Upload folder using huggingface_hub

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,99 @@
+---
+language:
+- en
+- ur
+tags:
+- translation
+license: cc-by-4.0
+datasets:
+- quickmt/quickmt-train.ur-en
+model-index:
+- name: quickmt-ur-en
+  results:
+  - task:
+      name: Translation urd-eng
+      type: translation
+      args: urd-eng
+    dataset:
+      name: flores101-devtest
+      type: flores_101
+      args: urd_Arab eng_Latn devtest
+    metrics:
+    - name: BLEU
+      type: bleu
+      value: 31.48
+    - name: CHRF
+      type: chrf
+      value: 58.35
+    - name: COMET
+      type: comet
+      value: 84.25
+---
+
+
+# `quickmt-ur-en` Neural Machine Translation Model 
+
+`quickmt-ur-en` is a reasonably fast and reasonably accurate neural machine translation model for translation from `ur` into `en`.
+
+
+## Model Information
+
+* Trained using [`eole`](https://github.com/eole-nlp/eole)
+* 195M parameter transformer 'big' with 8 encoder layers and 2 decoder layers
+* 20k seperate Sentencepiece vocabs
+* Expested for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
+* Training data: https://huggingface.co/datasets/quickmt/quickmt-train.ur-en/tree/main
+
+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-ur-en ./quickmt-ur-en
+```
+
+Finally use the model in python:
+
+```python
+from quickmt impest Translator
+
+# Auto-detects GPU, set to "cpu" to force CPU inference
+t = Translator("./quickmt-ur-en/", device="auto")
+
+# Translate - set beam size to 1 for faster speed (but lower quality)
+sample_text = 'ہالیفیکس، نووا اسکاٹیا کی ڈلہوزی یونیورسٹی میں طب کے پروفیسر اور کینیڈین ڈائبیٹک ایسوسی ایشن کے طبی و سائنسی ڈویژن کے صدر ڈاکٹر ایہود یوآر نے متنبہ کیا ہے کہ تحقیق ابھی تک اپنے ابتدائی مراحل میں ہے۔'
+
+t(sample_text, beam_size=5)
+```
+
+> 'Dr Ehud UR, a professor of medicine at Dalhousie University in Halifax, Nova Scotia and president of the Medical and Scientific Division of the Canadian Diabetes Association, warned that the research is still in its early stages.'
+
+```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 Uyer, professor of medicine at Dalhousie University in Halifax, Nova Scotia, president of the Medical and Science Division of the Canadian Dietetic Association, warned that research was still "a preliminary course" ahead.'
+
+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`.
+
+## 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) ("urd_Arab"->"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 the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32 (faster speed is possible using a larger batch size).
+
+|                                  |   bleu |   chrf2 |   comet22 |   Time (s) |
+|:---------------------------------|-------:|--------:|----------:|-----------:|
+| quickmt/quickmt-ur-en            |  31.48 |   58.35 |     84.25 |       1.28 |
+| Helsinki-NLP/opus-mt-ur-en       |  10.37 |   35.21 |     62.54 |       4.79 |
+| facebook/nllb-200-distilled-600M |  33.02 |   60.04 |     86.21 |      21.2  |
+| facebook/nllb-200-distilled-1.3B |  36.68 |   62.83 |     87.55 |      36.97 |
+| facebook/m2m100_418M             |  15.65 |   43.64 |     69.94 |      19.58 |
+| facebook/m2m100_1.2B             |  23.78 |   51.67 |     79.44 |      35.35 |

.ipynb_checkpoints/eole-config-checkpoint.yaml

@@ -0,0 +1,96 @@
+## IO
+save_data: data
+overwrite: True
+seed: 1234
+report_every: 100
+valid_metrics: ["BLEU"]
+tensorboard: true
+tensorboard_log_dir: tensorboard
+
+### Vocab
+src_vocab: ur.eole.vocab
+tgt_vocab: en.eole.vocab
+src_vocab_size: 20000
+tgt_vocab_size: 20000
+vocab_size_multiple: 8
+share_vocab: false
+n_sample: 0
+
+data:
+    corpus_1:
+        path_src: hf://quickmt/quickmt-train.ur-en/ur
+        path_tgt: hf://quickmt/quickmt-train.ur-en/en
+        path_sco: hf://quickmt/quickmt-train.ur-en/sco
+    valid:
+        path_src: valid.ur
+        path_tgt: valid.en
+
+transforms: [sentencepiece, filtertoolong]
+transforms_configs:
+  sentencepiece:
+    src_subword_model: "ur.spm.model"
+    tgt_subword_model: "en.spm.model"
+  filtertoolong:
+    src_seq_length: 256
+    tgt_seq_length: 256
+
+training:
+    # Run configuration
+    model_path: quickmt-ur-en-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: 8000
+    valid_batch_size: 4096
+    batch_size_multiple: 8
+    accum_count: [10]
+    accum_steps: [0]
+
+    # Optimizer & Compute
+    compute_dtype: "fp16"
+    optim: "adamw"
+    #use_amp: False
+    learning_rate: 2.0
+    warmup_steps: 4000
+    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: false
+    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 +1,99 @@
----
-license: cc-by-4.0
----
+---
+language:
+- en
+- ur
+tags:
+- translation
+license: cc-by-4.0
+datasets:
+- quickmt/quickmt-train.ur-en
+model-index:
+- name: quickmt-ur-en
+  results:
+  - task:
+      name: Translation urd-eng
+      type: translation
+      args: urd-eng
+    dataset:
+      name: flores101-devtest
+      type: flores_101
+      args: urd_Arab eng_Latn devtest
+    metrics:
+    - name: BLEU
+      type: bleu
+      value: 31.48
+    - name: CHRF
+      type: chrf
+      value: 58.35
+    - name: COMET
+      type: comet
+      value: 84.25
+---
+
+
+# `quickmt-ur-en` Neural Machine Translation Model 
+
+`quickmt-ur-en` is a reasonably fast and reasonably accurate neural machine translation model for translation from `ur` into `en`.
+
+
+## Model Information
+
+* Trained using [`eole`](https://github.com/eole-nlp/eole)
+* 195M parameter transformer 'big' with 8 encoder layers and 2 decoder layers
+* 20k seperate Sentencepiece vocabs
+* Expested for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
+* Training data: https://huggingface.co/datasets/quickmt/quickmt-train.ur-en/tree/main
+
+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-ur-en ./quickmt-ur-en
+```
+
+Finally use the model in python:
+
+```python
+from quickmt impest Translator
+
+# Auto-detects GPU, set to "cpu" to force CPU inference
+t = Translator("./quickmt-ur-en/", device="auto")
+
+# Translate - set beam size to 1 for faster speed (but lower quality)
+sample_text = 'ہالیفیکس، نووا اسکاٹیا کی ڈلہوزی یونیورسٹی میں طب کے پروفیسر اور کینیڈین ڈائبیٹک ایسوسی ایشن کے طبی و سائنسی ڈویژن کے صدر ڈاکٹر ایہود یوآر نے متنبہ کیا ہے کہ تحقیق ابھی تک اپنے ابتدائی مراحل میں ہے۔'
+
+t(sample_text, beam_size=5)
+```
+
+> 'Dr Ehud UR, a professor of medicine at Dalhousie University in Halifax, Nova Scotia and president of the Medical and Scientific Division of the Canadian Diabetes Association, warned that the research is still in its early stages.'
+
+```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 Uyer, professor of medicine at Dalhousie University in Halifax, Nova Scotia, president of the Medical and Science Division of the Canadian Dietetic Association, warned that research was still "a preliminary course" ahead.'
+
+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`.
+
+## 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) ("urd_Arab"->"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 the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32 (faster speed is possible using a larger batch size).
+
+|                                  |   bleu |   chrf2 |   comet22 |   Time (s) |
+|:---------------------------------|-------:|--------:|----------:|-----------:|
+| quickmt/quickmt-ur-en            |  31.48 |   58.35 |     84.25 |       1.28 |
+| Helsinki-NLP/opus-mt-ur-en       |  10.37 |   35.21 |     62.54 |       4.79 |
+| facebook/nllb-200-distilled-600M |  33.02 |   60.04 |     86.21 |      21.2  |
+| facebook/nllb-200-distilled-1.3B |  36.68 |   62.83 |     87.55 |      36.97 |
+| facebook/m2m100_418M             |  15.65 |   43.64 |     69.94 |      19.58 |
+| facebook/m2m100_1.2B             |  23.78 |   51.67 |     79.44 |      35.35 |

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 +1,96 @@
+## IO
+save_data: data
+overwrite: True
+seed: 1234
+report_every: 100
+valid_metrics: ["BLEU"]
+tensorboard: true
+tensorboard_log_dir: tensorboard
+
+### Vocab
+src_vocab: ur.eole.vocab
+tgt_vocab: en.eole.vocab
+src_vocab_size: 20000
+tgt_vocab_size: 20000
+vocab_size_multiple: 8
+share_vocab: false
+n_sample: 0
+
+data:
+    corpus_1:
+        path_src: hf://quickmt/quickmt-train.ur-en/ur
+        path_tgt: hf://quickmt/quickmt-train.ur-en/en
+        path_sco: hf://quickmt/quickmt-train.ur-en/sco
+    valid:
+        path_src: valid.ur
+        path_tgt: valid.en
+
+transforms: [sentencepiece, filtertoolong]
+transforms_configs:
+  sentencepiece:
+    src_subword_model: "ur.spm.model"
+    tgt_subword_model: "en.spm.model"
+  filtertoolong:
+    src_seq_length: 256
+    tgt_seq_length: 256
+
+training:
+    # Run configuration
+    model_path: quickmt-ur-en-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: 8000
+    valid_batch_size: 4096
+    batch_size_multiple: 8
+    accum_count: [10]
+    accum_steps: [0]
+
+    # Optimizer & Compute
+    compute_dtype: "fp16"
+    optim: "adamw"
+    #use_amp: False
+    learning_rate: 2.0
+    warmup_steps: 4000
+    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: false
+    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 @@
+{
+  "valid_metrics": [
+    "BLEU"
+  ],
+  "transforms": [
+    "sentencepiece",
+    "filtertoolong"
+  ],
+  "report_every": 100,
+  "src_vocab_size": 20000,
+  "vocab_size_multiple": 8,
+  "tensorboard": true,
+  "tgt_vocab_size": 20000,
+  "overwrite": true,
+  "tensorboard_log_dir_dated": "tensorboard/Jun-17_13-39-10",
+  "save_data": "data",
+  "src_vocab": "ur.eole.vocab",
+  "seed": 1234,
+  "share_vocab": false,
+  "n_sample": 0,
+  "tgt_vocab": "en.eole.vocab",
+  "tensorboard_log_dir": "tensorboard",
+  "training": {
+    "dropout": [
+      0.1
+    ],
+    "learning_rate": 2.0,
+    "batch_type": "tokens",
+    "normalization": "tokens",
+    "max_grad_norm": 0.0,
+    "accum_steps": [
+      0
+    ],
+    "label_smoothing": 0.1,
+    "num_workers": 0,
+    "valid_steps": 5000,
+    "warmup_steps": 4000,
+    "bucket_size": 128000,
+    "average_decay": 0.0001,
+    "param_init_method": "xavier_uniform",
+    "keep_checkpoint": 4,
+    "save_checkpoint_steps": 5000,
+    "prefetch_factor": 32,
+    "adam_beta2": 0.998,
+    "batch_size_multiple": 8,
+    "batch_size": 8000,
+    "attention_dropout": [
+      0.1
+    ],
+    "gpu_ranks": [
+      0
+    ],
+    "dropout_steps": [
+      0
+    ],
+    "accum_count": [
+      10
+    ],
+    "valid_batch_size": 4096,
+    "optim": "adamw",
+    "compute_dtype": "torch.float16",
+    "decay_method": "noam",
+    "world_size": 1,
+    "train_steps": 100000,
+    "model_path": "quickmt-ur-en-eole-model"
+  },
+  "data": {
+    "corpus_1": {
+      "transforms": [
+        "sentencepiece",
+        "filtertoolong"
+      ],
+      "path_tgt": "train.en",
+      "path_align": null,
+      "path_src": "train.ur"
+    },
+    "valid": {
+      "transforms": [
+        "sentencepiece",
+        "filtertoolong"
+      ],
+      "path_tgt": "valid.en",
+      "path_align": null,
+      "path_src": "valid.ur"
+    }
+  },
+  "model": {
+    "hidden_size": 1024,
+    "share_decoder_embeddings": false,
+    "architecture": "transformer",
+    "position_encoding_type": "SinusoidalInterleaved",
+    "share_embeddings": false,
+    "heads": 8,
+    "transformer_ff": 4096,
+    "embeddings": {
+      "tgt_word_vec_size": 1024,
+      "src_word_vec_size": 1024,
+      "position_encoding_type": "SinusoidalInterleaved",
+      "word_vec_size": 1024
+    },
+    "decoder": {
+      "tgt_word_vec_size": 1024,
+      "hidden_size": 1024,
+      "n_positions": null,
+      "layers": 2,
+      "position_encoding_type": "SinusoidalInterleaved",
+      "decoder_type": "transformer",
+      "heads": 8,
+      "transformer_ff": 4096
+    },
+    "encoder": {
+      "hidden_size": 1024,
+      "encoder_type": "transformer",
+      "n_positions": null,
+      "layers": 8,
+      "src_word_vec_size": 1024,
+      "position_encoding_type": "SinusoidalInterleaved",
+      "heads": 8,
+      "transformer_ff": 4096
+    }
+  },
+  "transforms_configs": {
+    "sentencepiece": {
+      "tgt_subword_model": "${MODEL_PATH}/en.spm.model",
+      "src_subword_model": "${MODEL_PATH}/ur.spm.model"
+    },
+    "filtertoolong": {
+      "src_seq_length": 256,
+      "tgt_seq_length": 256
+    }
+  }
+}

eole-model/en.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27f59c3dc803ba0fc30ac42666969bbf9ea2316c4e3bfc2ac9456127a0c745ee
+size 587475

eole-model/model.00.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07da6e1c561b698f84976ccee82daadde8ab0fe3b6d7fd3d1a1d4bdab81623bf
+size 823882912

eole-model/ur.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a9592d23b51ae0fafbdd8b1ee171039731ebaa1689b31164384d17bc7d3af2d
+size 624096

model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3276d2649bc9f5443751ecdf4dcfd9620c5490bae6904e2a318a539cdf14327b
+size 401699775

src.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a9592d23b51ae0fafbdd8b1ee171039731ebaa1689b31164384d17bc7d3af2d
+size 624096

tgt.spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27f59c3dc803ba0fc30ac42666969bbf9ea2316c4e3bfc2ac9456127a0c745ee
+size 587475