Upload converted EOLE COMET model

README.md

@@ -0,0 +1,102 @@
+---
+library_name: eole
+tags:
+- eole
+- comet
+- machine-translation-evaluation
+- quality-estimation
+- converted
+base_model: Unbabel/wmt22-comet-da
+license: apache-2.0
+pipeline_tag: text-classification
+---
+
+# WMT22 COMET DA (EOLE)
+
+This is [Unbabel/wmt22-comet-da](https://huggingface.co/Unbabel/wmt22-comet-da) converted to [EOLE](https://github.com/eole-nlp/eole) format.
+
+No model weights were modified. This repository contains a format conversion for use with EOLE's native `transformer_encoder_scorer` implementation.
+
+This is not an upstream `unbabel-comet` checkpoint layout. Use it with EOLE.
+
+## Model Details
+
+| | |
+|---|---|
+| Original model | [Unbabel/wmt22-comet-da](https://huggingface.co/Unbabel/wmt22-comet-da) |
+| EOLE architecture | `transformer_encoder_scorer` |
+| Scoring type | `comet` |
+| Class identifier | `regression_metric` |
+| EOLE scorer | `EOLE-COMET` |
+| Requires reference | Yes |
+| Encoder | XLM-R large style encoder |
+
+## Usage
+
+Requires an EOLE version with native COMET scorer support.
+
+### Validation Metric
+
+```yaml
+valid_metrics: ["EOLE-COMET"]
+comet_model: eole-nlp/wmt22-comet-da-eole
+comet_batch_size: 64
+```
+
+### Direct Scoring
+
+```bash
+eole predict \
+  --model_path eole-nlp/wmt22-comet-da-eole \
+  --src /path/to/src.txt \
+  --tgt /path/to/mt.txt \
+  --ref /path/to/ref.txt \
+  --output /path/to/scores.txt \
+  --with_score
+```
+
+By default, direct scoring writes one segment score per input line. To emit a
+single aggregate system score, add `--score_level system`:
+
+```bash
+eole predict \
+  --model_path eole-nlp/wmt22-comet-da-eole \
+  --src /path/to/src.txt \
+  --tgt /path/to/mt.txt \
+  --ref /path/to/ref.txt \
+  --output /path/to/system-score.txt \
+  --with_score \
+  --score_level system
+```
+
+`score_level: system` writes one numeric line containing the arithmetic mean of
+the segment scores, similar to Unbabel COMET's `--only_system` mode but using
+EOLE's score-file format.
+
+## Conversion
+
+Converted with:
+
+```bash
+eole convert COMET \
+  --model Unbabel/wmt22-comet-da \
+  --output wmt22-comet-da-eole
+```
+
+## Parity
+
+This conversion was parity-checked against the upstream Unbabel COMET runtime. On a 100-line WMT17 slice, native EOLE scoring matched the upstream baseline within floating point tolerance:
+
+| Metric | Value |
+|---|---:|
+| System score delta | `1.85e-08` |
+| Sentence MAE | `5.42e-08` |
+| Sentence max abs | `7.15e-07` |
+
+## Original Model and Attribution
+
+This model is a conversion of [Unbabel/wmt22-comet-da](https://huggingface.co/Unbabel/wmt22-comet-da). The model weights were converted without modification.
+
+This converted repository follows the upstream model license: Apache-2.0. Please refer to the original model card and license for training data, intended use, limitations, and citation details.
+
+Unbabel COMET: https://github.com/Unbabel/COMET

config.json

@@ -0,0 +1,62 @@
+{
+  "src_vocab": "dummy",
+  "tgt_vocab": "dummy",
+  "data": {},
+  "share_vocab": true,
+  "training": {
+    "compute_dtype": "fp32"
+  },
+  "model": {
+    "architecture": "transformer_encoder_scorer",
+    "scoring_type": "comet",
+    "class_identifier": "regression_metric",
+    "requires_reference": true,
+    "pretrained_model": "Unbabel/wmt22-comet-da",
+    "pool": "avg",
+    "layer": "mix",
+    "layer_transformation": "sparsemax",
+    "layer_norm": false,
+    "input_segments": [
+      "mt",
+      "src",
+      "ref"
+    ],
+    "hidden_sizes": [
+      3072,
+      1024
+    ],
+    "activations": "Tanh",
+    "final_activation": null,
+    "dropout": 0.1,
+    "encoder_architecture": "XLMRobertaForMaskedLM",
+    "embeddings": {
+      "embedding_type": "roberta",
+      "src_word_vec_size": 1024,
+      "tgt_word_vec_size": 1024,
+      "word_vec_size": 1024,
+      "position_encoding_type": "Learned",
+      "n_positions": 514,
+      "position_shift": 0,
+      "embedding_layer_norm": true,
+      "token_type_vocab_size": 1
+    },
+    "encoder": {
+      "encoder_type": "transformer",
+      "layers": 24,
+      "hidden_size": 1024,
+      "heads": 16,
+      "transformer_ff": 4096,
+      "mlp_activation_fn": "gelu",
+      "layer_norm": "standard",
+      "norm_eps": 1e-05,
+      "add_qkvbias": true,
+      "add_key_bias": true,
+      "add_final_linear_bias": true,
+      "add_ffnbias": true,
+      "position_encoding_type": "Learned",
+      "n_positions": 514,
+      "encoder_layer_style": "postnorm",
+      "final_encoder_layer_norm": false
+    }
+  }
+}

model.00.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d6d3b5f7503afe02c684f22802c126ad124a836b60e7ed02f7fd6ca51716f3b
+size 2323518060

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

tokenizer_config.json

@@ -0,0 +1 @@
+{"model_max_length": 512}