w_rms', 'w_lddt', 'w_blen', 'w_bang', 'w_aa']:189        seq2str_param[param] = getattr(args, param)190    191    # take care of str2seq params here192    #str2seq_param = {193    #            'w_aa'  : 1.0,194    #            'w_str' : 0.5,195    #            'w_bang': 0.05, 196    #            'w_blen': 0.05, 197    #            'w_dist': 0.05,198    #            'w_lddt': 0.05,199    #            'w_rms' : 0.05,200    #        }201    202    # TODO: Make single set of params203    mixed_param = {204                'w_aa'  : args.w_aa,205                'w_str' : args.w_str,206                'w_bang': args.w_bang,207                'w_blen': args.w_blen,208                'w_dist': args.w_dist,209                'w_lddt': args.w_lddt,210                'w_rms' : 0.05, # IGNORE211            }212    # put both into loss param 213    #loss_param['str2seq'] = str2seq_param214    #loss_param['seq2str'] = seq2str_param215    loss_param['str2seq'] = mixed_param216    loss_param['seq2str'] = mixed_param217    # add git hash of current commit218    args.commit = subprocess.check_output(f'git --git-dir {script_dir}/../../.git rev-parse HEAD',219                                          shell=True).decode().strip()220    with open(os.path.join(args.outdir, 'args.txt'), 'w') as fp:221        fp.write('Arguments from training session\n')222        for key,val in args.__dict__.items():223            fp.write(': '.join([str(key), str(val)])+'\n')...

reserver.py

Source:reserver.py

...70            contentType = "text/plain;charset=utf-8"7172        self.sendResponseWithOutput(response, contentType, out)7374    def _seq2str(self, seq):75        """76        converts sequence of strings or tuples to string to be returned77        to AJAX call78        """79        out = u''80        for toplevel in seq:81            if toplevel is not None:82                if type(toplevel) in types.StringTypes:83                    toplevel = (toplevel, )84                out += u', '.join(85                    ["'%s'" % x.replace("'", "\\'") for x in toplevel])86                out += '\n'87        return out8889    def _dict2str(self, d):90        """91        converts dict to string to be returned to AJAX call92        """93        return u''.join([u"%s: '%s'\n" % (k, v) for k, v in d.items()])9495    def do_re(self, *args, **params):96        """97        return self._dict2str(params) + '\n' + self._seq2str(params.items())98        """99        txt = unicode(params['txt'], 'utf-8')100        regex = unicode(params['regex'], 'utf-8')101        method = params['method']102        options = re.UNICODE103104        optionList = (105             ('dotall', re.DOTALL),106             ('ignorecase', re.IGNORECASE),107             ('multiline', re.MULTILINE),108             ('verbose', re.VERBOSE))109110        for s, bitflag in optionList:111            if params.get(s, '') == 'true':112                options |= bitflag113114        try:115            cre = re.compile(regex, options)116        except Exception, e:117            out = u'ERROR: %s' % e118        else:119            results = getattr(cre, method)(txt)120            121            if results:122                # list result123                if type(results) == types.ListType:124                    out = self._seq2str(results)125                # match object126                else:127                    out = u'[TEXT MATCHES]\n'128                    ng = self._dict2str(results.groupdict())129                    if ng:130                        out += u'### NAMED GROUPS ###\n%s' % ng131                    g = self._seq2str(results.groups())132                    if g:133                        out += u'### GROUPS ###\n%s' % g134            else:135                out = u'[NO MATCH OR NOTHING FOUND]'136        return out137138    def do_POST(self):139        """140        test a re141        """142        length = int(self.headers.getheader('content-length'))        143        qs = self.rfile.read(length)144        params = dict(cgi.parse_qsl(qs, keep_blank_values=1))145        out = self.do_re(**params)
...

distance.py

Source:distance.py

1"""2Distance module3Find distance between sequences4Written by Marshall Beddoe <mbeddoe@baselineresearch.net>5Copyright (c) 2004 Baseline Research6Licensed under the LGPL7"""8#9# Note: Gaps are denoted by the integer value 256 as to avoid '_' problems10#11import align, zlib12from Numeric import *13__all__ = [ "Distance", "Entropic", "PairwiseIdentity", "LocalAlignment" ]14class Distance:15    """Implementation of classify base class"""16    def __init__(self, sequences):17        self.sequences = sequences18        self.N = len(sequences)19        # NxN Distance matrix20        self.dmx = zeros((self.N, self.N), Float)21        for i in range(len(sequences)):22            for j in range(len(sequences)):23                self.dmx[i][j] = -124        self._go()25    def __repr__(self):26        return "%s" % self.dmx27    def __getitem__(self, i):28        return self.dmx[i]29    def __len__(self):30        return len(self.dmx)31    def _go(self):32        """Perform distance calculations"""33        pass34class Entropic(Distance):35    """Distance calculation based off compression ratios"""36    def _go(self):37        # Similarity matrix38        similar = zeros((self.N, self.N), Float)39        for i in range(self.N):40            for j in range(self.N):41                similar[i][j] = -142        #43        # Do compression ratio calculations44        #45        for i in range(self.N):46            for j in range(self.N):47                if similar[i][j] >= 0:48                    continue49                seq1 = self.sequences[i][1]50                seq2 = self.sequences[j][1]51                # Convert sequences to strings, gaps denoted by '_'52                seq1str = ""53                for x in seq1:54                    if x == 256:55                        seq1str += '_'56                    else:57                        seq1str += chr(x)58                seq2str = ""59                for x in seq2:60                    if x == 256:61                        seq2str += '_'62                    else:63                        seq2str += chr(x)64                comp1 = zlib.compress(seq1str)65                comp2 = zlib.compress(seq2str)66                if len(comp1) > len(comp2):67                    score = len(comp2) * 1.0 / len(comp1) * 1.068                else:69                    score = len(comp1) * 1.0 / len(comp2) * 1.070                similar[i][j] = similar[j][i] = score71        #72        # Distance matrix73        #74        for i in range(self.N):75            for j in range(self.N):76                self.dmx[i][j] = similar[i][i] - similar[i][j]77class PairwiseIdentity(Distance):78    """Distance through basic pairwise similarity"""79    def _go(self):80        # Similarity matrix81        similar = zeros((self.N, self.N), Float)82        for i in range(self.N):83            for j in range(self.N):84                similar[i][j] = -185        #86        # Find pairs87        #88        for i in range(self.N):89            for j in range(self.N):90                if similar[i][j] >= 0:91                    continue92                seq1 = self.sequences[i][1]93                seq2 = self.sequences[j][1]94                minlen = min(len(seq1), len(seq2))95                len1 = len2 = idents = 096                for x in range(minlen):97                    if seq1[x] != 256:98                        len1 += 1.099                        if seq1[x] == seq2[x]:100                            idents += 1.0101                    if seq2[x] != 256:102                        len2 += 1.0103                m = max(len1, len2)104                similar[i][j] = idents / m105        #106        # Distance matrix107        #108        for i in range(self.N):109            for j in range(self.N):110                self.dmx[i][j] = similar[i][i] - similar[i][j]111class LocalAlignment(Distance):112    """Distance through local alignment similarity"""113    def __init__(self, sequences, smx=None):114        self.smx = smx115        # If similarity matrix is None, make a quick identity matrix116        if self.smx == None:117            self.smx = zeros((257, 257), Float)118            for i in range(257):119                for j in range(257):120                    if i == j:121                        self.smx[i][j] = 1.0122                    else:123                        self.smx[i][j] = 0.0124        Distance.__init__(self, sequences)125    def _go(self):126        # Similarity matrix127        similar = zeros((self.N, self.N), Float)128        for i in range(self.N):129            for j in range(self.N):130                similar[i][j] = -1131        #132        # Compute similarity matrix of SW scores133        #134        for i in range(self.N):135            for j in range(self.N):136                if similar[i][j] >= 0:137                    continue138                seq1 = self.sequences[i][1]139                seq2 = self.sequences[j][1]140                (nseq1, nseq2, edits1, edits2, score, gaps) = \141                    align.SmithWaterman(seq1, seq2, self.smx, 0, 0)142                similar[i][j] = similar[j][i] = score143        #144        # Compute distance matrix of SW scores145        #146        for i in range(self.N):147            for j in range(self.N):148                if self.dmx[i][j] >= 0:149                    continue150                self.dmx[i][j] = 1 - (similar[i][j] / similar[i][i])...

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.