Age | Commit message (Collapse) | Author | |
---|---|---|---|
2021-01-23 | Remove the third term in canonical consistency loss | Jordan Gong | |
2021-01-23 | Add late start support for non-disentangling parts | Jordan Gong | |
2021-01-23 | Transform all frames together in evaluation | Jordan Gong | |
2021-01-23 | Evaluation bug fixes and code review | Jordan Gong | |
1. Return full cached clip in evaluation 2. Add multi-iter checkpoints support in evaluation 3. Remove duplicated code while transforming | |||
2021-01-22 | Handle unexpected restore iter | Jordan Gong | |
1. Skip finished model before load it 2. Raise error when restore iter is greater than total iter | |||
2021-01-21 | Print average losses after 100 iters | Jordan Gong | |
2021-01-21 | Bug fixes | Jordan Gong | |
1. Turn off autograd while decoding canonical and pose features 2. Change default batch size to (4, 8) | |||
2021-01-20 | Bump up version for scikit-learn | Jordan Gong | |
2021-01-14 | Enable optimizer fine tuning | Jordan Gong | |
2021-01-14 | Remove DataParallel | Jordan Gong | |
2021-01-13 | Tensorboard typo fix | Jordan Gong | |
2021-01-13 | Add server startup script | Jordan Gong | |
2021-01-13 | Update config file and convert int to str when joining | Jordan Gong | |
2021-01-13 | Add multiple checkpoints for different model and set default config value | Jordan Gong | |
2021-01-12 | Move the model to GPU before constructing optimizer | Jordan Gong | |
2021-01-12 | Some changes in hyperparameter config | Jordan Gong | |
1. Separate hyperparameter configs in model, optimizer and scheduler 2. Add more tunable hyperparameters in optimizer and scheduler | |||
2021-01-12 | Some type hint fixes | Jordan Gong | |
2021-01-12 | Print result after evaluation | Jordan Gong | |
2021-01-12 | Typo correct in evaluate function | Jordan Gong | |
2021-01-12 | Bump up version for tqdm | Jordan Gong | |
2021-01-11 | Add evaluation script, code review and fix some bugs | Jordan Gong | |
1. Add new `train_all` method for one shot calling 2. Print time used in 1k iterations 3. Correct label dimension in predict function 4. Transpose distance matrix for convenient indexing 5. Sort dictionary before generate signature 6. Extract visible CUDA setting function | |||
2021-01-11 | Implement evaluator | Jordan Gong | |
2021-01-10 | Make predict function transform samples different conditions in a single shot | Jordan Gong | |
2021-01-09 | Add prototype predict function | Jordan Gong | |
2021-01-09 | Change auto-encoder input in evaluation | Jordan Gong | |
2021-01-09 | Fix NaN when separate sum is zero | Jordan Gong | |
2021-01-07 | Train different models in different conditions | Jordan Gong | |
2021-01-07 | Add typical training script and some bug fixes | Jordan Gong | |
1. Resolve deprecated scheduler stepping issue 2. Make losses in the same scale(replace mean with sum in separate triplet loss, enlarge pose similarity loss 10x) 3. Add ReLU when compute distance in triplet loss 4. Remove classes except Model from `models` package init | |||
2021-01-07 | Change device config and add enable multi-GPU computing | Jordan Gong | |
1. Add `disable_acc` switch for disabling accelerator. When it is off, system will automatically choosing accelerator. 2. Enable multi-GPU training using torch.nn.DataParallel | |||
2021-01-06 | Add CUDA support | Jordan Gong | |
2021-01-06 | Add TensorBoard support | Jordan Gong | |
2021-01-05 | Implement checkpoint mechanism | Jordan Gong | |
2021-01-05 | Implement Batch All Triplet Loss | Jordan Gong | |
2021-01-05 | Change and improve weight initialization | Jordan Gong | |
1. Change initial weights for Conv layers 2. Find a way to init last fc in init_weights | |||
2021-01-03 | Separate last fc matrix from weight init function | Jordan Gong | |
Recursive apply will override other parameters too | |||
2021-01-03 | Delete dead training judge | Jordan Gong | |
2021-01-03 | Implement weight initialization | Jordan Gong | |
2021-01-03 | Update hyperparameter configuration, implement prototype fit function | Jordan Gong | |
2021-01-03 | Add separate fully connected layers | Jordan Gong | |
2021-01-03 | Unit testing on auto-encoder, HPM and Part Net | Jordan Gong | |
2021-01-03 | Bump up version for pillow | Jordan Gong | |
2021-01-02 | Separate training and evaluating | Jordan Gong | |
2021-01-02 | Correct feature dims after disentanglement and HPM backbone removal | Jordan Gong | |
1. Features used in HPM is decoded canonical embedding without transpose convolution 2. Decode pose embedding to image for Part Net 3. Backbone seems to be redundant, we can use feature map given by auto-decoder | |||
2021-01-02 | Change type of pose similarity loss to tensor | Jordan Gong | |
2020-12-31 | Implement some parts of RGB-GaitPart wrapper | Jordan Gong | |
1. Triplet loss function and weight init function haven't been implement yet 2. Tuplize features returned by auto-encoder for later unpack 3. Correct comment error in auto-encoder 4. Swap batch_size dim and time dim in HPM and PartNet in case of redundant transpose 5. Find backbone problems in HPM and disable it temporarily 6. Make feature structure by HPM consistent to that by PartNet 7. Fix average pooling dimension issue and incorrect view change in HP | |||
2020-12-31 | Make HPM capable of processing frames in all batches | Jordan Gong | |
2020-12-31 | Make super class constructor revoke consistent | Jordan Gong | |
2020-12-31 | Bug Fixes in HPM and PartNet | Jordan Gong | |
1. Register list of torch.nn.Module to the network using torch.nn.ModuleList 2. Fix operation error in squeeze list of tensor 3. Replace squeeze with view in HP in case batch size is 1 | |||
2020-12-30 | Correct and refine PartNet | Jordan Gong | |
1. Let FocalConv block capable of processing frames in all batches 2. Correct input dims of TFA and output dims of HP 3. Change torch.unsqueeze and torch.cat to torch.stack | |||
2020-12-30 | Combine FPFE and TFA to PartNet | Jordan Gong | |