ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Commit graph

  • aa31da4797 fix Logsoftmax infinity error (#1576) Xin Qiu 2017-09-21 16:02:03 +0800
  • f7ea6315e5 Support load tf unet (#1575) Ian Wong 2017-09-21 11:53:21 +0800
  • 956593ad02 Handle null array (#1571) Jerry Wu 2017-09-21 10:16:16 +0800
  • cc21eaf0cf Add slice and onehot (#1547) Yao Zhang 2017-09-20 21:47:37 +0800
  • 4cb832c141 Add assert, image decoders, parseExample operation (#1559) Yang Wang 2017-09-20 10:02:05 +0800
  • 42fafc3ba0 support different type in tensorflow loader (#1569) Yang Wang 2017-09-19 18:51:59 +0800
  • 4d0aaf96ea optimize copy in recurrent (#1556) zhangxiaoli73 2017-09-19 13:34:00 +0800
  • 29626d863a support multi output in graph follow up (#1564) Yang Wang 2017-09-18 13:02:57 +0800
  • 1b5ad05658 Fix failed tensorflow unit test (#1563) Ian Wong 2017-09-15 17:23:04 +0800
  • e7e78e5ea9 Implement and test Conv2d operation (#1539) Yao Zhang 2017-09-15 15:41:49 +0800
  • 88e2e4583b Add Floor L2Loss RandomUniform Rank MatMul SoftMax operations (#1533) Yao Zhang 2017-09-15 10:34:58 +0800
  • b4919497fc Support switch and merge operation (#1555) Ian Wong 2017-09-14 13:21:05 +0800
  • e16f246980 Fix operation logical related bugs (#1549) Ian Wong 2017-09-12 13:39:27 +0800
  • d213b4df98 Display bigdl model in tensorboard (#1545) Ian Wong 2017-09-12 11:32:51 +0800
  • 416dbb4e4b Add Greater and Less operations (#1525) Yao Zhang 2017-09-07 09:52:58 +0800
  • 810059432e Make layer suppor different input and output numeric type and Add Equal Operation as an example (#1520) Yao Zhang 2017-09-06 14:45:12 +0800
  • 6f368cfa35 support loading a subset of tf graph and support control dependency in graph (#1499) Yang Wang 2017-09-04 14:38:04 +0800
  • 34baa6cfc4 Allow module output multiple tensor in Graph (#1507) Ian Wong 2017-09-04 11:19:57 +0800
  • 6fedfc5ca1 Add Squeeze, Prod, Sum, Reshape, Identity, ReLU (#1477) Yao Zhang 2017-09-01 16:34:21 +0800
  • 23dfc7a222 1D Max Pooling (#1436) Yao Zhang 2017-09-01 15:29:46 +0800
  • 33ab8e8fe5 Add LogicalNot, LogicalOr and LogicalAnd (#1495) Yao Zhang 2017-09-01 15:27:13 +0800
  • 19737f3a6e serialization - add custom data conversion support (#1511) Jerry Wu 2017-09-01 15:22:10 +0800
  • d345ad10a4 add isInputWithBias and isHiddenWithBias for RNNCell (#1515) Yan Wan 2017-08-30 11:20:08 +0800
  • 391d3feaf0 Remove the code dealing with translating tensorflow NHWC format to BigDL NCHW format (#1506) Yang Wang 2017-08-29 01:18:04 -0700
  • 170f72c534 add isSplitInput in BiRecurrent (#1514) Yan Wan 2017-08-29 08:40:36 +0800
  • a9ba423241 BifurcateSplitTable (#1513) Yan Wan 2017-08-28 14:58:45 +0800
  • daf7344ef5 add BatchParams for BiRecurrent layer (#1512) Yan Wan 2017-08-28 12:59:18 +0800
  • 370291a66b change tanh backward (#1502) zhangxiaoli73 2017-08-28 09:19:07 +0800
  • 9415d4f3d0 Recurrent supports BatchNormalization (#1176) Yan Wan 2017-08-25 11:09:27 +0800
  • e71a4dcf55 Serialization - version control & python API (#1491) Jerry Wu 2017-08-25 08:25:40 +0800
  • b87cf0f95d support NHWC format in SpatialAveragePooling (#1488) Yang Wang 2017-08-24 00:27:39 -0700
  • 4586cfa84b Give Detailed information of nn exception (#1482) Hawkwood 2017-08-24 11:22:24 +0800
  • f1b8f790cd fix inplace layer clearstate (#1496) Xianyan 2017-08-23 13:47:54 +0800
  • 56d4d42712 [Issue #1033 #450] Disable backpropagation for general layers (#1181) Xianyan 2017-08-23 10:44:11 +0800
  • 2f75079c81 Support NHWC format in SpatialMaxPooling (#1444) Yang Wang 2017-08-20 19:51:48 -0700
  • 69fe0e2722 support layerwise serialization (#1472) Jerry Wu 2017-08-18 17:06:11 +0800
  • 2a29808ffc add getTimes in recurrent and TimeDistributed layer (#1470) Yan Wan 2017-08-17 16:24:28 +0800
  • 377d90cd4d optimize Sigmoid (#1471) zhangxiaoli73 2017-08-17 13:43:41 +0800
  • 0f372cd19c support same padding in maxpooling and avgpooling (#1458) Yang Wang 2017-08-16 19:06:11 -0700
  • c358955cef Add abstract Operation, Add, Substract, Multiply and Divide operations (#1462) Yao Zhang 2017-08-16 17:12:24 +0800
  • 058a1cac5e add mkl.tanh (#1450) zhangxiaoli73 2017-08-16 15:54:39 +0800
  • 6baef96116 remove set operation (#1464) Yan Wan 2017-08-16 10:51:55 +0800
  • 76007c1a30 support nhwc format in SpatialConvolution (#1434) Yang Wang 2017-08-14 05:55:26 -0700
  • 7d8612d806 optimize logsoftmax (#1424) Yan Wan 2017-08-14 15:25:24 +0800
  • 78e5653d2a fix copyStatus in TimeDistributed layer (#1451) Yan Wan 2017-08-14 14:05:41 +0800
  • 7ac4bbfcc5 add funcional model definition (#1200) zhangxiaoli73 2017-08-14 11:05:38 +0800
  • 59aa377c8b revise LSTMPeephole to Graph (#1430) Yan Wan 2017-08-11 15:07:57 +0800
  • 914216b258 GRU to Graph (#1429) Yan Wan 2017-08-11 13:57:28 +0800
  • ce22aae616 SimpleRNN to Graph (#1431) Yan Wan 2017-08-11 13:57:10 +0800
  • 86371f3907 LSTM to Graph (#1428) Yan Wan 2017-08-11 13:56:37 +0800
  • 38251491a6 Set hiddenstate (#1439) dding3 2017-08-10 20:34:53 -0700
  • fcdb4f0355 add nhwc im2col and col2im kernel (#1419) Yang Wang 2017-08-07 01:58:46 -0700
  • 3e610768be Add 3D de-convolution (#1401) Yao Zhang 2017-08-03 11:04:21 +0800
  • 2aa51233a8 Check target size in ClassNll (#1420) Xin Qiu 2017-08-02 14:27:18 +0800
  • fc4fd2005b add regularizer for convlstmpeephole3d (#1395) dding3 2017-07-27 23:10:48 -0700
  • 1713f68127 add regulaizer to volumetricconvolution (#1371) Xianyan 2017-07-27 16:45:20 +0800
  • 4e5e25662a Fix temporalconv (#1359) Jerry Wu 2017-07-27 14:49:26 +0800
  • 4bf88dacb5 support get final state and cell (#1369) dding3 2017-07-26 22:37:11 -0700
  • 61d443e51c add SpatialWithinChannelLRN, update the mapping of caffe lrn (#1364) Xianyan 2017-07-26 16:18:23 +0800
  • 9757dbd1a7 Support global average pooling (#1365) Xianyan 2017-07-26 15:35:22 +0800
  • 543a3762af Convlstm3d (#1361) dding3 2017-07-25 22:17:34 -0700
  • 13ea525adc remove unnecessary contiguous check in volumetricconvolution (#1357) dding3 2017-07-24 20:59:26 -0700
  • 0ed660313a Recurrent: fix clearState (#1306) Yan Wan 2017-07-20 14:43:13 +0800
  • ea3f8cf661 Add scala tensorflow load/save examples (#1291) Ian Wong 2017-07-20 00:39:06 -0500
  • dc0d060505 fix batchSize non-consistency bug (#1295) Yan Wan 2017-07-20 12:36:13 +0800
  • b51c466961 add CosineDistanceCriterion (#1195) Yan Wan 2017-07-19 20:28:29 +0800
  • 7772dece03 Refine tensorflow doc (a follow up of pr #1253 to meet code review) (#1272) Yang Wang 2017-07-18 22:45:28 -0700
  • e16bfe4599 Graph doc (#1275) Ian Wong 2017-07-19 00:19:56 -0500
  • c91f5242ab update shared convolution (#1144) Xin Qiu 2017-07-19 08:39:30 +0800
  • f658e378ce fix SpatialConvolution's hashcode and clearState will throw NullPointerException (#1268) Xin Qiu 2017-07-18 15:24:54 +0800
  • 7aa7bd16e8 add load caffe model doc, remove useless parameter, add python api (#1266) Xianyan 2017-07-18 15:10:31 +0800
  • 9c449c1ac7 optimize joinTable perf (#1248) zhangxiaoli73 2017-07-18 12:13:43 +0800
  • 1c2d630932 Fix the bug of JoinTable after calling clearState() (#1257) Xianyan 2017-07-18 10:28:47 +0800
  • dfdb03cc17 fix concat when the dimension concatenated along is larger than 2 (#1230) Yang Wang 2017-07-13 23:29:36 -0700
  • ee994fb14d convert back convlstm optimization ding 2017-07-13 11:05:49 -0400
  • df873d648c BatchNormalization accepts default values (#1216) Yan Wan 2017-07-13 15:17:57 +0800
  • 5b665b64cf fix inception and xavier (#1220) Xin Qiu 2017-07-13 12:18:09 +0800
  • d733f90901 Tree LSTM with Sentiment classification example (#1217) Yao Zhang 2017-07-13 06:23:49 +0800
  • 1033e735b7 Make ClassNLL can process -1 label for padding, not update the model. (#1215) Yao Zhang 2017-07-12 17:21:31 +0800
  • 52fc2bcba5 update local jvm with comments jenniew 2017-07-10 13:16:45 -0700
  • 1373b79587 update abstract module jenniew 2017-07-07 00:44:54 -0700
  • 76411f929a create localModule and update local predictor jenniew 2017-07-07 00:35:14 -0700
  • 98717d6d8b Fix maptable bug (#1214) Yao Zhang 2017-07-12 16:44:41 +0800
  • 9909c1e630 Add 1D Convolution (TemporalConvolution) (#1180) Yao Zhang 2017-07-12 15:51:00 +0800
  • fe05012ea9 Fix a bug in the MV layer (#1171) Yao Zhang 2017-07-12 14:28:31 +0800
  • 3e90228f9d improve perf and add md ding 2017-07-10 11:16:52 -0400
  • 381a6089fe add optimation for convlstm ding 2017-07-10 09:00:27 -0400
  • 01e8b8d2d8 save/load model from s3 (#1136) Xin Qiu 2017-07-11 21:23:44 +0800
  • f1c300d730 RNN cells do not execute preTopology in Dropout mode JerryYanWan 2017-07-10 10:59:42 +0800
  • 36685c3226 Caffe persister (#1095) Jerry Wu 2017-07-11 14:35:31 +0800
  • dfc5958854 [Issue 864] save module weights, load (all / part) module weights (#1102) Xianyan 2017-07-11 13:57:59 +0800
  • f4170c89ef fix bugs in convLSTM JerryYanWan 2017-07-10 13:46:31 +0800
  • b15db4690b fix comments ding 2017-07-06 18:16:37 -0400
  • 2b9d72caea support convlstm and convolution without bias ding 2017-06-26 11:08:01 -0400
  • ea6ad9dd90 fix comments ding 2017-07-06 14:41:41 -0400
  • 7bb0d92a68 support spatial convultion without bias ding 2017-07-05 11:56:46 -0400
  • 6d7f846894 Remove unnecessary varialbe in Recurrent object construtor (#1127) Yao Zhang 2017-07-07 10:12:11 +0800
  • 9f00c87f7d Make SpatialFullConvolution declare with implict numeric type (#1128) Yao Zhang 2017-07-06 18:47:07 +0800
  • 2f25b70660 Optimize LSTMPeephole by pre-calculating the input2hidden linear (#1158) Yan Wan 2017-07-06 16:55:12 +0800
  • ffae7fa1ba optimize GRU by pre-calculating three linear operations (#1150) Yan Wan 2017-07-06 15:37:28 +0800