Merge branch 'main' into cohere

.github/workflows/slow_tests_gaudi2.yml

@@ -3,7 +3,7 @@ name: (Gaudi2) Non-regression tests
 on:
   workflow_dispatch:
   schedule:
-    - cron: '0 23 * * 3,6'  # every Wednesday and Saturday at 1am CET (midnight winter time)
+    - cron: '0 23 * * *'  # every Wednesday and Saturday at 1am CET (midnight winter time)
 
 concurrency:
   group: ${{ github.workflow }}
@@ -63,7 +63,7 @@ jobs:
             --net=host \
             --ipc=host \
             vault.habana.ai/gaudi-docker/1.17.0/ubuntu22.04/habanalabs/pytorch-installer-2.3.1:latest \
-            pip install huggingface_hub && huggingface-cli login --token ${{ secrets.TEXT_GENERATION_CI_HUB_TOKEN }} && /bin/bash tests/ci/slow_tests_deepspeed.sh
+            /bin/bash tests/ci/slow_tests_deepspeed.sh ${{ secrets.TEXT_GENERATION_CI_HUB_TOKEN }}
   fsdp:
     name: Test FSDP models
     if: ${{ !cancelled() && (success() || failure()) }}

Makefile

@@ -39,6 +39,7 @@ fast_tests:
 # Run unit and integration tests related to Diffusers
 fast_tests_diffusers:
 	python -m pip install .[tests]
+	python -m pip install -r examples/stable-diffusion/requirements.txt
 	python -m pytest tests/test_diffusers.py
 
 # Run single-card non-regression tests on image classification models
@@ -94,12 +95,13 @@ slow_tests_deepspeed: test_installs
 	python -m pytest tests/test_examples.py -v -s -k "deepspeed"
 
 slow_tests_diffusers: test_installs
-	python -m pytest tests/test_diffusers.py -v -s -k "test_no_"
+	python -m pip install -r examples/stable-diffusion/requirements.txt
 	python -m pytest tests/test_diffusers.py -v -s -k "test_textual_inversion"
 	python -m pip install peft==0.7.0
 	python -m pytest tests/test_diffusers.py -v -s -k "test_train_text_to_image_"
 	python -m pytest tests/test_diffusers.py -v -s -k "test_train_controlnet"
 	python -m pytest tests/test_diffusers.py -v -s -k "test_deterministic_image_generation"
+	python -m pytest tests/test_diffusers.py -v -s -k "test_no_"
 
 # Run text-generation non-regression tests
 slow_tests_text_generation_example: test_installs
@@ -119,8 +121,8 @@ slow_tests_fsdp: test_installs
 	python -m pytest tests/test_fsdp_examples.py -v -s --token $(TOKEN)
 
 slow_tests_trl: test_installs
-	python -m pip install trl==0.8.6
-	python -m pip install peft==0.7.0
+	python -m pip install trl==0.9.6
+	python -m pip install peft==0.12.0
 	python -m pytest tests/test_trl.py -v -s -k "test_calculate_loss"
 
 slow_tests_object_segmentation: test_installs

README.md

@@ -212,6 +212,7 @@ The following model architectures, tasks and device distributions have been vali
 | Mixtral |   | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Persimmon |   | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Qwen2 | <div style="text-align:left"><li>Single card</li></div> | <div style="text-align:left"><li>Single card</li></div> | <li>[language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)</li><li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
+| Qwen2-MoE |   | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Gemma | :heavy_check_mark:  | <div style="text-align:left"><li>Single card</li></div> | <li>[language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)</li><li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Cohere       |          | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | T5 / Flan T5 | :heavy_check_mark: | :heavy_check_mark: | <li>[summarization](https://github.com/huggingface/optimum-habana/tree/main/examples/summarization)</li><li>[translation](https://github.com/huggingface/optimum-habana/tree/main/examples/translation)</li><li>[question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering#fine-tuning-t5-on-squad20)</li> |

docs/source/index.mdx

@@ -58,6 +58,7 @@ In the tables below, ✅ means single-card, multi-card and DeepSpeed have all be
 | Mixtral      |          | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Gemma        | ✅       | <div style="text-align:left"><li>Single card</li></div> | <li>[language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)</li><li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Qwen2        | <div style="text-align:left"><li>Single card</li></div> | <div style="text-align:left"><li>Single card</li></div> | <li>[language modeling](https://github.com/huggingface/optimum-habana/tree/main/examples/language-modeling)</li><li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
+| Qwen2-MoE    |          | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Persimmon    |          | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | Cohere       |          | <div style="text-align:left"><li>Single card</li></div> | <li>[text generation](https://github.com/huggingface/optimum-habana/tree/main/examples/text-generation)</li> |
 | T5 / Flan T5 | ✅       | ✅        | <li>[summarization](https://github.com/huggingface/optimum-habana/tree/main/examples/summarization)</li><li>[translation](https://github.com/huggingface/optimum-habana/tree/main/examples/translation)</li><li>[question answering](https://github.com/huggingface/optimum-habana/tree/main/examples/question-answering#fine-tuning-t5-on-squad20)</li> |

examples/audio-classification/run_audio_classification.py

@@ -46,7 +46,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.14.0", "To fix: pip install -r examples/pytorch/audio-classification/requirements.txt")

examples/contrastive-image-text/run_bridgetower.py

@@ -56,7 +56,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/contrastive-image-text/requirements.txt")

examples/contrastive-image-text/run_clip.py

@@ -61,7 +61,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/contrastive-image-text/requirements.txt")

examples/image-classification/run_image_classification.py

@@ -63,7 +63,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/image-classification/requirements.txt")

examples/image-to-text/run_pipeline.py

@@ -211,13 +211,15 @@ def main():
     )
     generate_kwargs = {
         "lazy_mode": True,
-        "use_cache": args.use_kv_cache,
         "hpu_graphs": args.use_hpu_graphs,
         "max_new_tokens": args.max_new_tokens,
         "ignore_eos": args.ignore_eos,
         "use_flash_attention": args.use_flash_attention,
         "flash_attention_recompute": args.flash_attention_recompute,
     }
+    if args.use_kv_cache:
+        generate_kwargs["use_cache"] = args.use_kv_cache
+
     if args.use_hpu_graphs:
         from habana_frameworks.torch.hpu import wrap_in_hpu_graph
 

examples/language-modeling/run_clm.py

@@ -62,7 +62,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")

examples/language-modeling/run_mlm.py

@@ -61,7 +61,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")

examples/language-modeling/run_multitask_prompt_tuning.py

@@ -60,7 +60,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risk.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")

examples/language-modeling/run_prompt_tuning_clm.py

@@ -62,7 +62,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")

examples/question-answering/run_qa.py

@@ -60,7 +60,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt")

examples/question-answering/run_seq2seq_qa.py

@@ -56,7 +56,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt")

examples/sentence-transformers-training/README.md

@@ -0,0 +1,24 @@
+# Examples for Sentence Transformers
+
+We provide 3 examples to show how to use Sentence Transformers with HPU devices.
+
+- **[training_stsbenchmark.py](https://github.com/huggingface/optimum-habana/tree/main/examples/sentence-transformers-training/sts)** - This example shows how to create a Sentence Transformers model from scratch by using a pre-trained transformer model (e.g. [`distilbert-base-uncased`](https://huggingface.co/distilbert/distilbert-base-uncased)) together with a pooling layer.
+
+- **[training_nli.py](https://github.com/huggingface/optimum-habana/tree/main/examples/sentence-transformers-training/nli)** - This example provides two sentences (a premise and a hypothesis), and the task of Natural Language Inference (NLI) is to determine whether the premise entails the hypothesis, contradicts it, or if they are neutral. Commonly the NLI dataset in [SNLI](https://huggingface.co/datasets/stanfordnlp/snli) and [MultiNLI](https://huggingface.co/datasets/nyu-mll/multi_nli) are used.
+
+- **[training_paraphrases.py](https://github.com/huggingface/optimum-habana/tree/main/examples/sentence-transformers-training/paraphrases)** - This example loads various datasets from Sentence Transformers. We construct batches by sampling examples from the respective dataset.
+
+### Tested Examples/Models and Configurations
+
+The following table contains examples supported and configurations we have validated on Gaudi2.
+
+| Examples                    |  General  | e5-mistral-7b-instruct | BF16 | Single Card | Multi-Cards |
+|-----------------------------|-----------|------------|------|-------------|-------------|
+| training_nli.py             |     ✔     |      ✔    |   ✔  |     ✔       |     ✔      |
+| training_stsbenchmark.py    |     ✔     |      ✔    |   ✔  |     ✔       |     ✔      |
+| training_paraphrases.py     |     ✔     |           |       |     ✔       |            |
+
+Notice:
+1. In the table, the column 'General' refers to general models like mpnet, MiniLM.
+2. When e5-mistral-7b-instruct model is enabled for the test, single card will use LoRA + gradient checkpointing and multi-card will use DeepSpeed ZeRO-2/ZeRO-3 to reduce the memory requirement.
+3. For the detailed instructions on how to run each example, you can refer to the README file located in each example folder.

examples/speech-recognition/run_speech_recognition_ctc.py

@@ -59,7 +59,7 @@ def check_optimum_habana_min_version(*a, **b):
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers and Optimum Habana are not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.18.0", "To fix: pip install -r examples/pytorch/speech-recognition/requirements.txt")

examples/speech-recognition/run_speech_recognition_seq2seq.py

@@ -55,7 +55,7 @@ def check_optimum_habana_min_version(*a, **b):
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.43.0")
+check_min_version("4.45.0")
 check_optimum_habana_min_version("1.14.0.dev0")
 
 require_version("datasets>=1.18.0", "To fix: pip install -r examples/pytorch/speech-recognition/requirements.txt")