Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+source.spm filter=lfs diff=lfs merge=lfs -text
+target.spm filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,65 @@
+---
+license: apache-2.0
+base_model: Helsinki-NLP/opus-mt-en-az
+base_model_relation: quantized
+tags:
+  - onnx
+  - translation
+  - marian
+library_name: transformers
+pipeline_tag: translation
+language:
+  - en
+  - az
+---
+
+# opus-mt-en-az-onnx
+
+ONNX export (fp32 + dynamic int8) of [`Helsinki-NLP/opus-mt-en-az`](https://huggingface.co/Helsinki-NLP/opus-mt-en-az),
+a Marian (en -> az) translation model from the Helsinki-NLP OPUS-MT project.
+
+**License:** apache-2.0, inherited unchanged from the base model.
+
+## Contents
+
+| Files | What it is |
+|---|---|
+| `encoder_model.onnx`, `decoder_model.onnx`, `decoder_with_past_model.onnx` | ONNX, float32 |
+| `int8/` | same three graphs, dynamic int8 (`QInt8`, `MatMul` only, `/lm_head/MatMul` excluded) |
+| `source.spm`, `target.spm`, `vocab.json` | `MarianTokenizer` over the original sentencepiece models |
+
+fp32 size: 485 MB (three graphs) | int8 size: 295 MB
+
+## Usage
+
+```python
+from optimum.onnxruntime import ORTModelForSeq2SeqLM
+from transformers import AutoTokenizer
+
+repo = "TigreGotico/opus-mt-en-az-onnx"
+tok = AutoTokenizer.from_pretrained(repo)
+model = ORTModelForSeq2SeqLM.from_pretrained(repo, use_cache=True, use_merged=False)  # fp32
+# int8: ORTModelForSeq2SeqLM.from_pretrained(repo, subfolder="int8", use_cache=True, use_merged=False)
+inputs = tok("The weather is very nice today.", return_tensors="pt")
+out = model.generate(**inputs, num_beams=4, max_new_tokens=64)
+print(tok.decode(out[0], skip_special_tokens=True))
+```
+
+## Parity with the original PyTorch model
+
+10 general-domain sentences, exact-string-match of generated output against
+`MarianMTModel.generate()` on the original `Helsinki-NLP/opus-mt-en-az` checkpoint.
+
+| Decoding | fp32 exact match | int8 exact match |
+|---|---|---|
+| greedy (num_beams=1) | 10/10 (100.0%) | 9/10 (90.0%) |
+| beam=4 | 10/10 (100.0%) | 4/10 (40.0%) |
+
+fp32 is a faithful reproduction of the original model at both decoding
+settings. int8 dynamic quantization noticeably degrades quality on this
+checkpoint under beam search - a spot check found real semantic drift on
+longer sentences (not just paraphrase), e.g. for "We need to discuss the
+budget for next quarter." the int8 beam-4 output diverged from both the
+reference and the fp32 ONNX output. Prefer fp32 for this pair; int8 is
+provided for size-constrained deployments where greedy decoding is used and
+some quality loss is acceptable.

config.json

@@ -0,0 +1,55 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "swish",
+  "add_bias_logits": false,
+  "add_final_layer_norm": false,
+  "architectures": [
+    "MarianMTModel"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 512,
+  "decoder_attention_heads": 8,
+  "decoder_ffn_dim": 2048,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 23287,
+  "decoder_vocab_size": 23288,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 2048,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 0,
+  "extra_pos_embeddings": 23288,
+  "forced_eos_token_id": 0,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": null,
+  "max_position_embeddings": 512,
+  "model_type": "marian",
+  "normalize_before": false,
+  "normalize_embedding": false,
+  "num_beams": null,
+  "num_hidden_layers": 6,
+  "pad_token_id": 23287,
+  "scale_embedding": true,
+  "share_encoder_decoder_embeddings": true,
+  "static_position_embeddings": true,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "vocab_size": 23288
+}

decoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df66ca072d83fc3fe9d1545cdf2aff14003b3069bd9bc2eb7b86ae0557de6d54
+size 197627460

decoder_with_past_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e31d3e07d0af2cde2bc2b19fabbc00b8de24a9374cc629b4a5424680820cac94
+size 184997853

encoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5cda9fc48dfa9d6e7924077d604674d409c56202e845f66e4c991115fab018c
+size 124473974

generation_config.json

@@ -0,0 +1,16 @@
+{
+  "bad_words_ids": [
+    [
+      23287
+    ]
+  ],
+  "bos_token_id": 0,
+  "decoder_start_token_id": 23287,
+  "eos_token_id": 0,
+  "forced_eos_token_id": 0,
+  "max_length": 512,
+  "num_beams": 4,
+  "pad_token_id": 23287,
+  "renormalize_logits": true,
+  "transformers_version": "4.57.6"
+}

int8/decoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:869df0af2f242e4d2782e49b8362ac6ee7656cbbe0bdf3fa86f30e63601b6a40
+size 122283040

int8/decoder_with_past_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6316f5b354c1e8685e9aeb31bbe0531a11f741af51ba5c0c5147099b1b4f8dca
+size 119071097

int8/encoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7872ef6fc986c9fa8cead3ada84511241a13ad1e343a71ec36611b1ab749455
+size 67920512

source.spm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0a4133c5eedaf9c3f880314bccc94a000251dc0a207e2df736b35273ccb68b4
+size 451246

special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

target.spm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b052899ce024e382af780e7350384e6879e34fb0e7d64cd785bdde7948052197
+size 470040

tokenizer_config.json

@@ -0,0 +1,39 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23287": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "separate_vocabs": false,
+  "source_lang": "eng",
+  "sp_model_kwargs": {},
+  "target_lang": "aze",
+  "tokenizer_class": "MarianTokenizer",
+  "unk_token": "<unk>"
+}