mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-17 03:28:22 +06:00
3f9f749325
7 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Younes Belkada
|
f7ea959b96
|
[core/ GC / tests] Stronger GC tests (#27124)
* stronger GC tests * better tests and skip failing tests * break down into 3 sub-tests * break down into 3 sub-tests * refactor a bit * more refactor * fix * last nit * credits contrib and suggestions * credits contrib and suggestions --------- Co-authored-by: Yih-Dar <2521628+ydshieh@users.noreply.github.com> Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> |
||
|
Younes Belkada
|
3ce3385c47
|
Revert "Fix gradient checkpointing + fp16 autocast for most models" (#24420)
Revert "Fix gradient checkpointing + fp16 autocast for most models (#24247)"
This reverts commit
|
||
|
Younes Belkada
|
285a48011d
|
Fix gradient checkpointing + fp16 autocast for most models (#24247)
* fix gc bug
* continue PoC on OPT
* fixes
* 🤯
* fix tests
* remove pytest.mark
* fixup
* forward contrib credits from discussions
* forward contrib credits from discussions
* reverting changes on untouched files.
---------
Co-authored-by: zhaoqf123 <zhaoqf123@users.noreply.github.com>
Co-authored-by: 7eu7d7 <7eu7d7@users.noreply.github.com>
|
||
|
Eli Simhayev
|
4b6a5a7caa
|
[Time-Series] Autoformer model (#21891)
* ran `transformers-cli add-new-model-like`
* added `AutoformerLayernorm` and `AutoformerSeriesDecomposition`
* added `decomposition_layer` in `init` and `moving_avg` to config
* added `AutoformerAutoCorrelation` to encoder & decoder
* removed caninical self attention `AutoformerAttention`
* added arguments in config and model tester. Init works! 😁
* WIP autoformer attention with autocorrlation
* fixed `attn_weights` size
* wip time_delay_agg_training
* fixing sizes and debug time_delay_agg_training
* aggregation in training works! 😁
* `top_k_delays` -> `top_k_delays_index` and added `contiguous()`
* wip time_delay_agg_inference
* finish time_delay_agg_inference 😎
* added resize to autocorrelation
* bug fix: added the length of the output signal to `irfft`
* `attention_mask = None` in the decoder
* fixed test: changed attention expected size, `test_attention_outputs` works!
* removed unnecessary code
* apply AutoformerLayernorm in final norm in enc & dec
* added series decomposition to the encoder
* added series decomp to decoder, with inputs
* added trend todos
* added autoformer to README
* added to index
* added autoformer.mdx
* remove scaling and init attention_mask in the decoder
* make style
* fix copies
* make fix-copies
* inital fix-copies
* fix from https://github.com/huggingface/transformers/pull/22076
* make style
* fix class names
* added trend
* added d_model and projection layers
* added `trend_projection` source, and decomp layer init
* added trend & seasonal init for decoder input
* AutoformerModel cannot be copied as it has the decomp layer too
* encoder can be copied from time series transformer
* fixed generation and made distrb. out more robust
* use context window to calculate decomposition
* use the context_window for decomposition
* use output_params helper
* clean up AutoformerAttention
* subsequences_length off by 1
* make fix copies
* fix test
* added init for nn.Conv1d
* fix IGNORE_NON_TESTED
* added model_doc
* fix ruff
* ignore tests
* remove dup
* fix SPECIAL_CASES_TO_ALLOW
* do not copy due to conv1d weight init
* remove unused imports
* added short summary
* added label_length and made the model non-autoregressive
* added params docs
* better doc for `factor`
* fix tests
* renamed `moving_avg` to `moving_average`
* renamed `factor` to `autocorrelation_factor`
* make style
* Update src/transformers/models/autoformer/configuration_autoformer.py
Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>
* Update src/transformers/models/autoformer/configuration_autoformer.py
Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>
* fix configurations
* fix integration tests
* Update src/transformers/models/autoformer/configuration_autoformer.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* fixing `lags_sequence` doc
* Revert "fixing `lags_sequence` doc"
This reverts commit
|
||
|
Yih-Dar
|
fa01127a67
|
update_pip_test_mapping (#22606)
* Add TFBlipForConditionalGeneration * update pipeline_model_mapping * Add import * Revert changes in GPTSanJapaneseTest --------- Co-authored-by: ydshieh <ydshieh@users.noreply.github.com> |
||
|
Eli Simhayev
|
9eae4aa576
|
[Time-Series] fix past_observed_mask type (#22076)
added > 0.5 to `past_observed_mask` |
||
|
Eli Simhayev
|
8abe4930d3
|
[Time-Series] informer model (#21099)
* added informer to gitignore * added informer to gitignore * WIP informer2020 * added checking that instantiate works * added config using gluonTS by kashif * WIP config * adding informeConfig. need to remove FeatureEmbedder * done InformerConfig, but need to change the names * Done informer model init. working on enc-dec * added things to address, after reading again enc-dec in the paper * done modeling - checking initialization work * added informer to gitignore * WIP informer2020 * added checking that instantiate works * added config using gluonTS by kashif * WIP config * adding informeConfig. need to remove FeatureEmbedder * done InformerConfig, but need to change the names * Done informer model init. working on enc-dec * added things to address, after reading again enc-dec in the paper * done modeling - checking initialization work * moved enc-dec init to InformerEncoder/Decoder init * added 'init_std' to config, now model init works! * WIP conversion script, and added code sources * WIP conversion script: loading original informer pth works * WIP conversion script: change defaults in the config * WIP conversion script: supporting Informer input embedding * WIP conversion script: added parameters for the informer embed * WIP conversion script: change dim_feedforward=2048 * WIP conversion script: remove unused args for loading checkpoint * just cleaning up * DataEmbedding removed, after thinking with Kashif * working on forward pass * WIP forward pass: trying to establish working batch for forward pass * cleaning and finalizing * adding HF names and docs * init after cleaning works * WIP in tests * added docs for the informer specific args * fix style * undo change * cleaning informer, now need to work only enc-dec * initial enc-dec classes * added encoder and decoder * added todo * add todos for conv_layers * added decoder docs from vanilla * added encoder docs from vanilla * remove encoder decoder from the original informer * removed AttentionLayer from the original paper * removed TriangularCausalMask, same as decoder_attention_mask * initial sparse attention * use conv_layers * fixed test_config test * fix parenthesis when itearting zip(layers, conv_layers) * error found in prob attention, added sizes as comments * fix sizes * added proposal for q_reduce indexing, and remove unused * WIP ProbMask, and changed factor=2 for testing * remove unused libs for this PR for creating the env * fix checking the attn_weights.size() after bmm * Q_reduce: changed from torch.gather to simple slicing * WIP calculate final attn_output * finish adding v_aggregated, attn_output ready * changed tgt_len to u in attention_mask, need to fix the size error * comment attention_mask for encoder, and fix if cond for v_agg * added ProbMask support (wip), removed old original code * finished ProbMask 😃 * Revert "remove unused libs for this PR for creating the env" This reverts commit |