aboutsummaryrefslogtreecommitdiffstats
Commit message (Expand)AuthorAgeFilesLines
* quantization doc updateHEADmasterManu Mathew2020-10-092-73/+155
* segmentation and classification docs updatedManu Mathew2020-10-093-189/+216
* use AddBlock for addition in fpnManu Mathew2020-09-151-1/+3
* Quantization documentation updatedManu Mathew2020-08-291-4/+9
* documentation update - added accuracies for regnetx based segmentation models.Manu Mathew2020-08-113-17/+40
* constrain_bias_start_iter - reverting to the correct valueManu Mathew2020-08-061-2/+2
* cleanedup STE for QAT. Added RegNetX modelsManu Mathew2020-08-0641-371/+1813
* renamed pytorch_jacinto_ai.vision to pytorch_jacinto_ai.xvisionManu Mathew2020-07-20120-176/+212
* updated python package requirements (don't need tensorflow for tensorboard). ...Manu Mathew2020-06-3010-29/+145
* updated quantization modules to support mmdetection, using Hardtanh for fixed...Manu Mathew2020-06-1620-111/+237
* re-implemented QuantTestModule using QuantTrainModule. constrain_bias added.Manu Mathew2020-05-2710-225/+434
* default of model_surgery_quantize is now True for QuantTestModule.Manu Mathew2020-05-239-40/+70
* support DataParallel for QuantTrainModuleManu Mathew2020-05-1911-131/+122
* cosmetic change to quantization shell scriptManu Mathew2020-05-191-29/+32
* minor update to quantization docs & scritps. support for external model in cl...Manu Mathew2020-05-194-7/+9
* quantization_example - RandomSampler is used when epoch_size!=0. epoch_size=0...Manu Mathew2020-05-142-22/+27
* epoch_size - meaning has changed to number of images instead of number of ite...Manu Mathew2020-05-144-13/+13
* added more options to the main scriptsManu Mathew2020-05-142-12/+14
* doc update - to clarify the use of model.train() and model.eval()Manu Mathew2020-05-142-7/+15
* constrain_weights - in default scenario, this is not used with per_channel_q.Manu Mathew2020-05-142-3/+4
* epoch_size and shuffling arguments for validationManu Mathew2020-05-143-7/+35
* class weights were not being used in segmentation loss due to a bug. fixed it.Manu Mathew2020-05-141-1/+5
* quantization - fix for ConvTranspose2d and BatchNorm merge. Do not merge weig...Manu Mathew2020-05-124-46/+65
* calibration - fix bugs there were introduced during the recent restructureManu Mathew2020-05-123-26/+19
* pretrained model names updateManu Mathew2020-05-122-4/+4
* changing extension of checkpoint files to .pth instead of .pth.tarManu Mathew2020-05-129-21/+22
* model path updateManu Mathew2020-05-122-7/+7
* minor document updateManu Mathew2020-05-111-3/+1
* minor document updateManu Mathew2020-05-111-2/+4
* cleanup of several modules in the repositoryManu Mathew2020-05-1111-437/+165
* documentation updateManu Mathew2020-05-082-13/+14
* bugfix for evaluationManu Mathew2020-05-084-12/+14
* better implementation for epoch_sizeManu Mathew2020-05-084-70/+96
* quantization docs updateManu Mathew2020-05-011-3/+6
* quantization docs updatedManu Mathew2020-04-301-0/+1
* quantization docs updateManu Mathew2020-04-301-3/+7
* quantization docs updateManu Mathew2020-04-301-7/+9
* quantization docs updatedManu Mathew2020-04-301-10/+7
* quantization docs updateManu Mathew2020-04-301-7/+10
* quantization doc updateManu Mathew2020-04-292-22/+12
* quantization doc updateManu Mathew2020-04-291-8/+8
* quantization doc updateManu Mathew2020-04-292-12/+12
* quantization doc updateManu Mathew2020-04-291-13/+22
* docs update and minor fixesManu Mathew2020-04-2920-472/+122
* quantization cleanup and minor fixesManu Mathew2020-04-2714-260/+685
* quantization fixes, docs update, resize_with()Manu Mathew2020-04-2315-59/+82
* quantization aware training - bugfix for merged weights becoming 0 (typically...Manu Mathew2020-03-273-41/+295
* torch.nn.ReLU is the recommended activation module. removed the custom define...Manu Mathew2020-03-1216-83/+507
* support Hardtanh activation function also in quantization aware trainingManu Mathew2020-03-127-14/+7
* ResizeWith, UpsampleWith classes that can export to onnx with scale_factor in...Manu Mathew2020-03-035-16/+24