sadat2307/MSciNLI · Datasets at Hugging Face

id stringlengths 7 12	sentence1 stringlengths 5 1.44k	sentence2 stringlengths 6 2.06k	label stringclasses 4 values	domain stringclasses 5 values
train_127300	Furthermore, spiking neurons are computationally more powerful than traditional neurons [15].	practitioners can construct small networks of spiking neurons to accomplish the same task as a traditional network of greater size.	reasoning	Software and its engineering
train_127301	Although the Add Product page-design does not have any PAGE IN-PUT blocks, the Edit Product page-design includes a PAGE INPUT block that .ows the ITEMNO data into the page.	clicking the Edit button brings up the Edit Product webpage, allowing the selected product to be updated.	reasoning	Software and its engineering
train_127302	Our system can prove a program safe, but if the program is modi.ed by making some functions non-deterministically diverge, then, we may no longer be able to prove safety.	in future work, it would be valuable to explore other ways to reconcile laziness and re.nement typing.	reasoning	Software and its engineering
train_127303	On the other, their work continues to use LattE as a lattice counting tool to estimate probabilities.	it is restricted to uniform distributions or discretized distributions over .nite data domains.	reasoning	Software and its engineering
train_127304	The design of the evaluation is motivated by the aim to involve external simulation tools for optimization into the framework.	pEPPA supports the handling of calculation extensive .tness evaluation.	reasoning	Software and its engineering
train_127305	For example, most previous approaches to DSE [38,54] typically use a .xed depth-.rst, backtracking search strategy, where the next execution path would always share the longest possible pre.x with the previous execution path.	much resource may be allocated to analyze small parts of the program under test before moving on.	reasoning	Software and its engineering
train_127306	In contrast, the rule processAllAttributes illustrates the binding of the variable A to all of the attributes of Class C (indicated by the use of {} decoration).	a is bound to an ordered collection of Attributes and for a single Class with any number of attributes, only one Something is created.	reasoning	Software and its engineering
train_127307	Amongother information, that database contains a register for each commit in the CVS repository for all the projects stored in SF.net (including additional information such as the date of the commit, author, .les a.ected, etc).	we could easily measure the daily number of changes for each project using the CVSAnalY dataset.	reasoning	Software and its engineering
train_127308	The message payload may contain concrete data, but also expressions on symbolic data.	the predicate PC is the disjunction of predicates path C1 .· · ·.path Cn , where each path Ci is a conjunction of .	reasoning	Software and its engineering
train_127309	By applying the Box and Jenkins approach described in Section 2.4.1, we obtain that the spectrum of the time series of S1 contains a frequency peak at 0.2 i.e., the series has a period every 1/0.2 = 5 days and that ARIMA(2,0,1) and ARIMA(1,0,0) are good models for the time series of S1 and S2 respectively.	we could forecast the future value of X for S1 and S2, by means of these ARIMA models, and .nally, we select the service between S1 and S2 with higher probability to meet the QoS constraint.	reasoning	Software and its engineering
train_127310	cases would only make the ratio even smaller.	we The Google Chrome debugger shows the general class of Object for foo and the Baz.Bar class for bar in their 2 We did not perform any preprocess to exclude third-party or repeated .les in the provided source code bundles.	reasoning	Software and its engineering
train_127311	Although fire followed by the event name is used for Java swing components, in general we can't even count on the existence of a method that throws the event.	we employ the following strategy: we use the fire method if it exists; otherwise we patch the standardized methods for adding and removing listeners, so that we are in control of the firing of the corresponding event.	reasoning	Software and its engineering
train_127312	These CM changes will result in all tests that execute either f() or g() to be identi.ed as a.ected.	all possible e.ects of this change are taken into account, even without the explicit representation of .ow of control due to exceptions in our call graphs.	reasoning	Software and its engineering
train_127313	Furthermore, the OMG does not provide test cases that would allow checking the conformity of implementations with respect to their specifications.	developers working in companies producing COS implementations have to overcome these inaccuracies and underspecifications using their own experience.	reasoning	Software and its engineering
train_127314	Clearly, the expected slowdown ratio (= E[T (x)]/x) [13] is the same for jobs with different service time requirements and equals 1/(1-.).	pS can be considered to be a fair scheme.	reasoning	Software and its engineering
train_127315	As shown in Figures 7 and 8, C-CLR views can be toggled to hide/display code that is irrelevant to a given flag setting.	though the source specifies multiple definitions for this macro, the C-CLR view shows only the one.	reasoning	Software and its engineering
train_127316	Furthermore, if the collector is not running in parallel (which is usually the case), then during the time the program threads are stopped, only one of the processors is utilized.	it is advantageous to use on-the-.y collectors [17, 19, 18].	reasoning	Software and its engineering
train_127317	Speci.cally, the computational cost of each point can be from 100 to 100,000 core-hours, even when executed on a state-of-the-art HPC cluster.	scheduling and coordination of the execution cannot be performed manually, and a single system cannot support it.	reasoning	Software and its engineering
train_127318	For example, in Fig ure 5 we would loose the information about the changed interface of the alternative components C2 and C3.	we would not be able to generate all variants of the software family, because only one interface type would be known.	reasoning	Software and its engineering
train_127319	Our decompiler maps registers in the assembly code to global variables of type int with 32-bit values; the 16bit and 8-bit registers (which are aliases of parts of the 32-bit registers) are also mapped to the appropriate 32bit global.	the 8-bit register %al and the 16-bit register %ax, which refer to the low 8 bits and the low 16 bits of the 32-bit register %eax respectively, are both mapped to the variable eax denoting the 32-bit register %eax.	reasoning	Software and its engineering

train_127300

Furthermore, spiking neurons are computationally more powerful than traditional neurons [15].

practitioners can construct small networks of spiking neurons to accomplish the same task as a traditional network of greater size.

reasoning

Software and its engineering

train_127301

Although the Add Product page-design does not have any PAGE IN-PUT blocks, the Edit Product page-design includes a PAGE INPUT block that .ows the ITEMNO data into the page.

clicking the Edit button brings up the Edit Product webpage, allowing the selected product to be updated.

reasoning

Software and its engineering

train_127302

Our system can prove a program safe, but if the program is modi.ed by making some functions non-deterministically diverge, then, we may no longer be able to prove safety.

in future work, it would be valuable to explore other ways to reconcile laziness and re.nement typing.

reasoning

Software and its engineering

train_127303

On the other, their work continues to use LattE as a lattice counting tool to estimate probabilities.

it is restricted to uniform distributions or discretized distributions over .nite data domains.

reasoning

Software and its engineering

train_127304

The design of the evaluation is motivated by the aim to involve external simulation tools for optimization into the framework.

pEPPA supports the handling of calculation extensive .tness evaluation.

reasoning

Software and its engineering

train_127305

For example, most previous approaches to DSE [38,54] typically use a .xed depth-.rst, backtracking search strategy, where the next execution path would always share the longest possible pre.x with the previous execution path.

much resource may be allocated to analyze small parts of the program under test before moving on.

reasoning

Software and its engineering

train_127306

In contrast, the rule processAllAttributes illustrates the binding of the variable A to all of the attributes of Class C (indicated by the use of {} decoration).

a is bound to an ordered collection of Attributes and for a single Class with any number of attributes, only one Something is created.

reasoning

Software and its engineering

train_127307

Amongother information, that database contains a register for each commit in the CVS repository for all the projects stored in SF.net (including additional information such as the date of the commit, author, .les a.ected, etc).

we could easily measure the daily number of changes for each project using the CVSAnalY dataset.

reasoning

Software and its engineering

train_127308

The message payload may contain concrete data, but also expressions on symbolic data.

the predicate PC is the disjunction of predicates path C1 .· · ·.path Cn , where each path Ci is a conjunction of .

reasoning

Software and its engineering

train_127309

By applying the Box and Jenkins approach described in Section 2.4.1, we obtain that the spectrum of the time series of S1 contains a frequency peak at 0.2 i.e., the series has a period every 1/0.2 = 5 days and that ARIMA(2,0,1) and ARIMA(1,0,0) are good models for the time series of S1 and S2 respectively.

we could forecast the future value of X for S1 and S2, by means of these ARIMA models, and .nally, we select the service between S1 and S2 with higher probability to meet the QoS constraint.

reasoning

Software and its engineering

train_127310

cases would only make the ratio even smaller.

we The Google Chrome debugger shows the general class of Object for foo and the Baz.Bar class for bar in their 2 We did not perform any preprocess to exclude third-party or repeated .les in the provided source code bundles.

reasoning

Software and its engineering

train_127311

Although fire followed by the event name is used for Java swing components, in general we can't even count on the existence of a method that throws the event.

we employ the following strategy: we use the fire method if it exists; otherwise we patch the standardized methods for adding and removing listeners, so that we are in control of the firing of the corresponding event.

reasoning

Software and its engineering

train_127312

These CM changes will result in all tests that execute either f() or g() to be identi.ed as a.ected.

all possible e.ects of this change are taken into account, even without the explicit representation of .ow of control due to exceptions in our call graphs.

reasoning

Software and its engineering

train_127313

Furthermore, the OMG does not provide test cases that would allow checking the conformity of implementations with respect to their specifications.

developers working in companies producing COS implementations have to overcome these inaccuracies and underspecifications using their own experience.

reasoning

Software and its engineering

train_127314

Clearly, the expected slowdown ratio (= E[T (x)]/x) [13] is the same for jobs with different service time requirements and equals 1/(1-.).

pS can be considered to be a fair scheme.

reasoning

Software and its engineering

train_127315

As shown in Figures 7 and 8, C-CLR views can be toggled to hide/display code that is irrelevant to a given flag setting.

though the source specifies multiple definitions for this macro, the C-CLR view shows only the one.

reasoning

Software and its engineering

train_127316

Furthermore, if the collector is not running in parallel (which is usually the case), then during the time the program threads are stopped, only one of the processors is utilized.

it is advantageous to use on-the-.y collectors [17, 19, 18].

reasoning

Software and its engineering

train_127317

Speci.cally, the computational cost of each point can be from 100 to 100,000 core-hours, even when executed on a state-of-the-art HPC cluster.

scheduling and coordination of the execution cannot be performed manually, and a single system cannot support it.

reasoning

Software and its engineering

train_127318

For example, in Fig ure 5 we would loose the information about the changed interface of the alternative components C2 and C3.

we would not be able to generate all variants of the software family, because only one interface type would be known.

reasoning

Software and its engineering

train_127319

Our decompiler maps registers in the assembly code to global variables of type int with 32-bit values; the 16bit and 8-bit registers (which are aliases of parts of the 32-bit registers) are also mapped to the appropriate 32bit global.

the 8-bit register %al and the 16-bit register %ax, which refer to the low 8 bits and the low 16 bits of the 32-bit register %eax respectively, are both mapped to the variable eax denoting the 32-bit register %eax.

reasoning

Software and its engineering