ó ¾÷Xc@`sdZddlmZmZmZdZddlZddlZejdƒZ ej ƒZ ej ddƒZ e je ƒe je ƒe jejƒddlmZdd lmZd Zdd lmZmZmZmZmZmZmZm Z m!Z!m"Z"m#Z#m$Z$m%Z%m&Z&m'Z'm(Z(m)Z)m*Z*m+Z+m,Z,m-Z-dd l.m/Z/m0Z0m1Z1m2Z2m3Z3m4Z4m5Z5m6Z6m7Z7m8Z8m9Z9m:Z:m;Z;m<Z<m=Z=m>Z>dd l?m@Z@ddlAmBZBmCZCddlDmEZEmFZFmGZGmHZHmIZImJZJmKZKddlLmMZMddlNmOZOmPZPmQZQmRZRddlSZTeUeTjVdƒrneTjVjWƒjXZXn d„ZXejYjZdƒs›ej[jZdƒrêddl\ZTeTj]j^j_rêddl`ZTejarçeTj]j^jVjbjcƒqçqênejYjZdƒsAejYjZdƒsAej[jZdƒsAej[jZdƒsAejddkrPddleZTnddlfZfejfjgdkrwdZin ejfjgZiejfjjdkrždZkn ejfjjZkejfjldkrÅdZmn ejfjlZmejfjndkrìdZon ejfjnZoejfjpdkrdZqn ejfjpZqefjrdeidekdemdeodeqƒ[i[k[m[o[qd„Zsd„Ztd „Zuevd!ƒdS("sw Theano is an optimizing compiler in Python, built to evaluate complicated expressions (especially matrix-valued ones) as quickly as possible. Theano compiles expression graphs (see :doc:`graph` ) that are built by Python code. The expressions in these graphs are called `Apply` nodes and the variables in these graphs are called `Variable` nodes. You compile a graph by calling `function`, which takes a graph, and returns a callable object. One of theano's most important features is that `function` can transform your graph before compiling it. It can replace simple expressions with faster or more numerically stable implementations. To learn more, check out: - Op List (:doc:`oplist`) The markup language used in the docstrings is ReStructured Text, which may be rendered with Sphinx. A rendered version is maintained at http://www.deeplearning.net/software/theano/library/ i(tabsolute_importtprint_functiontdivisionsrestructuredtext enNttheanotfmts%%(levelname)s (%(name)s): %(message)s(tversion(tconfigi(tCLinkert OpWiseCLinkert DualLinkertLinkert LocalLinkert PerformLinkert ContainertInconsistencyErrort FunctionGraphtApplytVariabletConstanttOptOpenMPOptoptttoolboxtTypetGenerictgenerictobject2tutils(t SymbolicInputtIntSymbolicOutputtOuttModetpredefined_modestpredefined_linkerstpredefined_optimizerst FunctionMakertfunctiont function_dumpt OpFromGrapht ProfileStatstParamtsharedtas_op(t_asarray(tpprinttpp(tscantmaptreducetfoldltfoldrtclonetscan_checkpoints(tOrderedUpdates(tRoptLoptgradt subgraph_gradtTheanoNoseTestercC`stdƒ‚dS(Ns@The nose module is not installed. It is needed for Theano tests.(t ImportError(((s//tmp/pip-build-isqEY4/theano/theano/__init__.pyttestgstgputcudatopenclttNonetalltdividetovertundertinvalidcC`sÔt}d\}}|tkr_t|dƒr_y|j|ƒ}Wq_tk r[}t}q_Xn|tkr¬t|dƒr¬y|j|ƒ}Wq¬tk r¨}t}q¬Xn|tkrÐtd||fƒ‚n|S(s5Return a symbolic matrix/dot product between l and r t__dot__t__rdot__s%Dot failed for the following reasons:N(NN(tNotImplementedRBthasattrRHt ExceptionRItNotImplementedError(tltrtrvalte0te1((s//tmp/pip-build-isqEY4/theano/theano/__init__.pytdot s     cC`sxdtƒkrkt|jtjƒrk|jdk rkt|jjtjƒrk|jj d}t j |ƒSnt j |ƒS(sUreturn the constant scalar(0-D) value underlying variable `v` If v is the output of dimshuffles, fills, allocs, rebroadcasts, cast this function digs through them. If theano.sparse is also there, we will look over CSM op. If `v` is not some view of constant data, then raise a tensor.basic.NotScalarConstantError. tsparseiN( tglobalst isinstancettypeRTt SparseTypetownerRBtoptCSMtinputsttensortget_scalar_constant_value(tvtdata((s//tmp/pip-build-isqEY4/theano/theano/__init__.pyR^µs $'cC`sFt|jjtjƒst‚|jjjdtƒ|jjŒ}|S(süThis function return a new variable whose gradient will be stored in a sparse format instead of dense. Currently only variable created by AdvancedSubtensor1 is supported. i.e. a_tensor_var[an_int_vector]. .. versionadded:: 0.6rc4 t sparse_grad( RVRYRZR]tAdvancedSubtensor1tAssertionErrort __class__tTrueR\(tvartret((s//tmp/pip-build-isqEY4/theano/theano/__init__.pyRaÈs $s"theano.tensor.shared_randomstreams(wt__doc__t __future__RRRt __docformat__tloggingtsyst getLoggert theano_loggert StreamHandlertlogging_default_handlert Formattertlogging_default_formattert setFormattert addHandlertsetLeveltWARNINGttheano.versionRt __version__ttheano.configdefaultsRt__api_version__t theano.gofRRR R R R R RRRRRRRRRRRRRRttheano.compileRRRRR R!R"R#R$R%R&R'R(R)R*R+ttheano.misc.safe_asarrayR,ttheano.printingR-R.ttheano.scan_moduleR/R0R1R2R3R4R5ttheano.updatesR6ttheano.gradientR7R8R9R:t theano.testsRRKttestsR;R=tdevicet startswithtinit_gpu_devicettheano.sandbox.cudatsandboxR?tcuda_availablet%theano.sandbox.cuda.tests.test_drivertenable_initial_driver_testt test_driverttest_nvidia_driver1tcontextsttheano.gpuarraytnumpyt seterr_allRBt_allt seterr_dividet_dividet seterr_overt_overt seterr_undert_undertseterr_invalidt_invalidtseterrRSR^Rat __import__(((s//tmp/pip-build-isqEY4/theano/theano/__init__.pyts€       ˆ j 4 "  $