Upload folder using huggingface_hub

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,105 @@
+---
+language:
+- en
+- da
+tags:
+- translation
+license: cc-by-4.0
+datasets:
+- quickmt/quickmt-train.da-en
+model-index:
+- name: quickmt-en-da
+  results:
+  - task:
+      name: Translation eng-dan
+      type: translation
+      args: eng-dan
+    dataset:
+      name: flores101-devtest
+      type: flores_101
+      args: eng_Latn dan_Latn devtest
+    metrics:
+    - name: BLEU
+      type: bleu
+      value: 46.61
+    - name: CHRF
+      type: chrf
+      value: 70.07
+    - name: COMET
+      type: comet
+      value: 89.49
+---
+
+
+# `quickmt-en-da` Neural Machine Translation Model 
+
+`quickmt-en-da` is a reasonably fast and reasonably accurate neural machine translation model for translation from `en` into `da`.
+
+
+## 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 transformer 'big' with 8 encoder layers and 2 decoder layers
+* 32k separate Sentencepiece vocabs
+* Exported for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
+
+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 ./quickmt/
+
+quickmt-model-download quickmt/quickmt-en-da ./quickmt-en-da
+```
+
+Finally use the model in python:
+
+```python
+from quickmt import Translator
+
+# Auto-detects GPU, set to "cpu" to force CPU inference
+t = Translator("./quickmt-en-da/", device="auto")
+
+# Translate - set beam size to 1 for faster speed (but lower quality)
+sample_text = '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 cautioned that the research is still in its early days.'
+
+t(sample_text, beam_size=5)
+```
+
+> 'Dr. Ehud Ur, professor i medicin ved Dalhousie University i Halifax, Nova Scotia og formand for den kliniske og videnskabelige afdeling af Canadian Diabetes Association advarede om, at forskningen stadig er i sine tidlige dage.'
+
+```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)
+```
+
+> 'Dr. Ehud Ur, professor i medicin på Dalhousie University i Halifax, Nova Scotia og formand for det kliniske og videnskabelige afdeling af Canadian Diabetes Association advarede om, at forskningen stadig er i sin tidlige dage.'
+
+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) ("eng_Latn"->"dan_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 the flores-devtest dataset (1012 sentences) on an Nvidia RTX 4070s GPU with batch size 32.
+
+
+|                                  |   bleu |   chrf2 |   comet22 |   Time (s) |
+|:---------------------------------|-------:|--------:|----------:|-----------:|
+| quickmt/quickmt-en-da            |  46.61 |   70.07 |     89.49 |       1.19 |
+| facebook/nllb-200-distilled-600M |  41.8  |   66.79 |     89.44 |      22.24 |
+| facebook/nllb-200-distilled-1.3B |  44.02 |   68.52 |     90.73 |      39.32 |
+| facebook/m2m100_418M             |  36.81 |   62.93 |     85.35 |      18.97 |
+| facebook/m2m100_1.2B             |  44.54 |   68.46 |     89.43 |      37.42 |

.ipynb_checkpoints/eole-config-checkpoint.yaml

@@ -0,0 +1,97 @@
+## IO
+save_data: data
+overwrite: True
+seed: 1234
+report_every: 100
+valid_metrics: ["BLEU"]
+tensorboard: true
+tensorboard_log_dir: tensorboard
+
+### Vocab
+src_vocab: en.eole.vocab
+tgt_vocab: da.eole.vocab
+src_vocab_size: 32000
+tgt_vocab_size: 32000
+vocab_size_multiple: 8
+share_vocab: false
+n_sample: 0
+
+data:
+    corpus_1:
+        path_src: hf://quickmt/quickmt-train.da-en/en
+        path_tgt: hf://quickmt/quickmt-train.da-en/da
+        path_sco: hf://quickmt/quickmt-train.da-en/sco
+
+    valid:
+        path_src: valid.en
+        path_tgt: valid.da
+
+transforms: [sentencepiece, filtertoolong]
+transforms_configs:
+  sentencepiece:
+    src_subword_model: "en.spm.model"
+    tgt_subword_model: "da.spm.model"
+  filtertoolong:
+    src_seq_length: 256
+    tgt_seq_length: 256
+
+training:
+    # Run configuration
+    model_path: quickmt-en-da-eole-model
+    #train_from: model
+    keep_checkpoint: 4
+    train_steps: 100000
+    save_checkpoint_steps: 5000
+    valid_steps: 5000
+    
+    # Train on a single GPU
+    world_size: 1
+    gpu_ranks: [0]
+
+    # Batching 10240
+    batch_type: "tokens"
+    batch_size: 6000
+    valid_batch_size: 2048
+    batch_size_multiple: 8
+    accum_count: [20]
+    accum_steps: [0]
+
+    # Optimizer & Compute
+    compute_dtype: "fp16"
+    optim: "adamw"
+    #use_amp: False
+    learning_rate: 3.0
+    warmup_steps: 5000
+    decay_method: "noam"
+    adam_beta2: 0.998
+
+    # Data loading
+    bucket_size: 128000
+    num_workers: 4
+    prefetch_factor: 32
+    
+    # Hyperparams
+    dropout_steps: [0]
+    dropout: [0.1]
+    attention_dropout: [0.1]
+    max_grad_norm: 0
+    label_smoothing: 0.1
+    average_decay: 0.0001
+    param_init_method: xavier_uniform
+    normalization: "tokens"
+
+model:
+    architecture: "transformer"
+    share_embeddings: false
+    share_decoder_embeddings: true
+    hidden_size: 1024
+    encoder:
+        layers: 8
+    decoder:
+        layers: 2
+    heads: 8
+    transformer_ff: 4096
+    embeddings:
+        word_vec_size: 1024
+        position_encoding_type: "SinusoidalInterleaved"
+

README.md

@@ -1,3 +1,105 @@
----
-license: cc-by-4.0
----
+---
+language:
+- en
+- da
+tags:
+- translation
+license: cc-by-4.0
+datasets:
+- quickmt/quickmt-train.da-en
+model-index:
+- name: quickmt-en-da
+  results:
+  - task:
+      name: Translation eng-dan
+      type: translation
+      args: eng-dan
+    dataset:
+      name: flores101-devtest
+      type: flores_101
+      args: eng_Latn dan_Latn devtest
+    metrics:
+    - name: BLEU
+      type: bleu
+      value: 46.61
+    - name: CHRF
+      type: chrf
+      value: 70.07
+    - name: COMET
+      type: comet
+      value: 89.49
+---
+
+
+# `quickmt-en-da` Neural Machine Translation Model 
+
+`quickmt-en-da` is a reasonably fast and reasonably accurate neural machine translation model for translation from `en` into `da`.
+
+
+## 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 transformer 'big' with 8 encoder layers and 2 decoder layers
+* 32k separate Sentencepiece vocabs
+* Exported for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
+
+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 ./quickmt/
+
+quickmt-model-download quickmt/quickmt-en-da ./quickmt-en-da
+```
+
+Finally use the model in python:
+
+```python
+from quickmt import Translator
+
+# Auto-detects GPU, set to "cpu" to force CPU inference
+t = Translator("./quickmt-en-da/", device="auto")
+
+# Translate - set beam size to 1 for faster speed (but lower quality)
+sample_text = '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 cautioned that the research is still in its early days.'
+
+t(sample_text, beam_size=5)
+```
+
+> 'Dr. Ehud Ur, professor i medicin ved Dalhousie University i Halifax, Nova Scotia og formand for den kliniske og videnskabelige afdeling af Canadian Diabetes Association advarede om, at forskningen stadig er i sine tidlige dage.'
+
+```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)
+```
+
+> 'Dr. Ehud Ur, professor i medicin på Dalhousie University i Halifax, Nova Scotia og formand for det kliniske og videnskabelige afdeling af Canadian Diabetes Association advarede om, at forskningen stadig er i sin tidlige dage.'
+
+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) ("eng_Latn"->"dan_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 the flores-devtest dataset (1012 sentences) on an Nvidia RTX 4070s GPU with batch size 32.
+
+
+|                                  |   bleu |   chrf2 |   comet22 |   Time (s) |
+|:---------------------------------|-------:|--------:|----------:|-----------:|
+| quickmt/quickmt-en-da            |  46.61 |   70.07 |     89.49 |       1.19 |
+| facebook/nllb-200-distilled-600M |  41.8  |   66.79 |     89.44 |      22.24 |
+| facebook/nllb-200-distilled-1.3B |  44.02 |   68.52 |     90.73 |      39.32 |
+| facebook/m2m100_418M             |  36.81 |   62.93 |     85.35 |      18.97 |
+| facebook/m2m100_1.2B             |  44.54 |   68.46 |     89.43 |      37.42 |

config.json

@@ -0,0 +1,10 @@
+{
+  "add_source_bos": false,
+  "add_source_eos": false,
+  "bos_token": "<s>",
+  "decoder_start_token": "<s>",
+  "eos_token": "</s>",
+  "layer_norm_epsilon": 1e-06,
+  "multi_query_attention": false,
+  "unk_token": "<unk>"
+}

eole-config.yaml

@@ -0,0 +1,97 @@
+## IO
+save_data: data
+overwrite: True
+seed: 1234
+report_every: 100
+valid_metrics: ["BLEU"]
+tensorboard: true
+tensorboard_log_dir: tensorboard
+
+### Vocab
+src_vocab: en.eole.vocab
+tgt_vocab: da.eole.vocab
+src_vocab_size: 32000
+tgt_vocab_size: 32000
+vocab_size_multiple: 8
+share_vocab: false
+n_sample: 0
+
+data:
+    corpus_1:
+        path_src: hf://quickmt/quickmt-train.da-en/en
+        path_tgt: hf://quickmt/quickmt-train.da-en/da
+        path_sco: hf://quickmt/quickmt-train.da-en/sco
+
+    valid:
+        path_src: valid.en
+        path_tgt: valid.da
+
+transforms: [sentencepiece, filtertoolong]
+transforms_configs:
+  sentencepiece:
+    src_subword_model: "en.spm.model"
+    tgt_subword_model: "da.spm.model"
+  filtertoolong:
+    src_seq_length: 256
+    tgt_seq_length: 256
+
+training:
+    # Run configuration
+    model_path: quickmt-en-da-eole-model
+    #train_from: model
+    keep_checkpoint: 4
+    train_steps: 100000
+    save_checkpoint_steps: 5000
+    valid_steps: 5000
+    
+    # Train on a single GPU
+    world_size: 1
+    gpu_ranks: [0]
+
+    # Batching 10240
+    batch_type: "tokens"
+    batch_size: 6000
+    valid_batch_size: 2048
+    batch_size_multiple: 8
+    accum_count: [20]
+    accum_steps: [0]
+
+    # Optimizer & Compute
+    compute_dtype: "fp16"
+    optim: "adamw"
+    #use_amp: False
+    learning_rate: 3.0
+    warmup_steps: 5000
+    decay_method: "noam"
+    adam_beta2: 0.998
+
+    # Data loading
+    bucket_size: 128000
+    num_workers: 4
+    prefetch_factor: 32
+    
+    # Hyperparams
+    dropout_steps: [0]
+    dropout: [0.1]
+    attention_dropout: [0.1]
+    max_grad_norm: 0
+    label_smoothing: 0.1
+    average_decay: 0.0001
+    param_init_method: xavier_uniform
+    normalization: "tokens"
+
+model:
+    architecture: "transformer"
+    share_embeddings: false
+    share_decoder_embeddings: true
+    hidden_size: 1024
+    encoder:
+        layers: 8
+    decoder:
+        layers: 2
+    heads: 8
+    transformer_ff: 4096
+    embeddings:
+        word_vec_size: 1024
+        position_encoding_type: "SinusoidalInterleaved"
+

eole-model/config.json

@@ -0,0 +1,132 @@
+{
+  "src_vocab_size": 32000,
+  "tgt_vocab": "da.eole.vocab",
+  "report_every": 100,
+  "seed": 1234,
+  "tensorboard_log_dir_dated": "tensorboard/Nov-02_18-34-56",
+  "tgt_vocab_size": 32000,
+  "transforms": [
+    "sentencepiece",
+    "filtertoolong"
+  ],
+  "n_sample": 0,
+  "save_data": "data",
+  "vocab_size_multiple": 8,
+  "share_vocab": false,
+  "tensorboard_log_dir": "tensorboard",
+  "valid_metrics": [
+    "BLEU"
+  ],
+  "src_vocab": "en.eole.vocab",
+  "overwrite": true,
+  "tensorboard": true,
+  "training": {
+    "compute_dtype": "torch.float16",
+    "prefetch_factor": 32,
+    "batch_size": 6000,
+    "num_workers": 0,
+    "adam_beta2": 0.998,
+    "model_path": "quickmt-en-da-eole-model",
+    "label_smoothing": 0.1,
+    "learning_rate": 3.0,
+    "param_init_method": "xavier_uniform",
+    "bucket_size": 128000,
+    "world_size": 1,
+    "accum_count": [
+      20
+    ],
+    "batch_size_multiple": 8,
+    "dropout_steps": [
+      0
+    ],
+    "accum_steps": [
+      0
+    ],
+    "decay_method": "noam",
+    "batch_type": "tokens",
+    "average_decay": 0.0001,
+    "normalization": "tokens",
+    "max_grad_norm": 0.0,
+    "train_steps": 100000,
+    "dropout": [
+      0.1
+    ],
+    "gpu_ranks": [
+      0
+    ],
+    "keep_checkpoint": 4,
+    "attention_dropout": [
+      0.1
+    ],
+    "optim": "adamw",
+    "warmup_steps": 5000,
+    "save_checkpoint_steps": 5000,
+    "valid_batch_size": 2048,
+    "valid_steps": 5000
+  },
+  "model": {
+    "architecture": "transformer",
+    "position_encoding_type": "SinusoidalInterleaved",
+    "share_decoder_embeddings": true,
+    "transformer_ff": 4096,
+    "share_embeddings": false,
+    "heads": 8,
+    "hidden_size": 1024,
+    "decoder": {
+      "decoder_type": "transformer",
+      "position_encoding_type": "SinusoidalInterleaved",
+      "n_positions": null,
+      "transformer_ff": 4096,
+      "heads": 8,
+      "tgt_word_vec_size": 1024,
+      "layers": 2,
+      "hidden_size": 1024
+    },
+    "encoder": {
+      "position_encoding_type": "SinusoidalInterleaved",
+      "n_positions": null,
+      "transformer_ff": 4096,
+      "heads": 8,
+      "src_word_vec_size": 1024,
+      "encoder_type": "transformer",
+      "layers": 8,
+      "hidden_size": 1024
+    },
+    "embeddings": {
+      "src_word_vec_size": 1024,
+      "position_encoding_type": "SinusoidalInterleaved",
+      "tgt_word_vec_size": 1024,
+      "word_vec_size": 1024
+    }
+  },
+  "data": {
+    "corpus_1": {
+      "path_tgt": "train.da",
+      "path_src": "train.en",
+      "path_align": null,
+      "transforms": [
+        "sentencepiece",
+        "filtertoolong"
+      ]
+    },
+    "valid": {
+      "path_tgt": "valid.da",
+      "path_src": "valid.en",
+      "path_align": null,
+      "transforms": [
+        "sentencepiece",
+        "filtertoolong"
+      ]
+    }
+  },
+  "transforms_configs": {
+    "filtertoolong": {
+      "tgt_seq_length": 256,
+      "src_seq_length": 256
+    },
+    "sentencepiece": {
+      "tgt_subword_model": "${MODEL_PATH}/da.spm.model",
+      "src_subword_model": "${MODEL_PATH}/en.spm.model"
+    }
+  }
+}

eole-model/da.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:688ecf3cf51d2cf7497720ba36e5f3c6bdfea75630c2c2c8afcf8a0de3580d06
+size 814697

eole-model/en.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eda8069d07624408f632a155bbfb6db9794b8bb7bdbc9040c8ec16d499b6ef2
+size 800426

eole-model/model.00.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0efd749a46f05b8062f4a8d3535449b90a373e3b64d60b9240a2c541870c6dbb
+size 840314816

model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfdd52ec90e4661131251f5a8a15882c2774df6a7a9970c3298b1c938f5b9681
+size 409915789

src.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eda8069d07624408f632a155bbfb6db9794b8bb7bdbc9040c8ec16d499b6ef2
+size 800426

tgt.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:688ecf3cf51d2cf7497720ba36e5f3c6bdfea75630c2c2c8afcf8a0de3580d06
+size 814697