Skip to content

Instantly share code, notes, and snippets.

@oak-tree

oak-tree/gist:ccec4bf5ec0931c29a11629e1a0f9d46

Created February 16, 2017 12:19
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save oak-tree/ccec4bf5ec0931c29a11629e1a0f9d46 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save oak-tree/ccec4bf5ec0931c29a11629e1a0f9d46 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
gistfile1.txt
1 #include <Python.h>
2 #include <iostream>
3 #include "theano_mod_helper.h"
4 #include "cuda_ndarray.cuh"
5 //////////////////////
6 //// Support Code
7 //////////////////////
8
9
10 namespace {
11 struct __struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604 {
12 PyObject* __ERROR;
13
14 PyObject* storage_V3;
15 PyObject* storage_V1;
16
17
18 __struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604() {
19 // This is only somewhat safe because we:
20 // 1) Are not a virtual class
21 // 2) Do not use any virtual classes in the members
22 // 3) Deal with mostly POD and pointers
23
24 // If this changes, we would have to revise this, but for
25 // now I am tired of chasing segfaults because
26 // initialization code had an error and some pointer has
27 // a junk value.
28 memset(this, 0, sizeof(*this));
29 }
30 ~__struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604(void) {
31 cleanup();
32 }
33
34 int init(PyObject* __ERROR, PyObject* storage_V3, PyObject* storage_V1) {
35 Py_XINCREF(storage_V3);
36 Py_XINCREF(storage_V1);
37 this->storage_V3 = storage_V3;
38 this->storage_V1 = storage_V1;
39
40
41
42
43 this->__ERROR = __ERROR;
44 return 0;
45 }
46 void cleanup(void) {
47 __label_1:
48
49 double __DUMMY_1;
50 __label_3:
51
52 double __DUMMY_3;
53 __label_6:
54
55 double __DUMMY_6;
56
57 Py_XDECREF(this->storage_V3);
58 Py_XDECREF(this->storage_V1);
59 }
60 int run(void) {
61 int __failure = 0;
62
63 PyObject* py_V1;
64 CudaNdarray * V1;
65 PyObject* py_V3;
66 CudaNdarray * V3;
67 {
68
69 py_V1 = PyList_GET_ITEM(storage_V1, 0);
70 {Py_XINCREF(py_V1);}
71
72 if (py_V1 == Py_None)
73 {
74 V1 = NULL;
75 }
76 else
77 {
78
79 assert(py_V1->ob_refcnt >= 2); // There should be at least one ref from the container object,
80 // and one ref from the local scope.
81
82 if (CudaNdarray_Check(py_V1))
83 {
84 //fprintf(stderr, "c_extract CNDA object w refcnt %p %i\n", py_V1, (py_V1->ob_refcnt));
85 V1 = (CudaNdarray*)py_V1;
86 //std::cerr << "c_extract " << V1 << '\n';
87
88
89 if (V1->nd != 2)
90 {
91 PyErr_Format(PyExc_RuntimeError,
92 "c_extract: Some CudaNdarray has rank %i, it was supposed to have rank 2",
93 V1->nd);
94 V1 = NULL;
95 {
96 __failure = 2;
97 if (!PyErr_Occurred()) {
98 PyErr_SetString(PyExc_RuntimeError,
99 "Unexpected error in an Op's C code. "
100 "No Python exception was set.");
101 }
102 goto __label_2;};
103 }
104 //std::cerr << "c_extract " << V1 << " nd check passed\n";
105
106
107 if (CudaNdarray_HOST_DIMS(V1)[1] != 1)
108 {
109 PyErr_Format(PyExc_RuntimeError,
110 "c_extract: Some CudaNdarray has dim %i on broadcastable dimension %i",
111 CudaNdarray_HOST_DIMS(V1)[1], 1);
112 V1 = NULL;
113 {
114 __failure = 2;
115 if (!PyErr_Occurred()) {
116 PyErr_SetString(PyExc_RuntimeError,
117 "Unexpected error in an Op's C code. "
118 "No Python exception was set.");
119 }
120 goto __label_2;};
121 }
122 //std::cerr << "c_extract " << V1 << "dim check 1 passed\n";
123 //std::cerr << "c_extract " << V1 << "checking bcast 1 <" << V1->str<< ">\n";
124 //std::cerr << "c_extract " << V1->str[1] << "\n";
125 if (CudaNdarray_HOST_STRIDES(V1)[1])
126 {
127 //std::cerr << "c_extract bad stride detected...\n";
128 PyErr_Format(PyExc_RuntimeError,
129 "c_extract: Some CudaNdarray has a nonzero stride %i on a broadcastable dimension %i",
130 CudaNdarray_HOST_STRIDES(V1)[1], 1);
131 V1 = NULL;
132 {
133 __failure = 2;
134 if (!PyErr_Occurred()) {
135 PyErr_SetString(PyExc_RuntimeError,
136 "Unexpected error in an Op's C code. "
137 "No Python exception was set.");
138 }
139 goto __label_2;};
140 }
141 //std::cerr << "c_extract " << V1 << "bcast check 1 passed\n";
142
143
144 assert(V1);
145 Py_INCREF(py_V1);
146 }
147 else if (py_V1 == Py_None)
148 {
149 PyErr_SetString(PyExc_TypeError,
150 "expected a CudaNdarray, not None");
151 V1 = NULL;
152 {
153 __failure = 2;
154 if (!PyErr_Occurred()) {
155 PyErr_SetString(PyExc_RuntimeError,
156 "Unexpected error in an Op's C code. "
157 "No Python exception was set.");
158 }
159 goto __label_2;};
160 }
161 else
162 {
163 //fprintf(stderr, "FAILING c_extract CNDA object w refcnt %p %i\n", py_V1, (py_V1->ob_refcnt));
164 PyErr_SetString(PyExc_TypeError, "Argument not a CudaNdarray");
165 V1 = NULL;
166 {
167 __failure = 2;
168 if (!PyErr_Occurred()) {
169 PyErr_SetString(PyExc_RuntimeError,
170 "Unexpected error in an Op's C code. "
171 "No Python exception was set.");
172 }
173 goto __label_2;};
174 }
175 //std::cerr << "c_extract done " << V1 << '\n';
176
177
178 }
179
180 {
181
182 py_V3 = PyList_GET_ITEM(storage_V3, 0);
183 {Py_XINCREF(py_V3);}
184
185 assert(py_V3->ob_refcnt >= 2); // There should be at least one ref from the container object,
186 // and one ref from the local scope.
187
188 if (CudaNdarray_Check(py_V3))
189 {
190 //fprintf(stderr, "c_extract CNDA object w refcnt %p %i\n", py_V3, (py_V3->ob_refcnt));
191 V3 = (CudaNdarray*)py_V3;
192 //std::cerr << "c_extract " << V3 << '\n';
193
194
195 if (V3->nd != 2)
196 {
197 PyErr_Format(PyExc_RuntimeError,
198 "c_extract: Some CudaNdarray has rank %i, it was supposed to have rank 2",
199 V3->nd);
200 V3 = NULL;
201 {
202 __failure = 4;
203 if (!PyErr_Occurred()) {
204 PyErr_SetString(PyExc_RuntimeError,
205 "Unexpected error in an Op's C code. "
206 "No Python exception was set.");
207 }
208 goto __label_4;};
209 }
210 //std::cerr << "c_extract " << V3 << " nd check passed\n";
211
212
213 if (CudaNdarray_HOST_DIMS(V3)[1] != 1)
214 {
215 PyErr_Format(PyExc_RuntimeError,
216 "c_extract: Some CudaNdarray has dim %i on broadcastable dimension %i",
217 CudaNdarray_HOST_DIMS(V3)[1], 1);
218 V3 = NULL;
219 {
220 __failure = 4;
221 if (!PyErr_Occurred()) {
222 PyErr_SetString(PyExc_RuntimeError,
223 "Unexpected error in an Op's C code. "
224 "No Python exception was set.");
225 }
226 goto __label_4;};
227 }
228 //std::cerr << "c_extract " << V3 << "dim check 1 passed\n";
229 //std::cerr << "c_extract " << V3 << "checking bcast 1 <" << V3->str<< ">\n";
230 //std::cerr << "c_extract " << V3->str[1] << "\n";
231 if (CudaNdarray_HOST_STRIDES(V3)[1])
232 {
233 //std::cerr << "c_extract bad stride detected...\n";
234 PyErr_Format(PyExc_RuntimeError,
235 "c_extract: Some CudaNdarray has a nonzero stride %i on a broadcastable dimension %i",
236 CudaNdarray_HOST_STRIDES(V3)[1], 1);
237 V3 = NULL;
238 {
239 __failure = 4;
240 if (!PyErr_Occurred()) {
241 PyErr_SetString(PyExc_RuntimeError,
242 "Unexpected error in an Op's C code. "
243 "No Python exception was set.");
244 }
245 goto __label_4;};
246 }
247 //std::cerr << "c_extract " << V3 << "bcast check 1 passed\n";
248
249
250 assert(V3);
251 Py_INCREF(py_V3);
252 }
253 else if (py_V3 == Py_None)
254 {
255 PyErr_SetString(PyExc_TypeError,
256 "expected a CudaNdarray, not None");
257 V3 = NULL;
258 {
259 __failure = 4;
260 if (!PyErr_Occurred()) {
261 PyErr_SetString(PyExc_RuntimeError,
262 "Unexpected error in an Op's C code. "
263 "No Python exception was set.");
264 }
265 goto __label_4;};
266 }
267 else
268 {
269 //fprintf(stderr, "FAILING c_extract CNDA object w refcnt %p %i\n", py_V3, (py_V3->ob_refcnt));
270 PyErr_SetString(PyExc_TypeError, "Argument not a CudaNdarray");
271 V3 = NULL;
272 {
273 __failure = 4;
274 if (!PyErr_Occurred()) {
275 PyErr_SetString(PyExc_RuntimeError,
276 "Unexpected error in an Op's C code. "
277 "No Python exception was set.");
278 }
279 goto __label_4;};
280 }
281 //std::cerr << "c_extract done " << V3 << '\n';
282
283
284 {
285 // Op class GpuElemwise
286
287 //std::cerr << "C_CODE RoundHalfToEven START\n";
288 //standard elemwise size checks
289
290
291 int dims[2] = {1,1};
292
293
294 int broadcasts_V3[2] = {0, 1};
295
296
297 //std::cerr << "C_CODE RoundHalfToEven checking input V3\n";
298 if (2 != V3->nd)
299 {
300 PyErr_Format(PyExc_TypeError,
301 "need 2 dims, not %i", V3->nd);
302 {
303 __failure = 5;
304 if (!PyErr_Occurred()) {
305 PyErr_SetString(PyExc_RuntimeError,
306 "Unexpected error in an Op's C code. "
307 "No Python exception was set.");
308 }
309 goto __label_5;};
310 }
311 for (int i = 0; i< 2; ++i)
312 {
313 dims[i] = (dims[i] == 1) ? CudaNdarray_HOST_DIMS(V3)[i] : dims[i];
314 if ((!(broadcasts_V3[i] &&
315 CudaNdarray_HOST_DIMS(V3)[i] == 1)) &&
316 (dims[i] != CudaNdarray_HOST_DIMS(V3)[i]))
317 {
318 //std::cerr << "C_CODE RoundHalfToEven checking input V3 failed\n";
319 PyErr_Format(PyExc_ValueError,
320 "GpuElemwise. Input dimension mis-match. Input"
321 " 0 (indices start at 0) has shape[%i] == %i"
322 ", but the output's size on that axis is %i.",
323 i,
324 CudaNdarray_HOST_DIMS(V3)[i],
325 dims[i]
326 );
327 {
328 __failure = 5;
329 if (!PyErr_Occurred()) {
330 PyErr_SetString(PyExc_RuntimeError,
331 "Unexpected error in an Op's C code. "
332 "No Python exception was set.");
333 }
334 goto __label_5;};
335 }
336 }
337
338
339 Py_XDECREF(V1);
340 V1 = V3;
341 Py_INCREF(V1);
342 for (int i = 0; (i< 2) && (V1); ++i) {
343 if (dims[i] != CudaNdarray_HOST_DIMS(V1)[i])
344 {
345 PyErr_Format(PyExc_ValueError,
346 "GpuElemwise. Output dimension mis-match. Output"
347 " 0 (indices start at 0), working inplace"
348 " on input 0, has shape[%i] == %i"
349 ", but the output's size on that axis is %i.",
350 i,
351 CudaNdarray_HOST_DIMS(V1)[i],
352 dims[i]
353 );
354 Py_DECREF(V1);
355 V1 = NULL;
356 {
357 __failure = 5;
358 if (!PyErr_Occurred()) {
359 PyErr_SetString(PyExc_RuntimeError,
360 "Unexpected error in an Op's C code. "
361 "No Python exception was set.");
362 }
363 goto __label_5;};
364 }
365 }
366 //std::cerr << "ELEMWISE NEW V1 nd" << V1->nd << "\n";
367 //std::cerr << "ELEMWISE NEW V1 data" << V1->devdata << "\n";
368
369
370 {
371 //new block so that failure gotos don't skip over variable initialization
372 //std::cerr << "calling callkernel\n";
373 if (callkernel_node_f4ef5cd450ad67f267508aba71bc9604_0(1, 0, dims
374
375
376 , CudaNdarray_DEV_DATA(V3), CudaNdarray_HOST_STRIDES(V3)
377
378
379 , CudaNdarray_DEV_DATA(V1), CudaNdarray_HOST_STRIDES(V1)
380
381
382 ))
383 {
384 // error
385
386
387 Py_DECREF(V1);
388 V1 = NULL;
389
390
391 {
392 __failure = 5;
393 if (!PyErr_Occurred()) {
394 PyErr_SetString(PyExc_RuntimeError,
395 "Unexpected error in an Op's C code. "
396 "No Python exception was set.");
397 }
398 goto __label_5;};
399 }
400 else // no error
401 {
402 }
403 }
404 //std::cerr << "C_CODE RoundHalfToEven END\n";
405
406 __label_5:
407
408 double __DUMMY_5;
409
410 }
411 __label_4:
412
413 //std::cerr << "cleanup " << py_V3 << " " << V3 << "\n";
414 //fprintf(stderr, "c_cleanup CNDA py_object w refcnt %p %i\n", py_V3, (py_V3->ob_refcnt));
415 if (V3)
416 {
417 //fprintf(stderr, "c_cleanup CNDA cn_object w refcnt %p %i\n", V3, (V3->ob_refcnt));
418 Py_XDECREF(V3);
419 }
420 //std::cerr << "cleanup done" << py_V3 << "\n";
421
422 {Py_XDECREF(py_V3);}
423
424 double __DUMMY_4;
425
426 }
427 __label_2:
428
429 if (!__failure) {
430
431 //std::cerr << "sync\n";
432 if (NULL == V1) {
433 // failure: sync None to storage
434 Py_XDECREF(py_V1);
435 py_V1 = Py_None;
436 Py_INCREF(py_V1);
437 }
438 else
439 {
440 if (py_V1 != (PyObject*)V1)
441 {
442 Py_XDECREF(py_V1);
443 py_V1 = (PyObject*)V1;
444 Py_INCREF(py_V1);
445 }
446 assert(py_V1->ob_refcnt);
447 }
448
449 PyObject* old = PyList_GET_ITEM(storage_V1, 0);
450 {Py_XINCREF(py_V1);}
451 PyList_SET_ITEM(storage_V1, 0, py_V1);
452 {Py_XDECREF(old);}
453 }
454
455 //std::cerr << "cleanup " << py_V1 << " " << V1 << "\n";
456 //fprintf(stderr, "c_cleanup CNDA py_object w refcnt %p %i\n", py_V1, (py_V1->ob_refcnt));
457 if (V1)
458 {
459 //fprintf(stderr, "c_cleanup CNDA cn_object w refcnt %p %i\n", V1, (V1->ob_refcnt));
460 Py_XDECREF(V1);
461 }
462 //std::cerr << "cleanup done" << py_V1 << "\n";
463
464 {Py_XDECREF(py_V1);}
465
466 double __DUMMY_2;
467
468 }
469
470
471 if (__failure) {
472 // When there is a failure, this code puts the exception
473 // in __ERROR.
474 PyObject* err_type = NULL;
475 PyObject* err_msg = NULL;
476 PyObject* err_traceback = NULL;
477 PyErr_Fetch(&err_type, &err_msg, &err_traceback);
478 if (!err_type) {err_type = Py_None;Py_INCREF(Py_None);}
479 if (!err_msg) {err_msg = Py_None; Py_INCREF(Py_None);}
480 if (!err_traceback) {err_traceback = Py_None; Py_INCREF(Py_None);}
481 PyObject* old_err_type = PyList_GET_ITEM(__ERROR, 0);
482 PyObject* old_err_msg = PyList_GET_ITEM(__ERROR, 1);
483 PyObject* old_err_traceback = PyList_GET_ITEM(__ERROR, 2);
484 PyList_SET_ITEM(__ERROR, 0, err_type);
485 PyList_SET_ITEM(__ERROR, 1, err_msg);
486 PyList_SET_ITEM(__ERROR, 2, err_traceback);
487 {Py_XDECREF(old_err_type);}
488 {Py_XDECREF(old_err_msg);}
489 {Py_XDECREF(old_err_traceback);}
490 }
491 // The failure code is returned to index what code block failed.
492 return __failure;
493
494 }
495 };
496 }
497
498
499 static int __struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604_executor(__struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604* self) {
500 return self->run();
501 }
502
503 static void __struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604_destructor(void* executor, void* self) {
504 delete ((__struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604*)self);
505 }
506
507 //////////////////////
508 //// Functions
509 //////////////////////
510 static PyObject * instantiate(PyObject * self, PyObject *argtuple) {
511 assert(PyTuple_Check(argtuple));
512 if (3 != PyTuple_Size(argtuple)){
513 PyErr_Format(PyExc_TypeError, "Wrong number of arguments, expected 3, got %i", (int)PyTuple_Size(argtuple));
514 return NULL;
515 }
516 __struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604* struct_ptr = new __struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604();
517 if (struct_ptr->init( PyTuple_GET_ITEM(argtuple, 0),PyTuple_GET_ITEM(argtuple, 1),PyTuple_GET_ITEM(argtuple, 2) ) != 0) {
518 delete struct_ptr;
519 return NULL;
520 }
521 PyObject* thunk = PyCObject_FromVoidPtrAndDesc((void*)(&__struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604_executor), struct_ptr, __struct_compiled_op_f4ef5cd450ad67f267508aba71bc9604_destructor);
522 return thunk; }
523
524 //////////////////////
525 //// Module init
526 //////////////////////
527 static PyMethodDef MyMethods[] = {
528 {"instantiate", instantiate, METH_VARARGS, "undocumented"} ,
529 {NULL, NULL, 0, NULL}
530 };
531 PyMODINIT_FUNC initf4ef5cd450ad67f267508aba71bc9604(void){
532 (void) Py_InitModule("f4ef5cd450ad67f267508aba71bc9604", MyMethods);
533 }
534
===============================
mod.cu(373): error: identifier "callkernel_node_f4ef5cd450ad67f267508aba71bc9604_0" is undefined
1 error detected in the compilation of "/tmp/tmpxft_00010e67_00000000-9_mod.cpp1.ii".
['nvcc', '-shared', '-O3', '--maxrregcount=32', '-arch=sm_37', '-m64', '-Xcompiler', '-fno-math-errno,-Wno-unused-label,-Wno-unused-variable,-Wno-write-strings,-DCUDA_NDARRAY_CUH=c72d035fdf91890f3b36710688069b2e,-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION,-fPIC,-fvisibility=hidden', '-Xlinker', '-rpath,/home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/cuda_ndarray', '-I/home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/cuda_ndarray', '-I/usr/local/cuda-8.0/include', '-I/home/oak/venv2/local/lib/python2.7/site-packages/numpy/core/include', '-I/usr/include/python2.7', '-I/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof', '-I/home/oak/venv2/local/lib/python2.7/site-packages/theano/sandbox/cuda', '-o', '/home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/tmpi0RVsA/f4ef5cd450ad67f267508aba71bc9604.so', 'mod.cu', '-L/home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/cuda_ndarray', '-L/usr/lib', '-lcudart', '-lcublas', '-lcuda_ndarray', '-lpython2.7']
Traceback (most recent call last):
File "train.py", line 25, in <module>
model.train(callbacks.get_callbacks(weights_path), gen, gen)
File "/home/oak/git//vsearch/trainer/model/resnet50_siamese_graph.py", line 131, in train
validation_data=valid_gen, callbacks=actions, verbose=verbose)
File "/home/oak/venv2/local/lib/python2.7/site-packages/keras/engine/training.py", line 1455, in fit_generator
self._make_train_function()
File "/home/oak/venv2/local/lib/python2.7/site-packages/keras/engine/training.py", line 768, in _make_train_function
**self._function_kwargs)
File "/home/oak/venv2/local/lib/python2.7/site-packages/keras/backend/theano_backend.py", line 969, in function
return Function(inputs, outputs, updates=updates, **kwargs)
File "/home/oak/venv2/local/lib/python2.7/site-packages/keras/backend/theano_backend.py", line 955, in __init__
**kwargs)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/compile/function.py", line 320, in function
output_keys=output_keys)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/compile/pfunc.py", line 479, in pfunc
output_keys=output_keys)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/compile/function_module.py", line 1777, in orig_function
defaults)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/compile/function_module.py", line 1641, in create
input_storage=input_storage_lists, storage_map=storage_map)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/link.py", line 690, in make_thunk
storage_map=storage_map)[:3]
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/vm.py", line 1003, in make_all
no_recycling))
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/sandbox/cuda/__init__.py", line 256, in make_thunk
compute_map, no_recycling)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/op.py", line 970, in make_thunk
no_recycling)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/op.py", line 879, in make_c_thunk
output_storage=node_output_storage)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/cc.py", line 1200, in make_thunk
keep_lock=keep_lock)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/cc.py", line 1143, in __compile__
keep_lock=keep_lock)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/cc.py", line 1595, in cthunk_factory
key=key, lnk=self, keep_lock=keep_lock)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/cmodule.py", line 1142, in module_from_key
module = lnk.compile_cmodule(location)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof/cc.py", line 1506, in compile_cmodule
preargs=preargs)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/sandbox/cuda/nvcc_compiler.py", line 399, in compile_str
'for cmd', ' '.join(cmd))
Exception: ('The following error happened while compiling the node', GpuElemwise{RoundHalfToEven}[(0, 0)](GpuElemwise{Composite{sqrt(clip(i0, i1, i2))}}[(0, 0)].0), '\n', 'nvcc return status', 2, 'for cmd', 'nvcc -shared -O3 --maxrregcount=32 -arch=sm_37 -m64 -Xcompiler -fno-math-errno,-Wno-unused-label,-Wno-unused-variable,-Wno-write-strings,-DCUDA_NDARRAY_CUH=c72d035fdf91890f3b36710688069b2e,-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION,-fPIC,-fvisibility=hidden -Xlinker -rpath,/home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/cuda_ndarray -I/home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/cuda_ndarray -I/usr/local/cuda-8.0/include -I/home/oak/venv2/local/lib/python2.7/site-packages/numpy/core/include -I/usr/include/python2.7 -I/home/oak/venv2/local/lib/python2.7/site-packages/theano/gof -I/home/oak/venv2/local/lib/python2.7/site-packages/theano/sandbox/cuda -o /home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/tmpi0RVsA/f4ef5cd450ad67f267508aba71bc9604.so mod.cu -L/home/oak/.theano/compiledir_Linux-4.4--generic-x86_64-with-Ubuntu-16.04-xenial-x86_64-2.7.12-64/cuda_ndarray -L/usr/lib -lcudart -lcublas -lcuda_ndarray -lpython2.7', '[GpuElemwise{RoundHalfToEven}[(0, 0)](<CudaNdarrayType(float32, col)>)]')
(venv2)$ python train.py
Using Theano backend.
Using gpu device 0: Tesla K80 (CNMeM is enabled with initial size: 95.0% of memory, cuDNN 5105)
/home/oak/venv2/local/lib/python2.7/site-packages/theano/sandbox/cuda/__init__.py:600: UserWarning: Your cuDNN version is more recent than the one Theano officially supports. If you see any problems, try updating Theano or downgrading cuDNN to version 5.
warnings.warn(warn)
Traceback (most recent call last):
File "train.py", line 7, in <module>
model = t_model.TModel()
File "/home/oak/git/vsearch/trainer/model/resnet50_siamese_graph.py", line 93, in __init__
out_a_base = example_network(input_a)
File "/home/oak/git/vsearch/trainer/model/resnet50_siamese_graph.py", line 46, in example_network
digit_input = Input(shape=(input_dim))
File "/home/oak/venv2/local/lib/python2.7/site-packages/keras/engine/topology.py", line 1194, in Input
batch_shape = (None,) + tuple(shape)
File "/home/oak/venv2/local/lib/python2.7/site-packages/theano/tensor/var.py", line 553, in __iter__
raise TypeError(('TensorType does not support iteration. '
TypeError: TensorType does not support iteration. Maybe you are using builtin.sum instead of theano.tensor.sum? (Maybe .max?)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment