Add Mega: Moving Average Equipped Gated Attention (#21766)
* add mega file structure and plain pytorch version of mega source code * added config class with old naming conventions * filled in mega documentation * added config class and embeddings with optional token types * updated notes * starting the conversion process, deleted intermediate and added use_cache back to config * renamed config attributes in modeling_mega.py * checkpointing before refactoring incremental decoding functions * removed stateful incremental key/values for EMA and self-attention * refactored MovingAverageGatedAttention to remove stateful k/v history and use unified attention mask * MovingAverageGatedAttention works with incremental decoding + past values, added sequence length enforcement * more comments in MovingAverageGatedAttention + checkpointing before GatedCrossAttention * bug fix in attention mask handling in MovingAverageGatedAttention * removed incremental state from GatedCrossAttention and removed IncrementalState class * finished gated cross attention and got MegaLayer working * fixed causal masking in mega decoder * fixed how padding and causal masks are passed through MegaLayer with and without k/v caching * finished MegaModel; tested with encoder, decoder-only, and cross-attention type inputs; started work on downstream classes; removed mentions of position_ids * added optional dense hidden layer for masked and causal LM classes * docstring updates in MultiHeadEMA and GatedCrossAttention, removed unnecessary inputs in cross-attention * removed before_attn_fn in Mega class and updated docstrings and comments up to there * bug fix in MovingAverageGatedAttention masking * working conversion of MLM checkpoint in scratchpad script -- perfect matches * moved arg for hidden dense layer in LM head to config; discovered issue where from_pretrained is renaming gamma and beta parameters * renamed gamma and beta parameters to avoid HF renaming when loading from checkpoint * finished checkpoint conversion script * cleanup old class in mega config script * removed 'copied from' statements and passing integration tests * added num_attention_heads=1 to config for integration compatibility, decoder tests working, generation tests failing * fixed tuple output of megamodel * all common tests passing after fixing issues in decoder, gradient retention, and initialization * added mega-specific tests, ready for more documentation and style checks * updated docstrings; checkpoint before style fixes * style and quality checks, fixed initialization problem in float_tensor, ready for PR * added mega to toctree * removed unnecessary arg in megaconfig * removed unused arg and fixed code samples with leftover roberta models * Apply suggestions from code review Applied all suggestions except the one renaming a class, as I'll need to update that througout Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com> * fixed issue where .view breaks batch dimension, conversion script fixed with absolute imports, updated readme with Mega->MEGA * removed asserts in Mega code, renamed sequencenorm, gatedcrossattention, and NFFN, replaced get_activation_fn with ACTFN, and added sequencenorm to layer norms * reformatted .forward() docstrings to match style and removed unused mask input in cross-attention * removed all reset_parameters() methods and rolled into MegaPreTrainedModel._init_weights() * renamed all single-letter variables and improved readability in tensor size comments, Mega->MEGA in 2 documentation files * variable names in NFFN * manual Mega->MEGA changes in docs * Mega->MEGA in config auto * style and quality fixes * Apply suggestions from code review Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com> * renamed parameters and variables with confusing names, added copied from statements, moved fft conv to its own method, other cleanup from PR comments * commit before dealing with merge conflicts * made new attention activation functions available in ACT2FN and added generation test from OPT * style and quality in activations and tests * documentation fixes, renaming variables in dropout and rotary positions, used built-in causal masking, encoders->layers in MegaModel, moved comments into docstrings * style and quality fixes after latest updates, before rotary position ids * causal mask in MegaBlock docstring + added missing device passing * Apply suggestions from code review Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com> * Update README.md Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * added Mega prefixes where missing, reverted MegaSequenceNorm to if-else, other module renaming requested in PR * style and quality fixes + readme updates pointing to main --------- Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com> Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
This commit is contained in:
Mitch Naylor
@@ -337,6 +337,7 @@ conda install -c huggingface transformers
|
||||
1. **[MaskFormer](https://huggingface.co/docs/transformers/model_doc/maskformer)** (मेटा और UIUC से) पेपर के साथ जारी किया गया [प्रति-पिक्सेल वर्गीकरण वह सब नहीं है जिसकी आपको सिमेंटिक सेगमेंटेशन की आवश्यकता है] (https://arxiv.org/abs/2107.06278) बोवेन चेंग, अलेक्जेंडर जी. श्विंग, अलेक्जेंडर किरिलोव द्वारा >>>>>> रिबेस ठीक करें
|
||||
1. **[mBART](https://huggingface.co/docs/transformers/model_doc/mbart)** (फेसबुक से) साथ में पेपर [न्यूरल मशीन ट्रांसलेशन के लिए मल्टीलिंगुअल डीनोइजिंग प्री-ट्रेनिंग](https://arxiv. org/abs/2001.08210) यिनहान लियू, जियाताओ गु, नमन गोयल, जियान ली, सर्गेई एडुनोव, मार्जन ग़ज़विनिनेजाद, माइक लुईस, ल्यूक ज़ेटलमॉयर द्वारा।
|
||||
1. **[mBART-50](https://huggingface.co/docs/transformers/model_doc/mbart)** (फेसबुक से) साथ में पेपर [एक्स्टेंसिबल बहुभाषी प्रीट्रेनिंग और फाइनट्यूनिंग के साथ बहुभाषी अनुवाद](https://arxiv युकिंग टैंग, चाउ ट्रान, जियान ली, पेंग-जेन चेन, नमन गोयल, विश्रव चौधरी, जियाताओ गु, एंजेला फैन द्वारा .org/abs/2008.00401)।
|
||||
1. **[MEGA](https://huggingface.co/docs/transformers/main/model_doc/mega)** (Facebook से) Xuezhe Ma, Chunting Zhou, Xiang Kong, Junxian He, Liangke Gui, Graham Neubig, Jonathan May, and Luke Zettlemoyer. द्वाराअनुसंधान पत्र [Mega: Moving Average Equipped Gated Attention](https://arxiv.org/abs/2209.10655) के साथ जारी किया गया
|
||||
1. **[Megatron-BERT](https://huggingface.co/docs/transformers/model_doc/megatron-bert)** (NVIDIA से) कागज के साथ [Megatron-LM: मॉडल का उपयोग करके बहु-अरब पैरामीटर भाषा मॉडल का प्रशिक्षण Parallelism](https://arxiv.org/abs/1909.08053) मोहम्मद शोएबी, मोस्टोफा पटवारी, राउल पुरी, पैट्रिक लेग्रेस्ले, जेरेड कैस्पर और ब्रायन कैटानज़ारो द्वारा।
|
||||
1. **[Megatron-GPT2](https://huggingface.co/docs/transformers/model_doc/megatron_gpt2)** (NVIDIA से) साथ वाला पेपर [Megatron-LM: ट्रेनिंग मल्टी-बिलियन पैरामीटर लैंग्वेज मॉडल्स यूजिंग मॉडल पैरेललिज़्म] (https://arxiv.org/abs/1909.08053) मोहम्मद शोएबी, मोस्टोफा पटवारी, राउल पुरी, पैट्रिक लेग्रेस्ले, जेरेड कैस्पर और ब्रायन कैटानज़ारो द्वारा पोस्ट किया गया।
|
||||
1. **[MGP-STR](https://huggingface.co/docs/transformers/model_doc/mgp-str)** (Alibaba Research से) Peng Wang, Cheng Da, and Cong Yao. द्वाराअनुसंधान पत्र [Multi-Granularity Prediction for Scene Text Recognition](https://arxiv.org/abs/2209.03592) के साथ जारी किया गया
|
||||
|
||||
Reference in New Issue
Block a user