Yvonne  Hickle

Yvonne Hickle

1662524700

Nimble: A Matcher Framework for Swift and Objective-C

Nimble

Use Nimble to express the expected outcomes of Swift or Objective-C expressions. Inspired by Cedar.

// Swift
expect(1 + 1).to(equal(2))
expect(1.2).to(beCloseTo(1.1, within: 0.1))
expect(3) > 2
expect("seahorse").to(contain("sea"))
expect(["Atlantic", "Pacific"]).toNot(contain("Mississippi"))
expect(ocean.isClean).toEventually(beTruthy())

How to Use Nimble

Some Background: Expressing Outcomes Using Assertions in XCTest

Apple's Xcode includes the XCTest framework, which provides assertion macros to test whether code behaves properly. For example, to assert that 1 + 1 = 2, XCTest has you write:

// Swift

XCTAssertEqual(1 + 1, 2, "expected one plus one to equal two")

Or, in Objective-C:

// Objective-C

XCTAssertEqual(1 + 1, 2, @"expected one plus one to equal two");

XCTest assertions have a couple of drawbacks:

  1. Not enough macros. There's no easy way to assert that a string contains a particular substring, or that a number is less than or equal to another.
  2. It's hard to write asynchronous tests. XCTest forces you to write a lot of boilerplate code.

Nimble addresses these concerns.

Nimble: Expectations Using expect(...).to

Nimble allows you to express expectations using a natural, easily understood language:

// Swift

import Nimble

expect(seagull.squawk).to(equal("Squee!"))
// Objective-C

@import Nimble;

expect(seagull.squawk).to(equal(@"Squee!"));

The expect function autocompletes to include file: and line:, but these parameters are optional. Use the default values to have Xcode highlight the correct line when an expectation is not met.

To perform the opposite expectation--to assert something is not equal--use toNot or notTo:

// Swift

import Nimble

expect(seagull.squawk).toNot(equal("Oh, hello there!"))
expect(seagull.squawk).notTo(equal("Oh, hello there!"))
// Objective-C

@import Nimble;

expect(seagull.squawk).toNot(equal(@"Oh, hello there!"));
expect(seagull.squawk).notTo(equal(@"Oh, hello there!"));

Custom Failure Messages

Would you like to add more information to the test's failure messages? Use the description optional argument to add your own text:

// Swift

expect(1 + 1).to(equal(3))
// failed - expected to equal <3>, got <2>

expect(1 + 1).to(equal(3), description: "Make sure libKindergartenMath is loaded")
// failed - Make sure libKindergartenMath is loaded
// expected to equal <3>, got <2>

Or the *WithDescription version in Objective-C:

// Objective-C

@import Nimble;

expect(@(1+1)).to(equal(@3));
// failed - expected to equal <3.0000>, got <2.0000>

expect(@(1+1)).toWithDescription(equal(@3), @"Make sure libKindergartenMath is loaded");
// failed - Make sure libKindergartenMath is loaded
// expected to equal <3.0000>, got <2.0000>

Type Safety

Nimble makes sure you don't compare two types that don't match:

// Swift

// Does not compile:
expect(1 + 1).to(equal("Squee!"))

Nimble uses generics--only available in Swift--to ensure type correctness. That means type checking is not available when using Nimble in Objective-C. :sob:

Operator Overloads

Tired of so much typing? With Nimble, you can use overloaded operators like == for equivalence, or > for comparisons:

// Swift

// Passes if squawk does not equal "Hi!":
expect(seagull.squawk) != "Hi!"

// Passes if 10 is greater than 2:
expect(10) > 2

Operator overloads are only available in Swift, so you won't be able to use this syntax in Objective-C. :broken_heart:

Lazily Computed Values

The expect function doesn't evaluate the value it's given until it's time to match. So Nimble can test whether an expression raises an exception once evaluated:

// Swift

// Note: Swift currently doesn't have exceptions.
//       Only Objective-C code can raise exceptions
//       that Nimble will catch.
//       (see https://github.com/Quick/Nimble/issues/220#issuecomment-172667064)
let exception = NSException(
    name: NSInternalInconsistencyException,
    reason: "Not enough fish in the sea.",
    userInfo: ["something": "is fishy"])
expect { exception.raise() }.to(raiseException())

// Also, you can customize raiseException to be more specific
expect { exception.raise() }.to(raiseException(named: NSInternalInconsistencyException))
expect { exception.raise() }.to(raiseException(
    named: NSInternalInconsistencyException,
    reason: "Not enough fish in the sea"))
expect { exception.raise() }.to(raiseException(
    named: NSInternalInconsistencyException,
    reason: "Not enough fish in the sea",
    userInfo: ["something": "is fishy"]))

Objective-C works the same way, but you must use the expectAction macro when making an expectation on an expression that has no return value:

// Objective-C

NSException *exception = [NSException exceptionWithName:NSInternalInconsistencyException
                                                 reason:@"Not enough fish in the sea."
                                               userInfo:nil];
expectAction(^{ [exception raise]; }).to(raiseException());

// Use the property-block syntax to be more specific.
expectAction(^{ [exception raise]; }).to(raiseException().named(NSInternalInconsistencyException));
expectAction(^{ [exception raise]; }).to(raiseException().
    named(NSInternalInconsistencyException).
    reason("Not enough fish in the sea"));
expectAction(^{ [exception raise]; }).to(raiseException().
    named(NSInternalInconsistencyException).
    reason("Not enough fish in the sea").
    userInfo(@{@"something": @"is fishy"}));

// You can also pass a block for custom matching of the raised exception
expectAction(exception.raise()).to(raiseException().satisfyingBlock(^(NSException *exception) {
    expect(exception.name).to(beginWith(NSInternalInconsistencyException));
}));

C Primitives

Some testing frameworks make it hard to test primitive C values. In Nimble, it just works:

// Swift

let actual: CInt = 1
let expectedValue: CInt = 1
expect(actual).to(equal(expectedValue))

In fact, Nimble uses type inference, so you can write the above without explicitly specifying both types:

// Swift

expect(1 as CInt).to(equal(1))

In Objective-C, Nimble only supports Objective-C objects. To make expectations on primitive C values, wrap then in an object literal:

expect(@(1 + 1)).to(equal(@2));

Asynchronous Expectations

In Nimble, it's easy to make expectations on values that are updated asynchronously. Just use toEventually or toEventuallyNot:

// Swift
DispatchQueue.main.async {
    ocean.add("dolphins")
    ocean.add("whales")
}
expect(ocean).toEventually(contain("dolphins", "whales"))
// Objective-C

dispatch_async(dispatch_get_main_queue(), ^{
    [ocean add:@"dolphins"];
    [ocean add:@"whales"];
});
expect(ocean).toEventually(contain(@"dolphins", @"whales"));

Note: toEventually triggers its polls on the main thread. Blocking the main thread will cause Nimble to stop the run loop. This can cause test pollution for whatever incomplete code that was running on the main thread. Blocking the main thread can be caused by blocking IO, calls to sleep(), deadlocks, and synchronous IPC.

In the above example, ocean is constantly re-evaluated. If it ever contains dolphins and whales, the expectation passes. If ocean still doesn't contain them, even after being continuously re-evaluated for one whole second, the expectation fails.

You can also test that a value always or never matches throughout the length of the timeout. Use toNever and toAlways for this:

// Swift
ocean.add("dolphins")
expect(ocean).toAlways(contain("dolphins"))
expect(ocean).toNever(contain("hares"))
// Objective-C
[ocean add:@"dolphins"]
expect(ocean).toAlways(contain(@"dolphins"))
expect(ocean).toNever(contain(@"hares"))

Sometimes it takes more than a second for a value to update. In those cases, use the timeout parameter:

// Swift

// Waits three seconds for ocean to contain "starfish":
expect(ocean).toEventually(contain("starfish"), timeout: .seconds(3))

// Evaluate someValue every 0.2 seconds repeatedly until it equals 100, or fails if it timeouts after 5.5 seconds.
expect(someValue).toEventually(equal(100), timeout: .milliseconds(5500), pollInterval: .milliseconds(200))
// Objective-C

// Waits three seconds for ocean to contain "starfish":
expect(ocean).withTimeout(3).toEventually(contain(@"starfish"));

You can also provide a callback by using the waitUntil function:

// Swift

waitUntil { done in
    ocean.goFish { success in
        expect(success).to(beTrue())
        done()
    }
}
// Objective-C

waitUntil(^(void (^done)(void)){
    [ocean goFishWithHandler:^(BOOL success){
        expect(success).to(beTrue());
        done();
    }];
});

waitUntil also optionally takes a timeout parameter:

// Swift

waitUntil(timeout: .seconds(10)) { done in
    ocean.goFish { success in
        expect(success).to(beTrue())
        done()
    }
}
// Objective-C

waitUntilTimeout(10, ^(void (^done)(void)){
    [ocean goFishWithHandler:^(BOOL success){
        expect(success).to(beTrue());
        done();
    }];
});

Note: waitUntil triggers its timeout code on the main thread. Blocking the main thread will cause Nimble to stop the run loop to continue. This can cause test pollution for whatever incomplete code that was running on the main thread. Blocking the main thread can be caused by blocking IO, calls to sleep(), deadlocks, and synchronous IPC.

In some cases (e.g. when running on slower machines) it can be useful to modify the default timeout and poll interval values. This can be done as follows:

// Swift

// Increase the global timeout to 5 seconds:
Nimble.AsyncDefaults.timeout = .seconds(5)

// Slow the polling interval to 0.1 seconds:
Nimble.AsyncDefaults.pollInterval = .milliseconds(100)

Objective-C Support

Nimble has full support for Objective-C. However, there are two things to keep in mind when using Nimble in Objective-C:

All parameters passed to the expect function, as well as matcher functions like equal, must be Objective-C objects or can be converted into an NSObject equivalent:

// Objective-C

@import Nimble;

expect(@(1 + 1)).to(equal(@2));
expect(@"Hello world").to(contain(@"world"));

// Boxed as NSNumber *
expect(2).to(equal(2));
expect(1.2).to(beLessThan(2.0));
expect(true).to(beTruthy());

// Boxed as NSString *
expect("Hello world").to(equal("Hello world"));

// Boxed as NSRange
expect(NSMakeRange(1, 10)).to(equal(NSMakeRange(1, 10)));

To make an expectation on an expression that does not return a value, such as -[NSException raise], use expectAction instead of expect:

// Objective-C

expectAction(^{ [exception raise]; }).to(raiseException());

The following types are currently converted to an NSObject type:

  • C Numeric types are converted to NSNumber *
  • NSRange is converted to NSValue *
  • char * is converted to NSString *

For the following matchers:

  • equal
  • beGreaterThan
  • beGreaterThanOrEqual
  • beLessThan
  • beLessThanOrEqual
  • beCloseTo
  • beTrue
  • beFalse
  • beTruthy
  • beFalsy
  • haveCount

If you would like to see more, file an issue.

Disabling Objective-C Shorthand

Nimble provides a shorthand for expressing expectations using the expect function. To disable this shorthand in Objective-C, define the NIMBLE_DISABLE_SHORT_SYNTAX macro somewhere in your code before importing Nimble:

#define NIMBLE_DISABLE_SHORT_SYNTAX 1

@import Nimble;

NMB_expect(^{ return seagull.squawk; }, __FILE__, __LINE__).to(NMB_equal(@"Squee!"));

Disabling the shorthand is useful if you're testing functions with names that conflict with Nimble functions, such as expect or equal. If that's not the case, there's no point in disabling the shorthand.

Built-in Matcher Functions

Nimble includes a wide variety of matcher functions.

Type Checking

Nimble supports checking the type membership of any kind of object, whether Objective-C conformant or not:

// Swift

protocol SomeProtocol{}
class SomeClassConformingToProtocol: SomeProtocol{}
struct SomeStructConformingToProtocol: SomeProtocol{}

// The following tests pass
expect(1).to(beAKindOf(Int.self))
expect("turtle").to(beAKindOf(String.self))

let classObject = SomeClassConformingToProtocol()
expect(classObject).to(beAKindOf(SomeProtocol.self))
expect(classObject).to(beAKindOf(SomeClassConformingToProtocol.self))
expect(classObject).toNot(beAKindOf(SomeStructConformingToProtocol.self))

let structObject = SomeStructConformingToProtocol()
expect(structObject).to(beAKindOf(SomeProtocol.self))
expect(structObject).to(beAKindOf(SomeStructConformingToProtocol.self))
expect(structObject).toNot(beAKindOf(SomeClassConformingToProtocol.self))
// Objective-C

// The following tests pass
NSMutableArray *array = [NSMutableArray array];
expect(array).to(beAKindOf([NSArray class]));
expect(@1).toNot(beAKindOf([NSNull class]));

Objects can be tested for their exact types using the beAnInstanceOf matcher:

// Swift

protocol SomeProtocol{}
class SomeClassConformingToProtocol: SomeProtocol{}
struct SomeStructConformingToProtocol: SomeProtocol{}

// Unlike the 'beKindOf' matcher, the 'beAnInstanceOf' matcher only
// passes if the object is the EXACT type requested. The following
// tests pass -- note its behavior when working in an inheritance hierarchy.
expect(1).to(beAnInstanceOf(Int.self))
expect("turtle").to(beAnInstanceOf(String.self))

let classObject = SomeClassConformingToProtocol()
expect(classObject).toNot(beAnInstanceOf(SomeProtocol.self))
expect(classObject).to(beAnInstanceOf(SomeClassConformingToProtocol.self))
expect(classObject).toNot(beAnInstanceOf(SomeStructConformingToProtocol.self))

let structObject = SomeStructConformingToProtocol()
expect(structObject).toNot(beAnInstanceOf(SomeProtocol.self))
expect(structObject).to(beAnInstanceOf(SomeStructConformingToProtocol.self))
expect(structObject).toNot(beAnInstanceOf(SomeClassConformingToProtocol.self))

Equivalence

// Swift

// Passes if 'actual' is equivalent to 'expected':
expect(actual).to(equal(expected))
expect(actual) == expected

// Passes if 'actual' is not equivalent to 'expected':
expect(actual).toNot(equal(expected))
expect(actual) != expected
// Objective-C

// Passes if 'actual' is equivalent to 'expected':
expect(actual).to(equal(expected))

// Passes if 'actual' is not equivalent to 'expected':
expect(actual).toNot(equal(expected))

Values must be Equatable, Comparable, or subclasses of NSObject. equal will always fail when used to compare one or more nil values.

Identity

// Swift

// Passes if 'actual' has the same pointer address as 'expected':
expect(actual).to(beIdenticalTo(expected))
expect(actual) === expected

// Passes if 'actual' does not have the same pointer address as 'expected':
expect(actual).toNot(beIdenticalTo(expected))
expect(actual) !== expected

It is important to remember that beIdenticalTo only makes sense when comparing types with reference semantics, which have a notion of identity. In Swift, that means types that are defined as a class.

This matcher will not work when comparing types with value semantics such as those defined as a struct or enum. If you need to compare two value types, consider what it means for instances of your type to be identical. This may mean comparing individual properties or, if it makes sense to do so, conforming your type to Equatable and using Nimble's equivalence matchers instead.

// Objective-C

// Passes if 'actual' has the same pointer address as 'expected':
expect(actual).to(beIdenticalTo(expected));

// Passes if 'actual' does not have the same pointer address as 'expected':
expect(actual).toNot(beIdenticalTo(expected));

Comparisons

// Swift

expect(actual).to(beLessThan(expected))
expect(actual) < expected

expect(actual).to(beLessThanOrEqualTo(expected))
expect(actual) <= expected

expect(actual).to(beGreaterThan(expected))
expect(actual) > expected

expect(actual).to(beGreaterThanOrEqualTo(expected))
expect(actual) >= expected
// Objective-C

expect(actual).to(beLessThan(expected));
expect(actual).to(beLessThanOrEqualTo(expected));
expect(actual).to(beGreaterThan(expected));
expect(actual).to(beGreaterThanOrEqualTo(expected));

Values given to the comparison matchers above must implement Comparable.

Because of how computers represent floating point numbers, assertions that two floating point numbers be equal will sometimes fail. To express that two numbers should be close to one another within a certain margin of error, use beCloseTo:

// Swift

expect(actual).to(beCloseTo(expected, within: delta))
// Objective-C

expect(actual).to(beCloseTo(expected).within(delta));

For example, to assert that 10.01 is close to 10, you can write:

// Swift

expect(10.01).to(beCloseTo(10, within: 0.1))
// Objective-C

expect(@(10.01)).to(beCloseTo(@10).within(0.1));

There is also an operator shortcut available in Swift:

// Swift

expect(actual) ≈ expected
expect(actual) ≈ (expected, delta)

(Type option+x to get on a U.S. keyboard)

The former version uses the default delta of 0.0001. Here is yet another way to do this:

// Swift

expect(actual) ≈ expected ± delta
expect(actual) == expected ± delta

(Type option+shift+= to get ± on a U.S. keyboard)

If you are comparing arrays of floating point numbers, you'll find the following useful:

// Swift

expect([0.0, 2.0]) ≈ [0.0001, 2.0001]
expect([0.0, 2.0]).to(beCloseTo([0.1, 2.1], within: 0.1))

Values given to the beCloseTo matcher must conform to FloatingPoint.

Types/Classes

// Swift

// Passes if 'instance' is an instance of 'aClass':
expect(instance).to(beAnInstanceOf(aClass))

// Passes if 'instance' is an instance of 'aClass' or any of its subclasses:
expect(instance).to(beAKindOf(aClass))
// Objective-C

// Passes if 'instance' is an instance of 'aClass':
expect(instance).to(beAnInstanceOf(aClass));

// Passes if 'instance' is an instance of 'aClass' or any of its subclasses:
expect(instance).to(beAKindOf(aClass));

Instances must be Objective-C objects: subclasses of NSObject, or Swift objects bridged to Objective-C with the @objc prefix.

For example, to assert that dolphin is a kind of Mammal:

// Swift

expect(dolphin).to(beAKindOf(Mammal))
// Objective-C

expect(dolphin).to(beAKindOf([Mammal class]));

beAnInstanceOf uses the -[NSObject isMemberOfClass:] method to test membership. beAKindOf uses -[NSObject isKindOfClass:].

Truthiness

// Passes if 'actual' is not nil, true, or an object with a boolean value of true:
expect(actual).to(beTruthy())

// Passes if 'actual' is only true (not nil or an object conforming to Boolean true):
expect(actual).to(beTrue())

// Passes if 'actual' is nil, false, or an object with a boolean value of false:
expect(actual).to(beFalsy())

// Passes if 'actual' is only false (not nil or an object conforming to Boolean false):
expect(actual).to(beFalse())

// Passes if 'actual' is nil:
expect(actual).to(beNil())
// Objective-C

// Passes if 'actual' is not nil, true, or an object with a boolean value of true:
expect(actual).to(beTruthy());

// Passes if 'actual' is only true (not nil or an object conforming to Boolean true):
expect(actual).to(beTrue());

// Passes if 'actual' is nil, false, or an object with a boolean value of false:
expect(actual).to(beFalsy());

// Passes if 'actual' is only false (not nil or an object conforming to Boolean false):
expect(actual).to(beFalse());

// Passes if 'actual' is nil:
expect(actual).to(beNil());

Swift Assertions

If you're using Swift, you can use the throwAssertion matcher to check if an assertion is thrown (e.g. fatalError()). This is made possible by @mattgallagher's CwlPreconditionTesting library.

// Swift

// Passes if 'somethingThatThrows()' throws an assertion, 
// such as by calling 'fatalError()' or if a precondition fails:
expect { try somethingThatThrows() }.to(throwAssertion())
expect { () -> Void in fatalError() }.to(throwAssertion())
expect { precondition(false) }.to(throwAssertion())

// Passes if throwing an NSError is not equal to throwing an assertion:
expect { throw NSError(domain: "test", code: 0, userInfo: nil) }.toNot(throwAssertion())

// Passes if the code after the precondition check is not run:
var reachedPoint1 = false
var reachedPoint2 = false
expect {
    reachedPoint1 = true
    precondition(false, "condition message")
    reachedPoint2 = true
}.to(throwAssertion())

expect(reachedPoint1) == true
expect(reachedPoint2) == false

Notes:

  • This feature is only available in Swift.
  • The tvOS simulator is supported, but using a different mechanism, requiring you to turn off the Debug executable scheme setting for your tvOS scheme's Test configuration.

Swift Error Handling

You can use the throwError matcher to check if an error is thrown.

// Swift

// Passes if 'somethingThatThrows()' throws an 'Error':
expect { try somethingThatThrows() }.to(throwError())

// Passes if 'somethingThatThrows()' throws an error within a particular domain:
expect { try somethingThatThrows() }.to(throwError { (error: Error) in
    expect(error._domain).to(equal(NSCocoaErrorDomain))
})

// Passes if 'somethingThatThrows()' throws a particular error enum case:
expect { try somethingThatThrows() }.to(throwError(NSCocoaError.PropertyListReadCorruptError))

// Passes if 'somethingThatThrows()' throws an error of a particular type:
expect { try somethingThatThrows() }.to(throwError(errorType: NimbleError.self))

When working directly with Error values, using the matchError matcher allows you to perform certain checks on the error itself without having to explicitly cast the error.

The matchError matcher allows you to check whether or not the error:

  • is the same type of error you are expecting.
  • represents a particular error value that you are expecting.

This can be useful when using Result or Promise types, for example.

// Swift

let actual: Error = ...

// Passes if 'actual' represents any error value from the NimbleErrorEnum type:
expect(actual).to(matchError(NimbleErrorEnum.self))

// Passes if 'actual' represents the case 'timeout' from the NimbleErrorEnum type:
expect(actual).to(matchError(NimbleErrorEnum.timeout))

// Passes if 'actual' contains an NSError equal to the one provided:
expect(actual).to(matchError(NSError(domain: "err", code: 123, userInfo: nil)))

Note: This feature is only available in Swift.

Exceptions

// Swift

// Passes if 'actual', when evaluated, raises an exception:
expect(actual).to(raiseException())

// Passes if 'actual' raises an exception with the given name:
expect(actual).to(raiseException(named: name))

// Passes if 'actual' raises an exception with the given name and reason:
expect(actual).to(raiseException(named: name, reason: reason))

// Passes if 'actual' raises an exception which passes expectations defined in the given closure:
// (in this case, if the exception's name begins with "a r")
expect { exception.raise() }.to(raiseException { (exception: NSException) in
    expect(exception.name).to(beginWith("a r"))
})
// Objective-C

// Passes if 'actual', when evaluated, raises an exception:
expect(actual).to(raiseException())

// Passes if 'actual' raises an exception with the given name
expect(actual).to(raiseException().named(name))

// Passes if 'actual' raises an exception with the given name and reason:
expect(actual).to(raiseException().named(name).reason(reason))

// Passes if 'actual' raises an exception and it passes expectations defined in the given block:
// (in this case, if name begins with "a r")
expect(actual).to(raiseException().satisfyingBlock(^(NSException *exception) {
    expect(exception.name).to(beginWith(@"a r"));
}));

Note: Swift currently doesn't have exceptions (see #220). Only Objective-C code can raise exceptions that Nimble will catch.

Collection Membership

// Swift

// Passes if all of the expected values are members of 'actual':
expect(actual).to(contain(expected...))

// Passes if 'actual' is empty (i.e. it contains no elements):
expect(actual).to(beEmpty())
// Objective-C

// Passes if expected is a member of 'actual':
expect(actual).to(contain(expected));

// Passes if 'actual' is empty (i.e. it contains no elements):
expect(actual).to(beEmpty());

In Swift contain takes any number of arguments. The expectation passes if all of them are members of the collection. In Objective-C, contain only takes one argument for now.

For example, to assert that a list of sea creature names contains "dolphin" and "starfish":

// Swift

expect(["whale", "dolphin", "starfish"]).to(contain("dolphin", "starfish"))
// Objective-C

expect(@[@"whale", @"dolphin", @"starfish"]).to(contain(@"dolphin"));
expect(@[@"whale", @"dolphin", @"starfish"]).to(contain(@"starfish"));

contain and beEmpty expect collections to be instances of NSArray, NSSet, or a Swift collection composed of Equatable elements.

To test whether a set of elements is present at the beginning or end of an ordered collection, use beginWith and endWith:

// Swift

// Passes if the elements in expected appear at the beginning of 'actual':
expect(actual).to(beginWith(expected...))

// Passes if the the elements in expected come at the end of 'actual':
expect(actual).to(endWith(expected...))
// Objective-C

// Passes if the elements in expected appear at the beginning of 'actual':
expect(actual).to(beginWith(expected));

// Passes if the the elements in expected come at the end of 'actual':
expect(actual).to(endWith(expected));

beginWith and endWith expect collections to be instances of NSArray, or ordered Swift collections composed of Equatable elements.

Like contain, in Objective-C beginWith and endWith only support a single argument for now.

For code that returns collections of complex objects without a strict ordering, there is the containElementSatisfying matcher:

// Swift

struct Turtle {
    let color: String
}

let turtles: [Turtle] = functionThatReturnsSomeTurtlesInAnyOrder()

// This set of matchers passes regardless of whether the array is 
// [{color: "blue"}, {color: "green"}] or [{color: "green"}, {color: "blue"}]:

expect(turtles).to(containElementSatisfying({ turtle in
    return turtle.color == "green"
}))
expect(turtles).to(containElementSatisfying({ turtle in
    return turtle.color == "blue"
}, "that is a turtle with color 'blue'"))

// The second matcher will incorporate the provided string in the error message
// should it fail
// Objective-C

@interface Turtle : NSObject
@property (nonatomic, readonly, nonnull) NSString *color;
@end

@implementation Turtle 
@end

NSArray<Turtle *> * __nonnull turtles = functionThatReturnsSomeTurtlesInAnyOrder();

// This set of matchers passes regardless of whether the array is 
// [{color: "blue"}, {color: "green"}] or [{color: "green"}, {color: "blue"}]:

expect(turtles).to(containElementSatisfying(^BOOL(id __nonnull object) {
    return [[turtle color] isEqualToString:@"green"];
}));
expect(turtles).to(containElementSatisfying(^BOOL(id __nonnull object) {
    return [[turtle color] isEqualToString:@"blue"];
}));

For asserting on if the given Comparable value is inside of a Range, use the beWithin matcher.

// Swift

// Passes if 5 is within the range 1 through 10, inclusive
expect(5).to(beWithin(1...10))

// Passes if 5 is not within the range 2 through 4.
expect(5).toNot(beWithin(2..<5))

Strings

// Swift

// Passes if 'actual' contains 'substring':
expect(actual).to(contain(substring))

// Passes if 'actual' begins with 'prefix':
expect(actual).to(beginWith(prefix))

// Passes if 'actual' ends with 'suffix':
expect(actual).to(endWith(suffix))

// Passes if 'actual' represents the empty string, "":
expect(actual).to(beEmpty())

// Passes if 'actual' matches the regular expression defined in 'expected':
expect(actual).to(match(expected))
// Objective-C

// Passes if 'actual' contains 'substring':
expect(actual).to(contain(expected));

// Passes if 'actual' begins with 'prefix':
expect(actual).to(beginWith(prefix));

// Passes if 'actual' ends with 'suffix':
expect(actual).to(endWith(suffix));

// Passes if 'actual' represents the empty string, "":
expect(actual).to(beEmpty());

// Passes if 'actual' matches the regular expression defined in 'expected':
expect(actual).to(match(expected))

Collection Elements

Nimble provides a means to check that all elements of a collection pass a given expectation.

Swift

In Swift, the collection must be an instance of a type conforming to Sequence.

// Swift

// Providing a custom function:
expect([1, 2, 3, 4]).to(allPass { $0 < 5 })

// Composing the expectation with another matcher:
expect([1, 2, 3, 4]).to(allPass(beLessThan(5)))

Objective-C

In Objective-C, the collection must be an instance of a type which implements the NSFastEnumeration protocol, and whose elements are instances of a type which subclasses NSObject.

Additionally, unlike in Swift, there is no override to specify a custom matcher function.

// Objective-C

expect(@[@1, @2, @3, @4]).to(allPass(beLessThan(@5)));

Collection Count

// Swift

// Passes if 'actual' contains the 'expected' number of elements:
expect(actual).to(haveCount(expected))

// Passes if 'actual' does _not_ contain the 'expected' number of elements:
expect(actual).notTo(haveCount(expected))
// Objective-C

// Passes if 'actual' contains the 'expected' number of elements:
expect(actual).to(haveCount(expected))

// Passes if 'actual' does _not_ contain the 'expected' number of elements:
expect(actual).notTo(haveCount(expected))

For Swift, the actual value must be an instance of a type conforming to Collection. For example, instances of Array, Dictionary, or Set.

For Objective-C, the actual value must be one of the following classes, or their subclasses:

  • NSArray,
  • NSDictionary,
  • NSSet, or
  • NSHashTable.

Notifications

// Swift
let testNotification = Notification(name: Notification.Name("Foo"), object: nil)

// Passes if the closure in expect { ... } posts a notification to the default
// notification center.
expect {
    NotificationCenter.default.post(testNotification)
}.to(postNotifications(equal([testNotification])))

// Passes if the closure in expect { ... } posts a notification to a given
// notification center
let notificationCenter = NotificationCenter()
expect {
    notificationCenter.post(testNotification)
}.to(postNotifications(equal([testNotification]), from: notificationCenter))

// Passes if the closure in expect { ... } posts a notification with the provided names to a given
// notification center. Make sure to use this when running tests on Catalina, 
// using DistributedNotificationCenter as there is currently no way 
// of observing notifications without providing specific names.
let distributedNotificationCenter = DistributedNotificationCenter()
expect {
    distributedNotificationCenter.post(testNotification)
}.toEventually(postDistributedNotifications(equal([testNotification]),
                                  from: distributedNotificationCenter,
                                  names: [testNotification.name]))

This matcher is only available in Swift.

Result

// Swift
let aResult: Result<String, Error> = .success("Hooray") 

// passes if result is .success
expect(aResult).to(beSuccess()) 

// passes if result value is .success and validates Success value
expect(aResult).to(beSuccess { value in
    expect(value).to(equal("Hooray"))
})


enum AnError: Error {
    case somethingHappened
}
let otherResult: Result<String, AnError> = .failure(.somethingHappened) 

// passes if result is .failure
expect(otherResult).to(beFailure()) 

// passes if result value is .failure and validates error
expect(otherResult).to(beFailure { error in
    expect(error).to(matchError(AnError.somethingHappened))
}) 

This matcher is only available in Swift.

Matching a value to any of a group of matchers

// Swift

// passes if actual is either less than 10 or greater than 20
expect(actual).to(satisfyAnyOf(beLessThan(10), beGreaterThan(20)))

// can include any number of matchers -- the following will pass
// **be careful** -- too many matchers can be the sign of an unfocused test
expect(6).to(satisfyAnyOf(equal(2), equal(3), equal(4), equal(5), equal(6), equal(7)))

// in Swift you also have the option to use the || operator to achieve a similar function
expect(82).to(beLessThan(50) || beGreaterThan(80))
// Objective-C

// passes if actual is either less than 10 or greater than 20
expect(actual).to(satisfyAnyOf(beLessThan(@10), beGreaterThan(@20)))

// can include any number of matchers -- the following will pass
// **be careful** -- too many matchers can be the sign of an unfocused test
expect(@6).to(satisfyAnyOf(equal(@2), equal(@3), equal(@4), equal(@5), equal(@6), equal(@7)))

Note: This matcher allows you to chain any number of matchers together. This provides flexibility, but if you find yourself chaining many matchers together in one test, consider whether you could instead refactor that single test into multiple, more precisely focused tests for better coverage.

Custom Validation

// Swift

// passes if .succeeded is returned from the closure
expect {
    guard case .enumCaseWithAssociatedValueThatIDontCareAbout = actual else {
        return .failed(reason: "wrong enum case")
    }

    return .succeeded
}.to(succeed())

// passes if .failed is returned from the closure
expect {
    guard case .enumCaseWithAssociatedValueThatIDontCareAbout = actual else {
        return .failed(reason: "wrong enum case")
    }

    return .succeeded
}.notTo(succeed())

The String provided with .failed() is shown when the test fails.

When using toEventually() be careful not to make state changes or run process intensive code since this closure will be ran many times.

Writing Your Own Matchers

In Nimble, matchers are Swift functions that take an expected value and return a Predicate closure. Take equal, for example:

// Swift

public func equal<T: Equatable>(expectedValue: T?) -> Predicate<T> {
    // Can be shortened to:
    //   Predicate { actual in  ... }
    //
    // But shown with types here for clarity.
    return Predicate { (actualExpression: Expression<T>) throws -> PredicateResult in
        let msg = ExpectationMessage.expectedActualValueTo("equal <\(expectedValue)>")
        if let actualValue = try actualExpression.evaluate() {
            return PredicateResult(
                bool: actualValue == expectedValue!,
                message: msg
            )
        } else {
            return PredicateResult(
                status: .fail,
                message: msg.appendedBeNilHint()
            )
        }
    }
}

The return value of a Predicate closure is a PredicateResult that indicates whether the actual value matches the expectation and what error message to display on failure.

The actual equal matcher function does not match when expected are nil; the example above has been edited for brevity.

Since matchers are just Swift functions, you can define them anywhere: at the top of your test file, in a file shared by all of your tests, or in an Xcode project you distribute to others.

If you write a matcher you think everyone can use, consider adding it to Nimble's built-in set of matchers by sending a pull request! Or distribute it yourself via GitHub.

For examples of how to write your own matchers, just check out the Matchers directory to see how Nimble's built-in set of matchers are implemented. You can also check out the tips below.

PredicateResult

PredicateResult is the return struct that Predicate return to indicate success and failure. A PredicateResult is made up of two values: PredicateStatus and ExpectationMessage.

Instead of a boolean, PredicateStatus captures a trinary set of values:

// Swift

public enum PredicateStatus {
// The predicate "passes" with the given expression
// eg - expect(1).to(equal(1))
case matches

// The predicate "fails" with the given expression
// eg - expect(1).toNot(equal(1))
case doesNotMatch

// The predicate never "passes" with the given expression, even if negated
// eg - expect(nil as Int?).toNot(equal(1))
case fail

// ...
}

Meanwhile, ExpectationMessage provides messaging semantics for error reporting.

// Swift

public indirect enum ExpectationMessage {
// Emits standard error message:
// eg - "expected to <string>, got <actual>"
case expectedActualValueTo(/* message: */ String)

// Allows any free-form message
// eg - "<string>"
case fail(/* message: */ String)

// ...
}

Predicates should usually depend on either .expectedActualValueTo(..) or .fail(..) when reporting errors. Special cases can be used for the other enum cases.

Finally, if your Predicate utilizes other Predicates, you can utilize .appended(details:) and .appended(message:) methods to annotate an existing error with more details.

A common message to append is failing on nils. For that, .appendedBeNilHint() can be used.

Lazy Evaluation

actualExpression is a lazy, memoized closure around the value provided to the expect function. The expression can either be a closure or a value directly passed to expect(...). In order to determine whether that value matches, custom matchers should call actualExpression.evaluate():

// Swift

public func beNil<T>() -> Predicate<T> {
    // Predicate.simpleNilable(..) automatically generates ExpectationMessage for
    // us based on the string we provide to it. Also, the 'Nilable' postfix indicates
    // that this Predicate supports matching against nil actualExpressions, instead of
    // always resulting in a PredicateStatus.fail result -- which is true for
    // Predicate.simple(..)
    return Predicate.simpleNilable("be nil") { actualExpression in
        let actualValue = try actualExpression.evaluate()
        return PredicateStatus(bool: actualValue == nil)
    }
}

In the above example, actualExpression is not nil -- it is a closure that returns a value. The value it returns, which is accessed via the evaluate() method, may be nil. If that value is nil, the beNil matcher function returns true, indicating that the expectation passed.

Type Checking via Swift Generics

Using Swift's generics, matchers can constrain the type of the actual value passed to the expect function by modifying the return type.

For example, the following matcher, haveDescription, only accepts actual values that implement the Printable protocol. It checks their description against the one provided to the matcher function, and passes if they are the same:

// Swift

public func haveDescription(description: String) -> Predicate<Printable?> {
    return Predicate.simple("have description") { actual in
        return PredicateStatus(bool: actual.evaluate().description == description)
    }
}

Customizing Failure Messages

When using Predicate.simple(..) or Predicate.simpleNilable(..), Nimble outputs the following failure message when an expectation fails:

// where `message` is the first string argument and
// `actual` is the actual value received in `expect(..)`
"expected to \(message), got <\(actual)>"

You can customize this message by modifying the way you create a Predicate.

Basic Customization

For slightly more complex error messaging, receive the created failure message with Predicate.define(..):

// Swift

public func equal<T: Equatable>(_ expectedValue: T?) -> Predicate<T> {
    return Predicate.define("equal <\(stringify(expectedValue))>") { actualExpression, msg in
        let actualValue = try actualExpression.evaluate()
        let matches = actualValue == expectedValue && expectedValue != nil
        if expectedValue == nil || actualValue == nil {
            if expectedValue == nil && actualValue != nil {
                return PredicateResult(
                    status: .fail,
                    message: msg.appendedBeNilHint()
                )
            }
            return PredicateResult(status: .fail, message: msg)
        }
        return PredicateResult(bool: matches, message: msg)
    }
}

In the example above, msg is defined based on the string given to Predicate.define. The code looks akin to:

// Swift

let msg = ExpectationMessage.expectedActualValueTo("equal <\(stringify(expectedValue))>")

Full Customization

To fully customize the behavior of the Predicate, use the overload that expects a PredicateResult to be returned.

Along with PredicateResult, there are other ExpectationMessage enum values you can use:

public indirect enum ExpectationMessage {
// Emits standard error message:
// eg - "expected to <message>, got <actual>"
case expectedActualValueTo(/* message: */ String)

// Allows any free-form message
// eg - "<message>"
case fail(/* message: */ String)

// Emits standard error message with a custom actual value instead of the default.
// eg - "expected to <message>, got <actual>"
case expectedCustomValueTo(/* message: */ String, /* actual: */ String)

// Emits standard error message without mentioning the actual value
// eg - "expected to <message>"
case expectedTo(/* message: */ String)

// ...
}

For matchers that compose other matchers, there are a handful of helper functions to annotate messages.

appended(message: String) is used to append to the original failure message:

// produces "expected to be true, got <actual> (use beFalse() for inverse)"
// appended message do show up inline in Xcode.
.expectedActualValueTo("be true").appended(message: " (use beFalse() for inverse)")

For a more comprehensive message that spans multiple lines, use appended(details: String) instead:

// produces "expected to be true, got <actual>\n\nuse beFalse() for inverse\nor use beNil()"
// details do not show inline in Xcode, but do show up in test logs.
.expectedActualValueTo("be true").appended(details: "use beFalse() for inverse\nor use beNil()")

Supporting Objective-C

To use a custom matcher written in Swift from Objective-C, you'll have to extend the NMBPredicate class, adding a new class method for your custom matcher. The example below defines the class method +[NMBPredicate beNilMatcher]:

// Swift

extension NMBPredicate {
    @objc public class func beNilMatcher() -> NMBPredicate {
        return NMBPredicate { actualExpression in
            return try beNil().satisfies(actualExpression).toObjectiveC()
        }
    }
}

The above allows you to use the matcher from Objective-C:

// Objective-C

expect(actual).to([NMBPredicate beNilMatcher]());

To make the syntax easier to use, define a C function that calls the class method:

// Objective-C

FOUNDATION_EXPORT NMBPredicate *beNil() {
    return [NMBPredicate beNilMatcher];
}

Properly Handling nil in Objective-C Matchers

When supporting Objective-C, make sure you handle nil appropriately. Like Cedar, most matchers do not match with nil. This is to bring prevent test writers from being surprised by nil values where they did not expect them.

Nimble provides the beNil matcher function for test writer that want to make expectations on nil objects:

// Objective-C

expect(nil).to(equal(nil)); // fails
expect(nil).to(beNil());    // passes

If your matcher does not want to match with nil, you use Predicate.define or Predicate.simple. Using those factory methods will automatically generate expected value failure messages when they're nil.

public func beginWith<S: Sequence>(_ startingElement: S.Element) -> Predicate<S> where S.Element: Equatable {
    return Predicate.simple("begin with <\(startingElement)>") { actualExpression in
        guard let actualValue = try actualExpression.evaluate() else { return .fail }

        var actualGenerator = actualValue.makeIterator()
        return PredicateStatus(bool: actualGenerator.next() == startingElement)
    }
}

extension NMBPredicate {
    @objc public class func beginWithMatcher(_ expected: Any) -> NMBPredicate {
        return NMBPredicate { actualExpression in
            let actual = try actualExpression.evaluate()
            let expr = actualExpression.cast { $0 as? NMBOrderedCollection }
            return try beginWith(expected).satisfies(expr).toObjectiveC()
        }
    }
}

Migrating from the Old Matcher API

Previously (<7.0.0), Nimble supported matchers via the following types:

  • Matcher
  • NonNilMatcherFunc
  • MatcherFunc

All of those types have been replaced by Predicate. The old API has been removed completely in Nimble v10.

Installing Nimble

Nimble can be used on its own, or in conjunction with its sister project, Quick. To install both Quick and Nimble, follow the installation instructions in the Quick Documentation.

Nimble can currently be installed in one of two ways: using CocoaPods, or with git submodules.

Installing Nimble as a Submodule

To use Nimble as a submodule to test your macOS, iOS or tvOS applications, follow these 4 easy steps:

  1. Clone the Nimble repository
  2. Add Nimble.xcodeproj to the Xcode workspace for your project
  3. Link Nimble.framework to your test target
  4. Start writing expectations!

For more detailed instructions on each of these steps, read How to Install Quick. Ignore the steps involving adding Quick to your project in order to install just Nimble.

Installing Nimble via CocoaPods

To use Nimble in CocoaPods to test your macOS, iOS or tvOS applications, add Nimble to your podfile and add the use_frameworks! line to enable Swift support for CocoaPods.

platform :ios, '8.0'

source 'https://github.com/CocoaPods/Specs.git'

# Whatever pods you need for your app go here

target 'YOUR_APP_NAME_HERE_Tests', :exclusive => true do
  use_frameworks!
  pod 'Nimble'
end

Finally run pod install.

Using Nimble without XCTest

Nimble is integrated with XCTest to allow it work well when used in Xcode test bundles, however it can also be used in a standalone app. After installing Nimble using one of the above methods, there are two additional steps required to make this work.

  1. Create a custom assertion handler and assign an instance of it to the global NimbleAssertionHandler variable. For example:
class MyAssertionHandler : AssertionHandler {
    func assert(assertion: Bool, message: FailureMessage, location: SourceLocation) {
        if (!assertion) {
            print("Expectation failed: \(message.stringValue)")
        }
    }
}
// Somewhere before you use any assertions
NimbleAssertionHandler = MyAssertionHandler()
  1. Add a post-build action to fix an issue with the Swift XCTest support library being unnecessarily copied into your app
  • Edit your scheme in Xcode, and navigate to Build -> Post-actions
  • Click the "+" icon and select "New Run Script Action"
  • Open the "Provide build settings from" dropdown and select your target
  • Enter the following script contents:
rm "${SWIFT_STDLIB_TOOL_DESTINATION_DIR}/libswiftXCTest.dylib"

You can now use Nimble assertions in your code and handle failures as you see fit.


Download Details:

Author: Quick
Source code: https://github.com/Quick/Nimble

License: Apache-2.0 license
#swift #objective-c 

What is GEEK

Buddha Community

Nimble: A Matcher Framework for Swift and Objective-C
Sasha  Lee

Sasha Lee

1650636000

Dl4clj: Clojure Wrapper for Deeplearning4j.

dl4clj

Port of deeplearning4j to clojure

Contact info

If you have any questions,

  • my email is will@yetanalytics.com
  • I'm will_hoyt in the clojurians slack
  • twitter is @FeLungz (don't check very often)

TODO

  • update examples dir
  • finish README
    • add in examples using Transfer Learning
  • finish tests
    • eval is missing regression tests, roc tests
    • nn-test is missing regression tests
    • spark tests need to be redone
    • need dl4clj.core tests
  • revist spark for updates
  • write specs for user facing functions
    • this is very important, match isnt strict for maps
    • provides 100% certianty of the input -> output flow
    • check the args as they come in, dispatch once I know its safe, test the pure output
  • collapse overlapping api namespaces
  • add to core use case flows

Features

Stable Features with tests

  • Neural Networks DSL
  • Early Stopping Training
  • Transfer Learning
  • Evaluation
  • Data import

Features being worked on for 0.1.0

  • Clustering (testing in progress)
  • Spark (currently being refactored)
  • Front End (maybe current release, maybe future release. Not sure yet)
  • Version of dl4j is 0.0.8 in this project. Current dl4j version is 0.0.9
  • Parallelism
  • Kafka support
  • Other items mentioned in TODO

Features being worked on for future releases

  • NLP
  • Computational Graphs
  • Reinforement Learning
  • Arbiter

Artifacts

NOT YET RELEASED TO CLOJARS

  • fork or clone to try it out

If using Maven add the following repository definition to your pom.xml:

<repository>
  <id>clojars.org</id>
  <url>http://clojars.org/repo</url>
</repository>

Latest release

With Leiningen:

n/a

With Maven:

n/a

<dependency>
  <groupId>_</groupId>
  <artifactId>_</artifactId>
  <version>_</version>
</dependency>

Usage

Things you need to know

All functions for creating dl4j objects return code by default

  • All of these functions have an option to return the dl4j object
    • :as-code? = false
  • This because all builders require the code representation of dl4j objects
    • this requirement is not going to change
  • INDarray creation fns default to objects, this is for convenience
    • :as-code? is still respected

API functions return code when all args are provided as code

API functions return the value of calling the wrapped method when args are provided as a mixture of objects and code or just objects

The tests are there to help clarify behavior, if you are unsure of how to use a fn, search the tests

  • for questions about spark, refer to the spark section bellow

Example of obj/code duality

(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]))

;; as code (the default)

(l/dense-layer-builder
 :activation-fn :relu
 :learning-rate 0.006
 :weight-init :xavier
 :layer-name "example layer"
 :n-in 10
 :n-out 1)

;; =>

(doto
 (org.deeplearning4j.nn.conf.layers.DenseLayer$Builder.)
 (.nOut 1)
 (.activation (dl4clj.constants/value-of {:activation-fn :relu}))
 (.weightInit (dl4clj.constants/value-of {:weight-init :xavier}))
 (.nIn 10)
 (.name "example layer")
 (.learningRate 0.006))

;; as an object

(l/dense-layer-builder
 :activation-fn :relu
 :learning-rate 0.006
 :weight-init :xavier
 :layer-name "example layer"
 :n-in 10
 :n-out 1
 :as-code? false)

;; =>

#object[org.deeplearning4j.nn.conf.layers.DenseLayer 0x69d7d160 "DenseLayer(super=FeedForwardLayer(super=Layer(layerName=example layer, activationFn=relu, weightInit=XAVIER, biasInit=NaN, dist=null, learningRate=0.006, biasLearningRate=NaN, learningRateSchedule=null, momentum=NaN, momentumSchedule=null, l1=NaN, l2=NaN, l1Bias=NaN, l2Bias=NaN, dropOut=NaN, updater=null, rho=NaN, epsilon=NaN, rmsDecay=NaN, adamMeanDecay=NaN, adamVarDecay=NaN, gradientNormalization=null, gradientNormalizationThreshold=NaN), nIn=10, nOut=1))"]

General usage examples

Importing data

Loading data from a file (here its a csv)


(ns my.ns
 (:require [dl4clj.datasets.input-splits :as s]
           [dl4clj.datasets.record-readers :as rr]
           [dl4clj.datasets.api.record-readers :refer :all]
           [dl4clj.datasets.iterators :as ds-iter]
           [dl4clj.datasets.api.iterators :refer :all]
           [dl4clj.helpers :refer [data-from-iter]]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; file splits (convert the data to records)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def poker-path "resources/poker-hand-training.csv")
;; this is not a complete dataset, it is just here to sever as an example

(def file-split (s/new-filesplit :path poker-path))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; record readers, (read the records created by the file split)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def csv-rr (initialize-rr! :rr (rr/new-csv-record-reader :skip-n-lines 0 :delimiter ",")
                                 :input-split file-split))

;; lets look at some data
(println (next-record! :rr csv-rr :as-code? false))
;; => #object[java.util.ArrayList 0x2473e02d [1, 10, 1, 11, 1, 13, 1, 12, 1, 1, 9]]
;; this is our first line from the csv


;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; record readers dataset iterators (turn our writables into a dataset)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def rr-ds-iter (ds-iter/new-record-reader-dataset-iterator
                 :record-reader csv-rr
                 :batch-size 1
                 :label-idx 10
                 :n-possible-labels 10))

;; we use our record reader created above
;; we want to see one example per dataset obj returned (:batch-size = 1)
;; we know our label is at the last index, so :label-idx = 10
;; there are 10 possible types of poker hands so :n-possible-labels = 10
;; you can also set :label-idx to -1 to use the last index no matter the size of the seq

(def other-rr-ds-iter (ds-iter/new-record-reader-dataset-iterator
                       :record-reader csv-rr
                       :batch-size 1
                       :label-idx -1
                       :n-possible-labels 10))

(str (next-example! :iter rr-ds-iter :as-code? false))
;; =>
;;===========INPUT===================
;;[1.00, 10.00, 1.00, 11.00, 1.00, 13.00, 1.00, 12.00, 1.00, 1.00]
;;=================OUTPUT==================
;;[0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 1.00]


;; and to show that :label-idx = -1 gives us the same output

(= (next-example! :iter rr-ds-iter :as-code? false)
   (next-example! :iter other-rr-ds-iter :as-code? false)) ;; => true

INDArrays and Datasets from clojure data structures


(ns my.ns
  (:require [nd4clj.linalg.factory.nd4j :refer [vec->indarray matrix->indarray
                                                indarray-of-zeros indarray-of-ones
                                                indarray-of-rand vec-or-matrix->indarray]]
            [dl4clj.datasets.new-datasets :refer [new-ds]]
            [dl4clj.datasets.api.datasets :refer [as-list]]
            [dl4clj.datasets.iterators :refer [new-existing-dataset-iterator]]
            [dl4clj.datasets.api.iterators :refer :all]
            [dl4clj.datasets.pre-processors :as ds-pp]
            [dl4clj.datasets.api.pre-processors :refer :all]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; INDArray creation
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;;TODO: consider defaulting to code

;; can create from a vector

(vec->indarray [1 2 3 4])
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x269df212 [1.00, 2.00, 3.00, 4.00]]

;; or from a matrix

(matrix->indarray [[1 2 3 4] [2 4 6 8]])
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x20aa7fe1
;; [[1.00, 2.00, 3.00, 4.00], [2.00, 4.00, 6.00, 8.00]]]


;; will fill in spareness with zeros

(matrix->indarray [[1 2 3 4] [2 4 6 8] [10 12]])
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x8b7796c
;;[[1.00, 2.00, 3.00, 4.00],
;; [2.00, 4.00, 6.00, 8.00],
;; [10.00, 12.00, 0.00, 0.00]]]

;; can create an indarray of all zeros with specified shape
;; defaults to :rows = 1 :columns = 1

(indarray-of-zeros :rows 3 :columns 2)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x6f586a7e
;;[[0.00, 0.00],
;; [0.00, 0.00],
;; [0.00, 0.00]]]

(indarray-of-zeros) ;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0xe59ffec 0.00]

;; and if only one is supplied, will get a vector of specified length

(indarray-of-zeros :rows 2)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x2899d974 [0.00, 0.00]]

(indarray-of-zeros :columns 2)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0xa5b9782 [0.00, 0.00]]

;; same considerations/defaults for indarray-of-ones and indarray-of-rand

(indarray-of-ones :rows 2 :columns 3)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x54f08662 [[1.00, 1.00, 1.00], [1.00, 1.00, 1.00]]]

(indarray-of-rand :rows 2 :columns 3)
;; all values are greater than 0 but less than 1
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x2f20293b [[0.85, 0.86, 0.13], [0.94, 0.04, 0.36]]]



;; vec-or-matrix->indarray is built into all functions which require INDArrays
;; so that you can use clojure data structures
;; but you still have the option of passing existing INDArrays

(def example-array (vec-or-matrix->indarray [1 2 3 4]))
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x5c44c71f [1.00, 2.00, 3.00, 4.00]]

(vec-or-matrix->indarray example-array)
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x607b03b0 [1.00, 2.00, 3.00, 4.00]]

(vec-or-matrix->indarray (indarray-of-rand :rows 2))
;; => #object[org.nd4j.linalg.cpu.nativecpu.NDArray 0x49143b08 [0.76, 0.92]]

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; data-set creation
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def ds-with-single-example (new-ds :input [1 2 3 4]
                                    :output [0.0 1.0 0.0]))

(as-list :ds ds-with-single-example :as-code? false)
;; =>
;; #object[java.util.ArrayList 0x5d703d12
;;[===========INPUT===================
;;[1.00, 2.00, 3.00, 4.00]
;;=================OUTPUT==================
;;[0.00, 1.00, 0.00]]]

(def ds-with-multiple-examples (new-ds
                                :input [[1 2 3 4] [2 4 6 8]]
                                :output [[0.0 1.0 0.0] [0.0 0.0 1.0]]))

(as-list :ds ds-with-multiple-examples :as-code? false)
;; =>
;;#object[java.util.ArrayList 0x29c7a9e2
;;[===========INPUT===================
;;[1.00, 2.00, 3.00, 4.00]
;;=================OUTPUT==================
;;[0.00, 1.00, 0.00],
;;===========INPUT===================
;;[2.00, 4.00, 6.00, 8.00]
;;=================OUTPUT==================
;;[0.00, 0.00, 1.00]]]

;; we can create a dataset iterator from the code which creates datasets
;; and set the labels for our outputs (optional)

(def ds-with-multiple-examples
  (new-ds
   :input [[1 2 3 4] [2 4 6 8]]
   :output [[0.0 1.0 0.0] [0.0 0.0 1.0]]))

;; iterator
(def training-rr-ds-iter
  (new-existing-dataset-iterator
   :dataset ds-with-multiple-examples
   :labels ["foo" "baz" "foobaz"]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; data-set normalization
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; this gathers statistics on the dataset and normalizes the data
;; and applies the transformation to all dataset objects in the iterator
(def train-iter-normalized
  (c/normalize-iter! :iter training-rr-ds-iter
                     :normalizer (ds-pp/new-standardize-normalization-ds-preprocessor)
                     :as-code? false))

;; above returns the normalized iterator
;; to get fit normalizer

(def the-normalizer
  (get-pre-processor train-iter-normalized))

Model configuration

Creating a neural network configuration with singe and multiple layers

(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.nn.conf.distributions :as dist]
            [dl4clj.nn.conf.input-pre-processor :as pp]
            [dl4clj.nn.conf.step-fns :as s-fn]))

;; nn/builder has 3 types of args
;; 1) args which set network configuration params
;; 2) args which set default values for layers
;; 3) args which set multi layer network configuration params

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; single layer nn configuration
;; here we are setting network configuration
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(nn/builder :optimization-algo :stochastic-gradient-descent
            :seed 123
            :iterations 1
            :minimize? true
            :use-drop-connect? false
            :lr-score-based-decay-rate 0.002
            :regularization? false
            :step-fn :default-step-fn
            :layers {:dense-layer {:activation-fn :relu
                                   :updater :adam
                                   :adam-mean-decay 0.2
                                   :adam-var-decay 0.1
                                   :learning-rate 0.006
                                   :weight-init :xavier
                                   :layer-name "single layer model example"
                                   :n-in 10
                                   :n-out 20}})

;; there are several options within a nn-conf map which can be configuration maps
;; or calls to fns
;; It doesn't matter which option you choose and you don't have to stay consistent
;; the list of params which can be passed as config maps or fn calls will
;; be enumerated at a later date

(nn/builder :optimization-algo :stochastic-gradient-descent
            :seed 123
            :iterations 1
            :minimize? true
            :use-drop-connect? false
            :lr-score-based-decay-rate 0.002
            :regularization? false
            :step-fn (s-fn/new-default-step-fn)
            :build? true
            ;; dont need to specify layer order, theres only one
            :layers (l/dense-layer-builder
                    :activation-fn :relu
                    :updater :adam
                    :adam-mean-decay 0.2
                    :adam-var-decay 0.1
                    :dist (dist/new-normal-distribution :mean 0 :std 1)
                    :learning-rate 0.006
                    :weight-init :xavier
                    :layer-name "single layer model example"
                    :n-in 10
                    :n-out 20))

;; these configurations are the same

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; multi-layer configuration
;; here we are also setting layer defaults
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; defaults will apply to layers which do not specify those value in their config

(nn/builder
 :optimization-algo :stochastic-gradient-descent
 :seed 123
 :iterations 1
 :minimize? true
 :use-drop-connect? false
 :lr-score-based-decay-rate 0.002
 :regularization? false
 :default-activation-fn :sigmoid
 :default-weight-init :uniform

 ;; we need to specify the layer order
 :layers {0 (l/activation-layer-builder
             :activation-fn :relu
             :updater :adam
             :adam-mean-decay 0.2
             :adam-var-decay 0.1
             :learning-rate 0.006
             :weight-init :xavier
             :layer-name "example first layer"
             :n-in 10
             :n-out 20)
          1 {:output-layer {:n-in 20
                            :n-out 2
                            :loss-fn :mse
                            :layer-name "example output layer"}}})

;; specifying multi-layer config params

(nn/builder
 ;; network args
 :optimization-algo :stochastic-gradient-descent
 :seed 123
 :iterations 1
 :minimize? true
 :use-drop-connect? false
 :lr-score-based-decay-rate 0.002
 :regularization? false

 ;; layer defaults
 :default-activation-fn :sigmoid
 :default-weight-init :uniform

 ;; the layers
 :layers {0 (l/activation-layer-builder
             :activation-fn :relu
             :updater :adam
             :adam-mean-decay 0.2
             :adam-var-decay 0.1
             :learning-rate 0.006
             :weight-init :xavier
             :layer-name "example first layer"
             :n-in 10
             :n-out 20)
          1 {:output-layer {:n-in 20
                            :n-out 2
                            :loss-fn :mse
                            :layer-name "example output layer"}}}
 ;; multi layer network args
 :backprop? true
 :backprop-type :standard
 :pretrain? false
 :input-pre-processors {0 (pp/new-zero-mean-pre-pre-processor)
                        1 {:unit-variance-processor {}}})

Configuration to Trained models

Multi Layer models

(ns my.ns
  (:require [dl4clj.datasets.iterators :as iter]
            [dl4clj.datasets.input-splits :as split]
            [dl4clj.datasets.record-readers :as rr]
            [dl4clj.optimize.listeners :as listener]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.nn.multilayer.multi-layer-network :as mln]
            [dl4clj.nn.api.model :refer [init! set-listeners!]]
            [dl4clj.nn.api.multi-layer-network :refer [evaluate-classification]]
            [dl4clj.datasets.api.record-readers :refer [initialize-rr!]]
            [dl4clj.eval.api.eval :refer [get-stats get-accuracy]]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; nn-conf -> multi-layer-network
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def nn-conf
  (nn/builder
   ;; network args
   :optimization-algo :stochastic-gradient-descent
   :seed 123 :iterations 1 :regularization? true

   ;; setting layer defaults
   :default-activation-fn :relu :default-l2 7.5e-6
   :default-weight-init :xavier :default-learning-rate 0.0015
   :default-updater :nesterovs :default-momentum 0.98

   ;; setting layer configuration
   :layers {0 {:dense-layer
               {:layer-name "example first layer"
                :n-in 784 :n-out 500}}
            1 {:dense-layer
               {:layer-name "example second layer"
                :n-in 500 :n-out 100}}
            2 {:output-layer
               {:n-in 100 :n-out 10
                ;; layer specific params
                :loss-fn :negativeloglikelihood
                :activation-fn :softmax
                :layer-name "example output layer"}}}

   ;; multi layer args
   :backprop? true
   :pretrain? false))

(def multi-layer-network (c/model-from-conf nn-conf))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; local cpu training with dl4j pre-built iterators
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; lets use the pre-built Mnist data set iterator

(def train-mnist-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? true
   :seed 123))

(def test-mnist-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? false
   :seed 123))

;; and lets set a listener so we can know how training is going

(def score-listener (listener/new-score-iteration-listener :print-every-n 5))

;; and attach it to our model

;; TODO: listeners are broken, look into log4j warnning
(def mln-with-listener (set-listeners! :model multi-layer-network
                                       :listeners [score-listener]))

(def trained-mln (mln/train-mln-with-ds-iter! :mln mln-with-listener
                                              :iter train-mnist-iter
                                              :n-epochs 15
                                              :as-code? false))

;; training happens because :as-code? = false
;; if it was true, we would still just have a data structure
;; we now have a trained model that has seen the training dataset 15 times
;; time to evaluate our model

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;Create an evaluation object
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def eval-obj (evaluate-classification :mln trained-mln
                                       :iter test-mnist-iter))

;; always remember that these objects are stateful, dont use the same eval-obj
;; to eval two different networks
;; we trained the model on a training dataset.  We evaluate on a test set

(println (get-stats :evaler eval-obj))
;; this will print the stats to standard out for each feature/label pair

;;Examples labeled as 0 classified by model as 0: 968 times
;;Examples labeled as 0 classified by model as 1: 1 times
;;Examples labeled as 0 classified by model as 2: 1 times
;;Examples labeled as 0 classified by model as 3: 1 times
;;Examples labeled as 0 classified by model as 5: 1 times
;;Examples labeled as 0 classified by model as 6: 3 times
;;Examples labeled as 0 classified by model as 7: 1 times
;;Examples labeled as 0 classified by model as 8: 2 times
;;Examples labeled as 0 classified by model as 9: 2 times
;;Examples labeled as 1 classified by model as 1: 1126 times
;;Examples labeled as 1 classified by model as 2: 2 times
;;Examples labeled as 1 classified by model as 3: 1 times
;;Examples labeled as 1 classified by model as 5: 1 times
;;Examples labeled as 1 classified by model as 6: 2 times
;;Examples labeled as 1 classified by model as 7: 1 times
;;Examples labeled as 1 classified by model as 8: 2 times
;;Examples labeled as 2 classified by model as 0: 3 times
;;Examples labeled as 2 classified by model as 1: 2 times
;;Examples labeled as 2 classified by model as 2: 1006 times
;;Examples labeled as 2 classified by model as 3: 2 times
;;Examples labeled as 2 classified by model as 4: 3 times
;;Examples labeled as 2 classified by model as 6: 3 times
;;Examples labeled as 2 classified by model as 7: 7 times
;;Examples labeled as 2 classified by model as 8: 6 times
;;Examples labeled as 3 classified by model as 2: 4 times
;;Examples labeled as 3 classified by model as 3: 990 times
;;Examples labeled as 3 classified by model as 5: 3 times
;;Examples labeled as 3 classified by model as 7: 3 times
;;Examples labeled as 3 classified by model as 8: 3 times
;;Examples labeled as 3 classified by model as 9: 7 times
;;Examples labeled as 4 classified by model as 2: 2 times
;;Examples labeled as 4 classified by model as 3: 1 times
;;Examples labeled as 4 classified by model as 4: 967 times
;;Examples labeled as 4 classified by model as 6: 4 times
;;Examples labeled as 4 classified by model as 7: 1 times
;;Examples labeled as 4 classified by model as 9: 7 times
;;Examples labeled as 5 classified by model as 0: 2 times
;;Examples labeled as 5 classified by model as 3: 6 times
;;Examples labeled as 5 classified by model as 4: 1 times
;;Examples labeled as 5 classified by model as 5: 874 times
;;Examples labeled as 5 classified by model as 6: 3 times
;;Examples labeled as 5 classified by model as 7: 1 times
;;Examples labeled as 5 classified by model as 8: 3 times
;;Examples labeled as 5 classified by model as 9: 2 times
;;Examples labeled as 6 classified by model as 0: 4 times
;;Examples labeled as 6 classified by model as 1: 3 times
;;Examples labeled as 6 classified by model as 3: 2 times
;;Examples labeled as 6 classified by model as 4: 4 times
;;Examples labeled as 6 classified by model as 5: 4 times
;;Examples labeled as 6 classified by model as 6: 939 times
;;Examples labeled as 6 classified by model as 7: 1 times
;;Examples labeled as 6 classified by model as 8: 1 times
;;Examples labeled as 7 classified by model as 1: 7 times
;;Examples labeled as 7 classified by model as 2: 4 times
;;Examples labeled as 7 classified by model as 3: 3 times
;;Examples labeled as 7 classified by model as 7: 1005 times
;;Examples labeled as 7 classified by model as 8: 2 times
;;Examples labeled as 7 classified by model as 9: 7 times
;;Examples labeled as 8 classified by model as 0: 3 times
;;Examples labeled as 8 classified by model as 2: 3 times
;;Examples labeled as 8 classified by model as 3: 2 times
;;Examples labeled as 8 classified by model as 4: 4 times
;;Examples labeled as 8 classified by model as 5: 3 times
;;Examples labeled as 8 classified by model as 6: 2 times
;;Examples labeled as 8 classified by model as 7: 4 times
;;Examples labeled as 8 classified by model as 8: 947 times
;;Examples labeled as 8 classified by model as 9: 6 times
;;Examples labeled as 9 classified by model as 0: 2 times
;;Examples labeled as 9 classified by model as 1: 2 times
;;Examples labeled as 9 classified by model as 3: 4 times
;;Examples labeled as 9 classified by model as 4: 8 times
;;Examples labeled as 9 classified by model as 6: 1 times
;;Examples labeled as 9 classified by model as 7: 4 times
;;Examples labeled as 9 classified by model as 8: 2 times
;;Examples labeled as 9 classified by model as 9: 986 times

;;==========================Scores========================================
;; Accuracy:        0.9808
;; Precision:       0.9808
;; Recall:          0.9807
;; F1 Score:        0.9807
;;========================================================================

;; can get the stats that are printed via fns in the evaluation namespace
;; after running eval-model-whole-ds

(get-accuracy :evaler evaler-with-stats) ;; => 0.9808

Model Tuning

Early Stopping (controlling training)

it is recommened you start here when designing models

using dl4clj.core


(ns my.ns
  (:require [dl4clj.earlystopping.termination-conditions :refer :all]
            [dl4clj.earlystopping.model-saver :refer [new-in-memory-saver]]
            [dl4clj.nn.api.multi-layer-network :refer [evaluate-classification]]
            [dl4clj.eval.api.eval :refer [get-stats]]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.datasets.iterators :as iter]
            [dl4clj.core :as c]))

(def nn-conf
  (nn/builder
   ;; network args
   :optimization-algo :stochastic-gradient-descent
   :seed 123
   :iterations 1
   :regularization? true

   ;; setting layer defaults
   :default-activation-fn :relu
   :default-l2 7.5e-6
   :default-weight-init :xavier
   :default-learning-rate 0.0015
   :default-updater :nesterovs
   :default-momentum 0.98

   ;; setting layer configuration
   :layers {0 {:dense-layer
               {:layer-name "example first layer"
                :n-in 784 :n-out 500}}
            1 {:dense-layer
               {:layer-name "example second layer"
                :n-in 500 :n-out 100}}
            2 {:output-layer
               {:n-in 100 :n-out 10
                ;; layer specific params
                :loss-fn :negativeloglikelihood
                :activation-fn :softmax
                :layer-name "example output layer"}}}

   ;; multi layer args
   :backprop? true
   :pretrain? false))

(def train-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? true
   :seed 123))

(def test-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? false
   :seed 123))

(def invalid-score-condition (new-invalid-score-iteration-termination-condition))

(def max-score-condition (new-max-score-iteration-termination-condition
                          :max-score 20.0))

(def max-time-condition (new-max-time-iteration-termination-condition
                         :max-time-val 10
                         :max-time-unit :minutes))

(def score-doesnt-improve-condition (new-score-improvement-epoch-termination-condition
                                     :max-n-epoch-no-improve 5))

(def target-score-condition (new-best-score-epoch-termination-condition
                             :best-expected-score 0.009))

(def max-number-epochs-condition (new-max-epochs-termination-condition :max-n 20))

(def in-mem-saver (new-in-memory-saver))

(def trained-mln
;; defaults to returning the model
  (c/train-with-early-stopping
   :nn-conf nn-conf
   :training-iter train-mnist-iter
   :testing-iter test-mnist-iter
   :eval-every-n-epochs 1
   :iteration-termination-conditions [invalid-score-condition
                                      max-score-condition
                                      max-time-condition]
   :epoch-termination-conditions [score-doesnt-improve-condition
                                  target-score-condition
                                  max-number-epochs-condition]
   :save-last-model? true
   :model-saver in-mem-saver
   :as-code? false))

(def model-evaler
  (evaluate-classification :mln trained-mln :iter test-mnist-iter))

(println (get-stats :evaler model-evaler))
  • explicit, step by step way of doing this
(ns my.ns
  (:require [dl4clj.earlystopping.early-stopping-config :refer [new-early-stopping-config]]
            [dl4clj.earlystopping.termination-conditions :refer :all]
            [dl4clj.earlystopping.model-saver :refer [new-in-memory-saver new-local-file-model-saver]]
            [dl4clj.earlystopping.score-calc :refer [new-ds-loss-calculator]]
            [dl4clj.earlystopping.early-stopping-trainer :refer [new-early-stopping-trainer]]
            [dl4clj.earlystopping.api.early-stopping-trainer :refer [fit-trainer!]]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.nn.multilayer.multi-layer-network :as mln]
            [dl4clj.utils :refer [load-model!]]
            [dl4clj.datasets.iterators :as iter]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; start with our network config
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def nn-conf
  (nn/builder
   ;; network args
   :optimization-algo :stochastic-gradient-descent
   :seed 123 :iterations 1 :regularization? true
   ;; setting layer defaults
   :default-activation-fn :relu :default-l2 7.5e-6
   :default-weight-init :xavier :default-learning-rate 0.0015
   :default-updater :nesterovs :default-momentum 0.98
   ;; setting layer configuration
   :layers {0 {:dense-layer
               {:layer-name "example first layer"
                :n-in 784 :n-out 500}}
            1 {:dense-layer
               {:layer-name "example second layer"
                :n-in 500 :n-out 100}}
            2 {:output-layer
               {:n-in 100 :n-out 10
                ;; layer specific params
                :loss-fn :negativeloglikelihood
                :activation-fn :softmax
                :layer-name "example output layer"}}}
   ;; multi layer args
   :backprop? true
   :pretrain? false))

(def mln (c/model-from-conf nn-conf))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; the training/testing data
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def train-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? true
   :seed 123))

(def test-iter
  (iter/new-mnist-data-set-iterator
   :batch-size 64
   :train? false
   :seed 123))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; we are going to need termination conditions
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; these allow us to control when we exit training

;; this can be based off of iterations or epochs

;; iteration termination conditions

(def invalid-score-condition (new-invalid-score-iteration-termination-condition))

(def max-score-condition (new-max-score-iteration-termination-condition
                          :max-score 20.0))

(def max-time-condition (new-max-time-iteration-termination-condition
                         :max-time-val 10
                         :max-time-unit :minutes))

;; epoch termination conditions

(def score-doesnt-improve-condition (new-score-improvement-epoch-termination-condition
                                     :max-n-epoch-no-improve 5))

(def target-score-condition (new-best-score-epoch-termination-condition :best-expected-score 0.009))

(def max-number-epochs-condition (new-max-epochs-termination-condition :max-n 20))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; we also need a way to save our model
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; can be in memory or to a local directory

(def in-mem-saver (new-in-memory-saver))

(def local-file-saver (new-local-file-model-saver :directory "resources/tmp/readme/"))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; set up your score calculator
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def score-calcer (new-ds-loss-calculator :iter test-iter
                                          :average? true))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; create an early stopping configuration
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; termination conditions
;; a way to save our model
;; a way to calculate the score of our model on the dataset

(def early-stopping-conf
  (new-early-stopping-config
   :epoch-termination-conditions [score-doesnt-improve-condition
                                  target-score-condition
                                  max-number-epochs-condition]
   :iteration-termination-conditions [invalid-score-condition
                                      max-score-condition
                                      max-time-condition]
   :eval-every-n-epochs 5
   :model-saver local-file-saver
   :save-last-model? true
   :score-calculator score-calcer))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; create an early stopping trainer from our data, model and early stopping conf
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def es-trainer (new-early-stopping-trainer :early-stopping-conf early-stopping-conf
                                            :mln mln
                                            :iter train-iter))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; fit and use our early stopping trainer
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def es-trainer-fitted (fit-trainer! es-trainer :as-code? false))

;; when the trainer terminates, you will see something like this
;;[nREPL-worker-24] BaseEarlyStoppingTrainer INFO  Completed training epoch 14
;;[nREPL-worker-24] BaseEarlyStoppingTrainer INFO  New best model: score = 0.005225599372851298,
;;                                                   epoch = 14 (previous: score = 0.018243224899038346, epoch = 7)
;;[nREPL-worker-24] BaseEarlyStoppingTrainer INFO Hit epoch termination condition at epoch 14.
;;                                           Details: BestScoreEpochTerminationCondition(0.009)

;; and if we look at the es-trainer-fitted object we see

;;#object[org.deeplearning4j.earlystopping.EarlyStoppingResult 0x5ab74f27 EarlyStoppingResult
;;(terminationReason=EpochTerminationCondition,details=BestScoreEpochTerminationCondition(0.009),
;; bestModelEpoch=14,bestModelScore=0.005225599372851298,totalEpochs=15)]

;; and our model has been saved to /resources/tmp/readme/bestModel.bin
;; there we have our model config, model params and our updater state

;; we can then load this model to use it or continue refining it

(def loaded-model (load-model! :path "resources/tmp/readme/bestModel.bin"
                               :load-updater? true))

Transfer Learning (freezing layers)


;; TODO: need to write up examples

Spark Training

dl4j Spark usage

How it is done in dl4clj

  • Uses dl4clj.core
    • This example uses a fn which takes care of most steps for you
      • allows you to pass args as code bc the fn accounts for the multiple spark contexts issue encountered when everything is just a data structure

(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.datasets.iterators :refer [new-iris-data-set-iterator]]
            [dl4clj.eval.api.eval :refer [get-stats]]
            [dl4clj.spark.masters.param-avg :as master]
            [dl4clj.spark.data.java-rdd :refer [new-java-spark-context
                                                java-rdd-from-iter]]
            [dl4clj.spark.api.dl4j-multi-layer :refer [eval-classification-spark-mln
                                                       get-spark-context]]
            [dl4clj.core :as c]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 1, create your model config
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def mln-conf
  (nn/builder
   :optimization-algo :stochastic-gradient-descent
   :default-learning-rate 0.006
   :layers {0 (l/dense-layer-builder :n-in 4 :n-out 2 :activation-fn :relu)
            1 {:output-layer
               {:loss-fn :negativeloglikelihood
                :n-in 2 :n-out 3
                :activation-fn :soft-max
                :weight-init :xavier}}}
   :backprop? true
   :backprop-type :standard))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 2, training master
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def training-master
  (master/new-parameter-averaging-training-master
   :build? true
   :rdd-n-examples 10
   :n-workers 4
   :averaging-freq 10
   :batch-size-per-worker 2
   :export-dir "resources/spark/master/"
   :rdd-training-approach :direct
   :repartition-data :always
   :repartition-strategy :balanced
   :seed 1234
   :save-updater? true
   :storage-level :none))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 3, spark context
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def your-spark-context
  (new-java-spark-context :app-name "example app"))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 4, training data
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def iris-iter
  (new-iris-data-set-iterator
   :batch-size 1
   :n-examples 5))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 5, spark mln
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def fitted-spark-mln
  (c/train-with-spark :spark-context your-spark-context
                      :mln-conf mln-conf
                      :training-master training-master
                      :iter iris-iter
                      :n-epochs 1
                      :as-code? false))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 5, use spark context from spark-mln to create rdd
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; TODO: eliminate this step

(def our-rdd
  (let [sc (get-spark-context fitted-spark-mln :as-code? false)]
    (java-rdd-from-iter :spark-context sc
                        :iter iris-iter)))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 6, evaluation model and print stats (poor performance of model expected)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def eval-obj
  (eval-classification-spark-mln
   :spark-mln fitted-spark-mln
   :rdd our-rdd))

(println (get-stats :evaler eval-obj))

  • this example demonstrates the dl4j workflow
    • NOTE: unlike the previous example, this one requires dl4j objects to be used
      • this is becaues spark only wants you to have one spark context at a time
(ns my.ns
  (:require [dl4clj.nn.conf.builders.layers :as l]
            [dl4clj.nn.conf.builders.nn :as nn]
            [dl4clj.datasets.iterators :refer [new-iris-data-set-iterator]]
            [dl4clj.eval.api.eval :refer [get-stats]]
            [dl4clj.spark.masters.param-avg :as master]
            [dl4clj.spark.data.java-rdd :refer [new-java-spark-context java-rdd-from-iter]]
            [dl4clj.spark.dl4j-multi-layer :as spark-mln]
            [dl4clj.spark.api.dl4j-multi-layer :refer [fit-spark-mln!
                                                       eval-classification-spark-mln]]))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 1, create your model
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def mln-conf
  (nn/builder
   :optimization-algo :stochastic-gradient-descent
   :default-learning-rate 0.006
   :layers {0 (l/dense-layer-builder :n-in 4 :n-out 2 :activation-fn :relu)
            1 {:output-layer
               {:loss-fn :negativeloglikelihood
                :n-in 2 :n-out 3
                :activation-fn :soft-max
                :weight-init :xavier}}}
   :backprop? true
   :as-code? false
   :backprop-type :standard))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 2, create a training master
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; not all options specified, but most are

(def training-master
  (master/new-parameter-averaging-training-master
   :build? true
   :rdd-n-examples 10
   :n-workers 4
   :averaging-freq 10
   :batch-size-per-worker 2
   :export-dir "resources/spark/master/"
   :rdd-training-approach :direct
   :repartition-data :always
   :repartition-strategy :balanced
   :seed 1234
   :as-code? false
   :save-updater? true
   :storage-level :none))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 3, create a Spark Multi Layer Network
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def your-spark-context
  (new-java-spark-context :app-name "example app" :as-code? false))

;; new-java-spark-context will turn an existing spark-configuration into a java spark context
;; or create a new java spark context with master set to "local[*]" and the app name
;; set to :app-name


(def spark-mln
  (spark-mln/new-spark-multi-layer-network
   :spark-context your-spark-context
   :mln mln-conf
   :training-master training-master
   :as-code? false))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 4, load your data
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;; one way is via a dataset-iterator
;; can make one directly from a dataset (iterator data-set)
;; see: nd4clj.linalg.dataset.api.data-set and nd4clj.linalg.dataset.data-set
;; we are going to use a pre-built one

(def iris-iter
  (new-iris-data-set-iterator
   :batch-size 1
   :n-examples 5
   :as-code? false))

;; now lets convert the data into a javaRDD

(def our-rdd
  (java-rdd-from-iter :spark-context your-spark-context
                      :iter iris-iter))

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;; Step 5, fit and evaluate the model
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

(def fitted-spark-mln
  (fit-spark-mln!
   :spark-mln spark-mln
   :rdd our-rdd
   :n-epochs 1))
;; this fn also has the option to supply :path-to-data instead of :rdd
;; that path should point to a directory containing a number of dataset objects

(def eval-obj
  (eval-classification-spark-mln
   :spark-mln fitted-spark-mln
   :rdd our-rdd))
;; we would want to have different testing and training rdd's but here we are using
;; the data we trained on

;; lets get the stats for how our model performed

(println (get-stats :evaler eval-obj))

Terminology

Coming soon

Packages to come back to:

Implement ComputationGraphs and the classes which use them

NLP

Parallelism

TSNE

UI


Author: yetanalytics
Source Code: https://github.com/yetanalytics/dl4clj
License: BSD-2-Clause License

#machine-learning #deep-learning 

Arvel  Parker

Arvel Parker

1591611780

How to Find Ulimit For user on Linux

How can I find the correct ulimit values for a user account or process on Linux systems?

For proper operation, we must ensure that the correct ulimit values set after installing various software. The Linux system provides means of restricting the number of resources that can be used. Limits set for each Linux user account. However, system limits are applied separately to each process that is running for that user too. For example, if certain thresholds are too low, the system might not be able to server web pages using Nginx/Apache or PHP/Python app. System resource limits viewed or set with the NA command. Let us see how to use the ulimit that provides control over the resources available to the shell and processes.

#[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object]

MEAN Stack Tutorial MongoDB ExpressJS AngularJS NodeJS

We are going to build a full stack Todo App using the MEAN (MongoDB, ExpressJS, AngularJS and NodeJS). This is the last part of three-post series tutorial.

MEAN Stack tutorial series:

AngularJS tutorial for beginners (Part I)
Creating RESTful APIs with NodeJS and MongoDB Tutorial (Part II)
MEAN Stack Tutorial: MongoDB, ExpressJS, AngularJS and NodeJS (Part III) 👈 you are here
Before completing the app, let’s cover some background about the this stack. If you rather jump to the hands-on part click here to get started.

#[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object]

Yoshiko  Jones

Yoshiko Jones

1598195340

How to configure AWS SES with Postfix MTA

How do I configure Amazon SES With Postfix mail server to send email under a CentOS/RHEL/Fedora/Ubuntu/Debian Linux server?

Amazon Simple Email Service (SES) is a hosted email service for you to send and receive email using your email addresses and domains. Typically SES used for sending bulk email or routing emails without hosting MTA. We can use Perl/Python/PHP APIs to send an email via SES. Another option is to configure Linux or Unix box running Postfix to route all outgoing emails via SES.

  • » Remove sendmail
  • » Install postfix
  • » Configuring postfix for SES
  • » Test postfix

Procedure to configure AWS SES with Postfix

Before getting started with Amazon SES and Postfix, you need to sign up for AWS, including SES. You need to verify your email address and other settings. Make sure you create a user for SES access and download credentials too.

Step 1 – Uninstall Sendmail if installed

If sendmail installed remove it. Debian/Ubuntu Linux user type the following apt command/apt-get command:

$`` sudo apt --purge remove sendmail

CentOS/RHEL user type the following yum command or dnf command on Fedora/CentOS/RHEL 8.x:

$`` sudo yum remove sendmail

$`` sudo dnf remove sendmail

Sample outputs from CentOS 8 server:

Dependencies resolved.
===============================================================================
 Package           Architecture  Version               Repository         Size
===============================================================================
Removing:
 sendmail          x86_64        8.15.2-32.el8         @AppStream        2.4 M
Removing unused dependencies:
 cyrus-sasl        x86_64        2.1.27-1.el8          @BaseOS           160 k
 procmail          x86_64        3.22-47.el8           @AppStream        369 k

Transaction Summary
===============================================================================
Remove  3 Packages

Freed space: 2.9 M
Is this ok [y/N]: y

#[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object]

Brain  Crist

Brain Crist

1595434320

Docker Applikationen mit Visual Studio Code debuggen

Mit dem integrierten Debugger von Visual Studio Code lassen sich ASP.NET Core bzw. .NET Core Applikationen einfach und problemlos debuggen. Der Debugger unterstützt auch Remote Debugging, somit lassen sich zum Beispiel .NET Core Programme, die in einem Docker-Container laufen, debuggen.

Als Beispiel Applikation reicht das Default-Template für MVC Applikationen dotnet new mvc

$ md docker-core-debugger
$ cd docker-core-debugger
$ dotnet new mvc

Mit dotnet run prüfen wir kurz, ob die Applikation läuft und unter der Adresse http://localhost:5000 erreichbar ist.

$ dotnet run
$ Hosting environment: Production
$ Content root path: D:\Temp\docker-aspnetcore
$ Now listening on: http://localhost:5000

Die .NET Core Applikation builden wir mit dotnet build und publishen alles mit Hilfe von dotnet publish

$ dotnet build
$ dotnet publish -c Debug -o out --runtime linux-x64

Dabei gilt es zu beachten, dass die Build Configuration mit -c Debug gesetzt ist und das Output Directory auf -o out. Sonst findet Docker die nötigen Binaries nicht. Für den Docker Container brauchen wir nun ein Dockerfile, dass beim Start vorgängig den .NET Core command line debugger (VSDBG) installiert. Das Installations-Script für VSDBG ist unter https://aka.ms/getvsdbgsh abfrufbar.

FROM microsoft/aspnetcore:latest
WORKDIR /app

RUN apt-get update \
    && apt-get install -y --no-install-recommends \
       unzip procps \
    && rm -rf /var/lib/apt/lists/* \
    && curl -sSL https://aka.ms/getvsdbgsh | bash /dev/stdin -v latest -l /vsdbg

COPY ./out .
ENTRYPOINT ["dotnet", "docker-core-debugger.dll"]

Den Docker Container erstellen wir mit dem docker build Kommando

$ docker build -t coreapp .

und starten die Applikation mit docker run.

$ docker run -d -p 8080:80 --name coreapp coreapp

Jetzt muss Visual Studio Code nur noch wissen, wo unsere Applikation läuft. Dazu definieren wir eine launch.json vom Typ attach und konfigurieren die nötigen Parameter für den Debugger.

{
    "version": "0.2.0",
    "configurations": [
         {
            "name": ".NET Core Remote Attach",
            "type": "coreclr",
            "request": "attach",
            "processId": "${command:pickRemoteProcess}",
            "pipeTransport": {
                "pipeProgram": "docker",
                "pipeArgs": ["exec", "-i coreapp ${debuggerCommand}"],
                "quoteArgs": false,
                "debuggerPath": "/vsdbg/vsdbg",
                "pipeCwd": "${workspaceRoot}"
            },

            "logging": {
                "engineLogging": true,
                "exceptions": true,
                "moduleLoad": true,
                "programOutput": true
            },
        }
    ]
}

Mit F5 starten wir den Debugger. Wenn alles klappt, sollte eine Auswahl der Prozesse des Docker-Containers sichtbar sein.

vscode

Nun muss der dotnet Prozess ausgewählt werden. Der Visual Studio Code Debugger verbindet sich darauf mit VSDBG und wir können wie gewohnt unseren Code debuggen. Dazu setzen wir einen Breakpoint in der Index-Action des HomeControllers und rufen mit dem Browser die URL http://localhost:8080/ auf.

vscode

#[object object] #[object object] #[object object] #[object object] #[object object] #[object object] #[object object]