Upload folder using huggingface_hub

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,121 @@
+---
+language:
+- fr
+- en
+tags:
+- translation
+license: cc-by-4.0
+datasets:
+- quickmt/quickmt-train.fr-en
+- quickmt/madlad400-en-backtranslated-fr
+- quickmt/newscrawl2024-en-backtranslated-fr
+- quickmt/canadian_hansard
+model-index:
+- name: quickmt-fr-en
+  results:
+  - task:
+      name: Translation fra-eng
+      type: translation
+      args: fra-eng
+    dataset:
+      name: flores101-devtest
+      type: flores_101
+      args: fra_Latn eng_Latn devtest
+    metrics:
+    - name: BLEU
+      type: bleu
+      value: 46.84
+    - name: CHRF
+      type: chrf
+      value: 69.87
+    - name: COMET
+      type: comet
+      value: 89.4
+---
+
+
+# `quickmt-fr-en` Neural Machine Translation Model 
+
+`quickmt-fr-en` is a reasonably fast and reasonably accurate neural machine translation model for translation from `fr` into `en`.
+
+`quickmt` models are roughly 3 times faster for GPU inference than OpusMT models and roughly [40 times](https://huggingface.co/spaces/quickmt/quickmt-vs-libretranslate) faster than [LibreTranslate](https://huggingface.co/spaces/quickmt/quickmt-vs-libretranslate)/[ArgosTranslate](github.com/argosopentech/argos-translate). 
+
+
+## *UPDATED VERSION!* 
+
+This model was augmented with back-translated data and has improved translation quality!
+
+
+## Try it on our Huggingface Space
+
+Give it a try before downloading here: https://huggingface.co/spaces/quickmt/QuickMT-Demo
+
+
+## Model Information
+
+* Trained using [`eole`](https://github.com/eole-nlp/eole) 
+* 200M parameter seq2seq transformer
+* 32k separate Sentencepiece vocabs
+* Exported for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
+* The pytorch model (for use with [`eole`](https://github.com/eole-nlp/eole)) is available in this repository in the `eole-model` folder
+
+See the `eole` model configuration in this repository for further details and the `eole-model` for the raw `eole` (pytorch) model.
+
+
+## Usage with `quickmt`
+
+You must install the Nvidia cuda toolkit first, if you want to do GPU inference.
+
+Next, install the `quickmt` python library and download the model:
+
+```bash
+git clone https://github.com/quickmt/quickmt.git
+pip install -e ./quickmt/
+
+quickmt-model-download quickmt/quickmt-fr-en ./quickmt-fr-en
+```
+
+Finally use the model in python:
+
+```python
+from quickmt import Translator
+
+# Auto-detects GPU, set to "cpu" to force CPU inference
+mt = Translator("./quickmt-fr-en/", device="auto")
+
+# Translate - set beam size to 1 for faster speed (but lower quality)
+sample_text = "Le Dr Ehud Ur, professeur de médecine à l'Université Dalhousie de Halifax (Nouvelle-Écosse) et président de la division clinique et scientifique de l'Association canadienne du diabète, a averti que la recherche en était encore à ses débuts."
+
+mt(sample_text, beam_size=5)
+```
+
+> 'Dr. Ehud Ur, professor of medicine at Dalhousie University in Halifax, Nova Scotia and chair of the clinical and scientific division of the Canadian Diabetes Association, warned that research was still in its infancy.'
+
+```python
+# Get alternative translations by sampling
+# You can pass any cTranslate2 `translate_batch` arguments
+mt([sample_text], sampling_temperature=1.2, beam_size=1, sampling_topk=50, sampling_topp=0.9)
+```
+
+> 'Dr. Ehud Ur, Professor of Medicine at Dalhousie University in Halifax, Nova Scotia, and Chair of the Clinical & Scientific Division of the Canadian Diabetes Association, warned the research was still in its infancy.'
+
+The model is in `ctranslate2` format, and the tokenizers are `sentencepiece`, so you can use `ctranslate2` directly instead of through `quickmt`. It is also possible  to get this model to work with e.g. [LibreTranslate](https://libretranslate.com/) which also uses `ctranslate2` and `sentencepiece`. A model in safetensors format to be used with `eole` is also provided.
+
+
+## Metrics
+
+`bleu` and `chrf2` are calculated with [sacrebleu](https://github.com/mjpost/sacrebleu) on the [Flores200 `devtest` test set](https://huggingface.co/datasets/facebook/flores). `comet22` with the [`comet`](https://github.com/Unbabel/COMET) library and the [default model](https://huggingface.co/Unbabel/wmt22-comet-da). "Time (s)" is the time in seconds to translate the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32.
+
+
+|                                       |   bleu |   chrf2 |   comet22 |   Time (s) |
+|:--------------------------------------|-------:|--------:|----------:|-----------:|
+| quickmt/quickmt-fr-en                 |  46.84 |   69.87 |     89.4  |       1.08 |
+| Helsinki-NLP/opus-mt-fr-en            |  41.71 |   66.84 |     88.31 |       3.49 |
+| facebook/nllb-200-distilled-600M      |  44.05 |   67.81 |     88.48 |      21.52 |
+| facebook/nllb-200-distilled-1.3B      |  46.24 |   69.32 |     89.24 |      37.25 |
+| facebook/m2m100_418M                  |  36.48 |   63.3  |     85.87 |      18.28 |
+| facebook/m2m100_1.2B                  |  41.69 |   66.51 |     88    |      35.1  |
+| tencent/HY-MT1.5-1.8B                 |  28.68 |   61.62 |     88.66 |      9     |
+| tencent/Hunyuan-MT-7B-fp8             |  35.66 |   64.94 |     89.72 |     22     |
+| CohereLabs/aya-expanse-8b (bnb quant) |  45.03 |   69.02 |     90.29 |     73.97  |
+

README.md

@@ -1,12 +1,15 @@
 ---
 language:
-- en
 - fr
+- en
 tags:
 - translation
 license: cc-by-4.0
 datasets:
 - quickmt/quickmt-train.fr-en
+- quickmt/madlad400-en-backtranslated-fr
+- quickmt/newscrawl2024-en-backtranslated-fr
+- quickmt/canadian_hansard
 model-index:
 - name: quickmt-fr-en
   results:
@@ -19,15 +22,15 @@ model-index:
       type: flores_101
       args: fra_Latn eng_Latn devtest
     metrics:
-    - name: CHRF
-      type: chrf
-      value: 66.77
     - name: BLEU
       type: bleu
-      value: 42.17
+      value: 46.84
+    - name: CHRF
+      type: chrf
+      value: 69.87
     - name: COMET
       type: comet
-      value: 58.10
+      value: 89.4
 ---
 
 
@@ -35,16 +38,28 @@ model-index:
 
 `quickmt-fr-en` is a reasonably fast and reasonably accurate neural machine translation model for translation from `fr` into `en`.
 
+`quickmt` models are roughly 3 times faster for GPU inference than OpusMT models and roughly [40 times](https://huggingface.co/spaces/quickmt/quickmt-vs-libretranslate) faster than [LibreTranslate](https://huggingface.co/spaces/quickmt/quickmt-vs-libretranslate)/[ArgosTranslate](github.com/argosopentech/argos-translate). 
+
+
+## *UPDATED VERSION!* 
+
+This model was augmented with back-translated data and has improved translation quality!
+
+
+## Try it on our Huggingface Space
+
+Give it a try before downloading here: https://huggingface.co/spaces/quickmt/QuickMT-Demo
+
 
 ## Model Information
 
-* Trained using [`eole`](https://github.com/eole-nlp/eole)
-* 185M parameter transformer 'big' with 8 encoder layers and 2 decoder layers
-* 50k joint Sentencepiece vocabulary
+* Trained using [`eole`](https://github.com/eole-nlp/eole) 
+* 200M parameter seq2seq transformer
+* 32k separate Sentencepiece vocabs
 * Exported for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
-* Training data: https://huggingface.co/datasets/quickmt/quickmt-train.fr-en/tree/main
+* The pytorch model (for use with [`eole`](https://github.com/eole-nlp/eole)) is available in this repository in the `eole-model` folder
 
-See the `eole` model configuration in this repository for further details. 
+See the `eole` model configuration in this repository for further details and the `eole-model` for the raw `eole` (pytorch) model.
 
 
 ## Usage with `quickmt`
@@ -55,7 +70,7 @@ Next, install the `quickmt` python library and download the model:
 
 ```bash
 git clone https://github.com/quickmt/quickmt.git
-pip install ./quickmt/
+pip install -e ./quickmt/
 
 quickmt-model-download quickmt/quickmt-fr-en ./quickmt-fr-en
 ```
@@ -66,31 +81,41 @@ Finally use the model in python:
 from quickmt import Translator
 
 # Auto-detects GPU, set to "cpu" to force CPU inference
-t = Translator("./quickmt-fr-en/", device="auto")
+mt = Translator("./quickmt-fr-en/", device="auto")
 
-# Translate - set beam size to 5 for higher quality (but slower speed)
-sample_text = "Résigny est une commune française située dans le département de l'Aisne, en région Hauts-de-France. "
-t(sample_text, beam_size=1)
+# Translate - set beam size to 1 for faster speed (but lower quality)
+sample_text = "Le Dr Ehud Ur, professeur de médecine à l'Université Dalhousie de Halifax (Nouvelle-Écosse) et président de la division clinique et scientifique de l'Association canadienne du diabète, a averti que la recherche en était encore à ses débuts."
+
+mt(sample_text, beam_size=5)
+```
 
+> 'Dr. Ehud Ur, professor of medicine at Dalhousie University in Halifax, Nova Scotia and chair of the clinical and scientific division of the Canadian Diabetes Association, warned that research was still in its infancy.'
+
+```python
 # Get alternative translations by sampling
 # You can pass any cTranslate2 `translate_batch` arguments
-t([sample_text], sampling_temperature=1.2, beam_size=1, sampling_topk=50, sampling_topp=0.9)
+mt([sample_text], sampling_temperature=1.2, beam_size=1, sampling_topk=50, sampling_topp=0.9)
 ```
 
-The model is in `ctranslate2` format, and the tokenizers are `sentencepiece`, so you can use `ctranslate2` directly instead of through `quickmt`. It is also possible  to get this model to work with e.g. [LibreTranslate](https://libretranslate.com/) which also uses `ctranslate2` and `sentencepiece`.
+> 'Dr. Ehud Ur, Professor of Medicine at Dalhousie University in Halifax, Nova Scotia, and Chair of the Clinical & Scientific Division of the Canadian Diabetes Association, warned the research was still in its infancy.'
+
+The model is in `ctranslate2` format, and the tokenizers are `sentencepiece`, so you can use `ctranslate2` directly instead of through `quickmt`. It is also possible  to get this model to work with e.g. [LibreTranslate](https://libretranslate.com/) which also uses `ctranslate2` and `sentencepiece`. A model in safetensors format to be used with `eole` is also provided.
 
 
 ## Metrics
 
-`bleu` and `chrf2` are calculated with [sacrebleu](https://github.com/mjpost/sacrebleu) on the [Flores200 `devtest` test set](https://huggingface.co/datasets/facebook/flores) ("fra_Latn"->"eng_Latn"). `comet22` with the [`comet`](https://github.com/Unbabel/COMET) library and the [default model](https://huggingface.co/Unbabel/wmt22-comet-da). "Time (s)" is the time in seconds to translate (using `ctranslate2`) the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32 (faster speed is possible using a large batch size).
+`bleu` and `chrf2` are calculated with [sacrebleu](https://github.com/mjpost/sacrebleu) on the [Flores200 `devtest` test set](https://huggingface.co/datasets/facebook/flores). `comet22` with the [`comet`](https://github.com/Unbabel/COMET) library and the [default model](https://huggingface.co/Unbabel/wmt22-comet-da). "Time (s)" is the time in seconds to translate the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32.
+
 
-| Model                            | chrf2 | bleu    | comet22 | Time (s) |
-| -------------------------------- | ----- | ------- | ------- | -------- |
-| quickmt/quickmt-fr-en            | 68.22 | 44.28   | 88.86   |  1.1     |
-| Helsinki-NLP/opus-mt-fr-en       | 66.85 | 41.71   | 88.31   |  3.6     |
-| facebook/m2m100_418M             | 64.39 | 36.49   | 85.87   | 18.0     |
-| facebook/m2m100_1.2B             | 66.51 | 41.69   | 88.00   | 34.6     |
-| facebook/nllb-200-distilled-600M | 67.82 | 44.04   | 88.47   | 21.7     |
-| facebook/nllb-200-distilled-1.3B | 69.30 | 46.22   | 89.24   | 37.1     |
+|                                       |   bleu |   chrf2 |   comet22 |   Time (s) |
+|:--------------------------------------|-------:|--------:|----------:|-----------:|
+| quickmt/quickmt-fr-en                 |  46.84 |   69.87 |     89.4  |       1.08 |
+| Helsinki-NLP/opus-mt-fr-en            |  41.71 |   66.84 |     88.31 |       3.49 |
+| facebook/nllb-200-distilled-600M      |  44.05 |   67.81 |     88.48 |      21.52 |
+| facebook/nllb-200-distilled-1.3B      |  46.24 |   69.32 |     89.24 |      37.25 |
+| facebook/m2m100_418M                  |  36.48 |   63.3  |     85.87 |      18.28 |
+| facebook/m2m100_1.2B                  |  41.69 |   66.51 |     88    |      35.1  |
+| tencent/HY-MT1.5-1.8B                 |  28.68 |   61.62 |     88.66 |      9     |
+| tencent/Hunyuan-MT-7B-fp8             |  35.66 |   64.94 |     89.72 |     22     |
+| CohereLabs/aya-expanse-8b (bnb quant) |  45.03 |   69.02 |     90.29 |     73.97  |
 
-`quickmt-fr-en` is the fastest and is higher quality than `opus-mt-fr-en`, `m2m100_418m`, `m2m100_1.2B` and `nllb-200-distilled-600M`.

eole-config.yaml

@@ -1,5 +1,5 @@
 ## IO
-save_data: fr-en/data_spm
+save_data: data
 overwrite: True
 seed: 1234
 report_every: 100
@@ -8,71 +8,84 @@ tensorboard: true
 tensorboard_log_dir: tensorboard
 
 ### Vocab
-src_vocab: fr-en/joint.eole.vocab
-tgt_vocab: fr-en/joint.eole.vocab
-src_vocab_size: 50000
-tgt_vocab_size: 50000
+src_vocab: fren/fr.eole.vocab
+tgt_vocab: fren/en.eole.vocab
+src_vocab_size: 32000
+tgt_vocab_size: 32000
 vocab_size_multiple: 8
-share_vocab: True
+share_vocab: false
 n_sample: 0
 
 data:
     corpus_1:
-        path_src: hf://quickmt/quickmt-train.fr-en/fr
-        path_tgt: hf://quickmt/quickmt-train.fr-en/en
-        path_sco: hf://quickmt/quickmt-train.fr-en/sco
+        path_src: fren/train.cleaned.filtered.fr
+        path_tgt: fren/train.cleaned.filtered.en
+        weight: 200
+    corpus_2:
+        path_src: ../data/newscrawl.backtrans.cleaned.filtered.fr
+        path_tgt: ../data/newscrawl.backtrans.cleaned.filtered.en
+        weight: 35
+    corpus_3:
+        path_src: ../data/madlad.backtrans.cleaned.filtered.fr
+        path_tgt: ../data/madlad.backtrans.cleaned.filtered.en
+        weight: 68
+    corpus_4:
+        path_src: ../data/hansard.fr
+        path_tgt: ../data/hansard.en
+        weight: 5
     valid:
-        path_src: fr-en/dev.src
-        path_tgt: fr-en/dev.tgt
+        path_src: fren/dev.fr
+        path_tgt: fren/dev.en
 
 transforms: [sentencepiece, filtertoolong]
 transforms_configs:
   sentencepiece:
-    src_subword_model: "fr-en/joint.spm.model"
-    tgt_subword_model: "fr-en/joint.spm.model"
+    src_subword_model: "fren/fr.spm.model"
+    tgt_subword_model: "fren/en.spm.model"
   filtertoolong:
     src_seq_length: 256
     tgt_seq_length: 256
 
 training:
     # Run configuration
-    model_path: fr-en/model
+    model_path: quickmt-fr-en-eole-model
     keep_checkpoint: 4
-    save_checkpoint_steps: 2000
-    train_steps: 100000
-    valid_steps: 2000
+    train_steps: 200000
+    save_checkpoint_steps: 5000
+    valid_steps: 5000
     
     # Train on a single GPU
     world_size: 1
     gpu_ranks: [0]
 
-    # Batching
+    # Batching 120,000 tokens
+    # For RTX 5090, 15000 batch size, accum_count 8
     batch_type: "tokens"
-    batch_size: 8192
-    valid_batch_size: 8192
+    batch_size: 6000
+    valid_batch_size: 2048
     batch_size_multiple: 8
-    accum_count: [16]
+    accum_count: [20]
     accum_steps: [0]
 
     # Optimizer & Compute
-    compute_dtype: "bf16"
-    optim: "pagedadamw8bit"
-    #optim: "adamw"
-    learning_rate: 2.0
-    warmup_steps: 10000
+    compute_dtype: "fp16"
+    optim: "adamw"
+    #use_amp: True
+    learning_rate: 3.0
+    warmup_steps: 5000
     decay_method: "noam"
     adam_beta2: 0.998
 
     # Data loading
-    bucket_size: 128000
+    bucket_size: 256000
     num_workers: 4
-    prefetch_factor: 100
+    prefetch_factor: 128
     
     # Hyperparams
     dropout_steps: [0]
     dropout: [0.1]
     attention_dropout: [0.1]
-    max_grad_norm: 2
+    max_grad_norm: 0
     label_smoothing: 0.1
     average_decay: 0.0001
     param_init_method: xavier_uniform
@@ -80,22 +93,22 @@ training:
 
 model:
     architecture: "transformer"
-    layer_norm: standard
-    share_embeddings: true
-    share_decoder_embeddings: true
-    add_ffnbias: true
-    mlp_activation_fn: gelu
+    share_embeddings: false
+    share_decoder_embeddings: false
     add_estimator: false
+    add_ffnbias: true
     add_qkvbias: false
-    norm_eps: 1e-6
-    hidden_size: 1024
+    layer_norm: standard
+    mlp_activation_fn: gelu
+    hidden_size: 768
     encoder:
-        layers: 8
+        layers: 12
     decoder:
         layers: 2
-    heads: 8
+    heads: 16
     transformer_ff: 4096
     embeddings:
-        word_vec_size: 1024
+        word_vec_size: 768
         position_encoding_type: "SinusoidalInterleaved"
 
+

eole-model/config.json

@@ -1,126 +1,73 @@
 {
+  "tensorboard": true,
+  "tensorboard_log_dir_dated": "tensorboard/Jan-11_15-01-39",
+  "src_vocab_size": 32000,
+  "src_vocab": "fren/fr.eole.vocab",
   "seed": 1234,
-  "transforms": [
-    "sentencepiece",
-    "filtertoolong"
-  ],
-  "report_every": 100,
-  "save_data": "fr-en/data_spm",
-  "src_vocab_size": 50000,
-  "share_vocab": true,
-  "overwrite": true,
-  "tgt_vocab": "fr-en/joint.eole.vocab",
   "valid_metrics": [
     "BLEU"
   ],
-  "tensorboard_log_dir_dated": "tensorboard/Feb-17_09-24-56",
-  "src_vocab": "fr-en/joint.eole.vocab",
+  "overwrite": true,
+  "share_vocab": false,
+  "save_data": "data",
+  "report_every": 100,
   "tensorboard_log_dir": "tensorboard",
-  "tensorboard": true,
+  "transforms": [
+    "sentencepiece",
+    "filtertoolong"
+  ],
+  "tgt_vocab": "fren/en.eole.vocab",
   "n_sample": 0,
-  "tgt_vocab_size": 50000,
   "vocab_size_multiple": 8,
+  "tgt_vocab_size": 32000,
   "training": {
-    "adam_beta2": 0.998,
-    "dropout_steps": [
-      0
+    "prefetch_factor": 128,
+    "optim": "adamw",
+    "keep_checkpoint": 4,
+    "world_size": 1,
+    "decay_method": "noam",
+    "attention_dropout": [
+      0.1
     ],
+    "max_grad_norm": 0.0,
     "param_init_method": "xavier_uniform",
-    "accum_steps": [
-      0
-    ],
-    "batch_size": 8192,
+    "normalization": "tokens",
     "batch_size_multiple": 8,
     "gpu_ranks": [
       0
     ],
-    "model_path": "fr-en/model3",
-    "learning_rate": 2.0,
-    "bucket_size": 128000,
-    "train_steps": 100000,
-    "label_smoothing": 0.1,
-    "num_workers": 0,
-    "world_size": 1,
-    "compute_dtype": "torch.bfloat16",
-    "save_checkpoint_steps": 2000,
-    "dropout": [
-      0.1
+    "accum_count": [
+      20
     ],
-    "decay_method": "noam",
-    "keep_checkpoint": 4,
-    "optim": "pagedadamw8bit",
-    "normalization": "tokens",
-    "valid_batch_size": 8192,
-    "batch_type": "tokens",
-    "warmup_steps": 10000,
     "average_decay": 0.0001,
-    "prefetch_factor": 100,
-    "valid_steps": 2000,
-    "accum_count": [
-      16
+    "batch_size": 6000,
+    "compute_dtype": "torch.float16",
+    "adam_beta2": 0.998,
+    "valid_steps": 5000,
+    "dropout_steps": [
+      0
     ],
-    "attention_dropout": [
+    "train_steps": 200000,
+    "warmup_steps": 5000,
+    "learning_rate": 3.0,
+    "num_workers": 0,
+    "save_checkpoint_steps": 5000,
+    "accum_steps": [
+      0
+    ],
+    "batch_type": "tokens",
+    "dropout": [
       0.1
     ],
-    "max_grad_norm": 2.0
-  },
-  "model": {
-    "share_decoder_embeddings": true,
-    "hidden_size": 1024,
-    "mlp_activation_fn": "gelu",
-    "add_estimator": false,
-    "add_ffnbias": true,
-    "share_embeddings": true,
-    "norm_eps": 1e-06,
-    "transformer_ff": 4096,
-    "position_encoding_type": "SinusoidalInterleaved",
-    "layer_norm": "standard",
-    "architecture": "transformer",
-    "add_qkvbias": false,
-    "heads": 8,
-    "encoder": {
-      "layer_norm": "standard",
-      "rope_config": null,
-      "encoder_type": "transformer",
-      "hidden_size": 1024,
-      "add_qkvbias": false,
-      "layers": 8,
-      "src_word_vec_size": 1024,
-      "add_ffnbias": true,
-      "n_positions": null,
-      "norm_eps": 1e-06,
-      "mlp_activation_fn": "gelu",
-      "heads": 8,
-      "transformer_ff": 4096,
-      "position_encoding_type": "SinusoidalInterleaved"
-    },
-    "embeddings": {
-      "word_vec_size": 1024,
-      "position_encoding_type": "SinusoidalInterleaved",
-      "src_word_vec_size": 1024,
-      "tgt_word_vec_size": 1024
-    },
-    "decoder": {
-      "layer_norm": "standard",
-      "decoder_type": "transformer",
-      "rope_config": null,
-      "tgt_word_vec_size": 1024,
-      "hidden_size": 1024,
-      "add_qkvbias": false,
-      "layers": 2,
-      "add_ffnbias": true,
-      "n_positions": null,
-      "norm_eps": 1e-06,
-      "mlp_activation_fn": "gelu",
-      "heads": 8,
-      "transformer_ff": 4096,
-      "position_encoding_type": "SinusoidalInterleaved"
-    }
+    "bucket_size": 256000,
+    "label_smoothing": 0.1,
+    "model_path": "quickmt-fr-en-eole-model",
+    "valid_batch_size": 2048
   },
   "transforms_configs": {
     "sentencepiece": {
-      "tgt_subword_model": "${MODEL_PATH}/joint.spm.model",
-      "src_subword_model": "${MODEL_PATH}/joint.spm.model"
+      "src_subword_model": "${MODEL_PATH}/fr.spm.model",
+      "tgt_subword_model": "${MODEL_PATH}/en.spm.model"
     },
     "filtertoolong": {
       "src_seq_length": 256,
@@ -129,22 +76,101 @@
   },
   "data": {
     "corpus_1": {
+      "path_src": "fren/train.cleaned.filtered.fr",
+      "path_tgt": "fren/train.cleaned.filtered.en",
+      "path_align": null,
       "transforms": [
         "sentencepiece",
         "filtertoolong"
       ],
+      "weight": 200
+    },
+    "corpus_2": {
+      "path_src": "../data/newscrawl.backtrans.cleaned.filtered.fr",
+      "path_tgt": "../data/newscrawl.backtrans.cleaned.filtered.en",
       "path_align": null,
-      "path_src": "fr-en/train.cleaned.src",
-      "path_tgt": "fr-en/train.cleaned.tgt"
+      "transforms": [
+        "sentencepiece",
+        "filtertoolong"
+      ],
+      "weight": 35
     },
-    "valid": {
+    "corpus_3": {
+      "path_src": "../data/madlad.backtrans.cleaned.filtered.fr",
+      "path_tgt": "../data/madlad.backtrans.cleaned.filtered.en",
+      "path_align": null,
+      "transforms": [
+        "sentencepiece",
+        "filtertoolong"
+      ],
+      "weight": 68
+    },
+    "corpus_4": {
+      "path_src": "../data/hansard.fr",
+      "path_tgt": "../data/hansard.en",
+      "path_align": null,
       "transforms": [
         "sentencepiece",
         "filtertoolong"
       ],
+      "weight": 5
+    },
+    "valid": {
+      "path_src": "fren/dev.fr",
+      "path_tgt": "fren/dev.en",
       "path_align": null,
-      "path_src": "fr-en/dev.src",
-      "path_tgt": "fr-en/dev.tgt"
+      "transforms": [
+        "sentencepiece",
+        "filtertoolong"
+      ]
+    }
+  },
+  "model": {
+    "position_encoding_type": "SinusoidalInterleaved",
+    "share_decoder_embeddings": false,
+    "add_qkvbias": false,
+    "architecture": "transformer",
+    "add_estimator": false,
+    "hidden_size": 768,
+    "share_embeddings": false,
+    "layer_norm": "standard",
+    "add_ffnbias": true,
+    "mlp_activation_fn": "gelu",
+    "heads": 16,
+    "transformer_ff": 4096,
+    "decoder": {
+      "transformer_ff": 4096,
+      "position_encoding_type": "SinusoidalInterleaved",
+      "add_qkvbias": false,
+      "tgt_word_vec_size": 768,
+      "n_positions": null,
+      "decoder_type": "transformer",
+      "hidden_size": 768,
+      "layer_norm": "standard",
+      "add_ffnbias": true,
+      "mlp_activation_fn": "gelu",
+      "heads": 16,
+      "layers": 2
+    },
+    "encoder": {
+      "encoder_type": "transformer",
+      "transformer_ff": 4096,
+      "position_encoding_type": "SinusoidalInterleaved",
+      "src_word_vec_size": 768,
+      "add_qkvbias": false,
+      "n_positions": null,
+      "hidden_size": 768,
+      "layer_norm": "standard",
+      "add_ffnbias": true,
+      "mlp_activation_fn": "gelu",
+      "heads": 16,
+      "layers": 12
+    },
+    "embeddings": {
+      "tgt_word_vec_size": 768,
+      "word_vec_size": 768,
+      "position_encoding_type": "SinusoidalInterleaved",
+      "src_word_vec_size": 768
     }
   }
 }

eole-model/en.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce51817f4aabdac074cccee54167581e681e9cbade82b563d70d64f7be958e4d
+size 799063

eole-model/fr.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e41eb3cab1f4e5caf55370cd23362ace5839cec791411da61d159365b6a6451
+size 816271

eole-model/model.00.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a28ad097a4ed4a2bd3f4b6da5731c5f4d2cf664cc28a54d62d2b150b1f68e0c
-size 762769904
+oid sha256:540388bd6955f125e9e6f79e8dd51aecb54dca6c982a8df30e1cad79bb2a1b1b
+size 829569112

model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a5c0a467de1c354122644c49dc5fad47b4c38b1eeb6ddfa841f3d6e3b2a699b
-size 381336824
+oid sha256:c0657044846861a905f959bb4950a155e0a41c2bf6187a2c47fa86e2a79d3582
+size 407101843

src.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e41eb3cab1f4e5caf55370cd23362ace5839cec791411da61d159365b6a6451
+size 816271

tgt.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce51817f4aabdac074cccee54167581e681e9cbade82b563d70d64f7be958e4d
+size 799063