Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Apply "symflower fix" to a "write-test" result of a model when it errors, so model responses can possibly be fixed #229

Merged
merged 2 commits into from
Jul 2, 2024
Merged

Apply "symflower fix" to a "write-test" result of a model when it errors, so model responses can possibly be fixed #229

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fcf95c4
refactor, Let the tasks return assessments per task identifiers, so e…
ruiAzevedo19 Jun 28, 2024
eb9e3f0
Apply "symflower fix" to a "write-test" result of a model when it err…
ruiAzevedo19 Jun 28, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
282 changes: 171 additions & 111 deletions cmd/eval-dev-quality/cmd/evaluate_test.go
View file
Open in desktop

Large diffs are not rendered by default.

7 changes: 4 additions & 3 deletions evaluate/evaluate.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -126,7 +126,7 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
}

assessment, ps, err := task.Run(temporaryRepository)
assessments.Add(model, language, repositoryPath, taskIdentifier, assessment)
assessments.AddAssessmentPerTask(model, language, repositoryPath, assessment)
if err != nil {
ps = append(ps, err)
}
Expand Down Expand Up @@ -226,7 +226,7 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
}

assessment, ps, err := task.Run(temporaryRepository)
assessments.Add(model, language, repositoryPath, taskIdentifier, assessment)
assessments.AddAssessmentPerTask(model, language, repositoryPath, assessment)
problemsPerModel[modelID] = append(problemsPerModel[modelID], ps...)
if err != nil {
ctx.Log.Printf("ERROR: Model %q encountered a hard error for language %q, repository %q: %+v", modelID, languageID, repositoryPath, err)
Expand All @@ -249,7 +249,8 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
}
}
if isOnlyPlainRepositories {
totalScore = uint64(len(ctx.Languages)) * uint64(ctx.Runs)
// For every write-test task in the plain repository, each model is also executed with the `symflower fix` which results in double the total results.
totalScore = 2 * uint64(len(ctx.Languages)) * uint64(ctx.Runs)
}

return assessments, totalScore
Expand Down
162 changes: 152 additions & 10 deletions evaluate/evaluate_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -139,7 +139,7 @@ func TestEvaluate(t *testing.T) {
return nil
}))

assert.Equal(t, tc.ExpectedAssessments, actualAssessments)
assert.ElementsMatch(t, tc.ExpectedAssessments, actualAssessments)
assert.Equal(t, tc.ExpectedTotalScore, actualTotalScore)

if tc.ExpectedOutputValidate != nil {
Expand Down Expand Up @@ -200,8 +200,15 @@ func TestEvaluate(t *testing.T) {
Task: evaluatetask.IdentifierWriteTests,
Assessment: metrics.Assessments{},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: metrics.Assessments{},
},
},
ExpectedTotalScore: 1,
ExpectedTotalScore: 2,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), mockedModel.ID(), "golang", "golang", "plain.log"): nil,
},
Expand Down Expand Up @@ -246,8 +253,15 @@ func TestEvaluate(t *testing.T) {
Task: evaluatetask.IdentifierWriteTests,
Assessment: metrics.Assessments{},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: metrics.Assessments{},
},
},
ExpectedTotalScore: 1,
ExpectedTotalScore: 2,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): func(t *testing.T, filePath, data string) {
assert.Contains(t, data, ErrEmptyResponseFromModel.Error())
Expand Down Expand Up @@ -301,8 +315,19 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 1,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyGenerateTestsForFileCharacterCount: 14,
metrics.AssessmentKeyResponseCharacterCount: 14,
metrics.AssessmentKeyResponseNoError: 1,
},
},
},
ExpectedTotalScore: 1,
ExpectedTotalScore: 2,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): func(t *testing.T, filePath, data string) {
assert.Contains(t, data, "Attempt 1/3: "+ErrEmptyResponseFromModel.Error())
Expand Down Expand Up @@ -355,8 +380,19 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 1,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyGenerateTestsForFileCharacterCount: 14,
metrics.AssessmentKeyResponseCharacterCount: 14,
metrics.AssessmentKeyResponseNoError: 1,
},
},
},
ExpectedTotalScore: 1,
ExpectedTotalScore: 2,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): func(t *testing.T, filePath, data string) {
assert.Contains(t, data, "DONE 0 tests, 1 error")
Expand Down Expand Up @@ -443,6 +479,17 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 1,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryNextPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 1,
metrics.AssessmentKeyResponseNoError: 1,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
Expand All @@ -454,6 +501,17 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 2,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPlainPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 2,
metrics.AssessmentKeyResponseNoError: 2,
},
},
},
ExpectedTotalScore: 0,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
Expand Down Expand Up @@ -516,6 +574,17 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 2,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryNextPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 2,
metrics.AssessmentKeyResponseNoError: 2,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
Expand All @@ -527,6 +596,17 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 1,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPlainPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 1,
metrics.AssessmentKeyResponseNoError: 1,
},
},
},
ExpectedTotalScore: 0,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
Expand Down Expand Up @@ -581,6 +661,13 @@ func TestEvaluate(t *testing.T) {
Task: evaluatetask.IdentifierWriteTests,
Assessment: map[metrics.AssessmentKey]uint64{},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPlainPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{},
},
},
ExpectedTotalScore: 0,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
Expand Down Expand Up @@ -634,8 +721,19 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 3,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 3,
metrics.AssessmentKeyResponseNoError: 3,
},
},
},
ExpectedTotalScore: 3,
ExpectedTotalScore: 6,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): nil,
},
Expand Down Expand Up @@ -690,8 +788,19 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 3,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 3,
metrics.AssessmentKeyResponseNoError: 3,
},
},
},
ExpectedTotalScore: 3,
ExpectedTotalScore: 6,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): nil,
},
Expand Down Expand Up @@ -776,8 +885,19 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 3,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 3,
metrics.AssessmentKeyResponseNoError: 3,
},
},
},
ExpectedTotalScore: 3,
ExpectedTotalScore: 6,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): nil,
},
Expand Down Expand Up @@ -845,8 +965,19 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 3,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 3,
metrics.AssessmentKeyResponseNoError: 3,
},
},
},
ExpectedTotalScore: 3,
ExpectedTotalScore: 6,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): nil,
},
Expand Down Expand Up @@ -895,8 +1026,19 @@ func TestEvaluate(t *testing.T) {
metrics.AssessmentKeyResponseNoError: 1,
},
},
&metricstesting.AssessmentTuple{
Model: mockedModel,
Language: languageGolang,
RepositoryPath: repositoryPath,
Task: evaluatetask.IdentifierWriteTestsSymflowerFix,
Assessment: map[metrics.AssessmentKey]uint64{
metrics.AssessmentKeyCoverage: 0,
metrics.AssessmentKeyFilesExecuted: 1,
metrics.AssessmentKeyResponseNoError: 1,
},
},
},
ExpectedTotalScore: 1,
ExpectedTotalScore: 2,
ExpectedResultFiles: map[string]func(t *testing.T, filePath string, data string){
filepath.Join(string(evaluatetask.IdentifierWriteTests), evalmodel.CleanModelNameForFileSystem(mockedModelID), "golang", "golang", "plain.log"): nil,
},
Expand Down
16 changes: 16 additions & 0 deletions evaluate/metrics/assessment.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -152,3 +152,19 @@ func (a Assessments) StringCSV() (row []string) {

return row
}

// CombineWithSymflowerFixAssessments combines the model assessments with the ones from "symflower fix".
func CombineWithSymflowerFixAssessments(model Assessments, fixed Assessments) (combined Assessments) {
combined = NewAssessments()

combined[AssessmentKeyCoverage] = fixed[AssessmentKeyCoverage]
combined[AssessmentKeyFilesExecuted] = fixed[AssessmentKeyFilesExecuted]
combined[AssessmentKeyGenerateTestsForFileCharacterCount] = model[AssessmentKeyGenerateTestsForFileCharacterCount]
combined[AssessmentKeyProcessingTime] = model[AssessmentKeyProcessingTime] + fixed[AssessmentKeyProcessingTime]
combined[AssessmentKeyResponseCharacterCount] = model[AssessmentKeyResponseCharacterCount]
combined[AssessmentKeyResponseNoError] = model[AssessmentKeyResponseNoError]
combined[AssessmentKeyResponseNoExcess] = model[AssessmentKeyResponseNoExcess]
combined[AssessmentKeyResponseWithCode] = model[AssessmentKeyResponseWithCode]

return combined
}
51 changes: 51 additions & 0 deletions evaluate/metrics/assessment_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -272,3 +272,54 @@ func TestAssessmentsScore(t *testing.T) {
ExpectedScore: uint64(9),
})
}

func TestCombineModelAndSymflowerFixAssessments(t *testing.T) {
type testCase struct {
Name string

ModelAssessment Assessments
SymflowerFixAssessments Assessments

ExpectedAssessments Assessments
}

validate := func(t *testing.T, tc *testCase) {
t.Run(tc.Name, func(t *testing.T) {
actualAssessments := CombineWithSymflowerFixAssessments(tc.ModelAssessment, tc.SymflowerFixAssessments)

assert.Equal(t, tc.ExpectedAssessments, actualAssessments)
})
}

validate(t, &testCase{
Name: "Simple",

ModelAssessment: Assessments{
AssessmentKeyFilesExecuted: 1,
AssessmentKeyProcessingTime: uint64(200),
AssessmentKeyCoverage: 0,
AssessmentKeyResponseCharacterCount: 100,
AssessmentKeyGenerateTestsForFileCharacterCount: 50,
AssessmentKeyResponseNoError: 0,
AssessmentKeyResponseWithCode: 1,
AssessmentKeyResponseNoExcess: 1,
},
SymflowerFixAssessments: Assessments{
AssessmentKeyFilesExecuted: 1,
AssessmentKeyProcessingTime: uint64(100),
AssessmentKeyCoverage: 10,
AssessmentKeyResponseNoError: 1,
},

ExpectedAssessments: Assessments{
AssessmentKeyFilesExecuted: 1,
AssessmentKeyProcessingTime: uint64(300),
AssessmentKeyCoverage: 10,
AssessmentKeyResponseCharacterCount: 100,
AssessmentKeyGenerateTestsForFileCharacterCount: 50,
AssessmentKeyResponseNoError: 0,
AssessmentKeyResponseWithCode: 1,
AssessmentKeyResponseNoExcess: 1,
},
})
}
Loading
Loading