Revert "Model templates encoder only (huggingface#8509)"

This reverts commit aa2012f.

.github/workflows/self-push.yml

@@ -4,12 +4,10 @@ on:
   push:
     branches:
       - master
-      - model-templates
     paths:
       - "src/**"
       - "tests/**"
       - ".github/**"
-      - "templates/**"
   # pull_request:
   repository_dispatch:
 
@@ -57,14 +55,6 @@ jobs:
           python -c "import torch; print('Cuda available:', torch.cuda.is_available())"
           python -c "import torch; print('Number of GPUs available:', torch.cuda.device_count())"
 
-      - name: Create model files
-        run: |
-          source .env/bin/activate
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/encoder-bert-tokenizer.json --path=templates/cookiecutter
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/pt-encoder-bert-tokenizer.json --path=templates/cookiecutter
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/standalone.json --path=templates/cookiecutter
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/tf-encoder-bert-tokenizer.json --path=templates/cookiecutter
-
       - name: Run all non-slow tests on GPU
         env:
           OMP_NUM_THREADS: 1
@@ -126,14 +116,6 @@ jobs:
           TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
           TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
 
-      - name: Create model files
-        run: |
-          source .env/bin/activate
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/encoder-bert-tokenizer.json --path=templates/cookiecutter
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/pt-encoder-bert-tokenizer.json --path=templates/cookiecutter
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/standalone.json --path=templates/cookiecutter
-          transformers-cli add-new-model --testing --testing_file=templates/cookiecutter/tests/tf-encoder-bert-tokenizer.json --path=templates/cookiecutter
-
       - name: Run all non-slow tests on GPU
         env:
           OMP_NUM_THREADS: 1

setup.py

@@ -98,21 +98,20 @@
 
 extras["tokenizers"] = ["tokenizers==0.9.2"]
 extras["onnxruntime"] = ["onnxruntime>=1.4.0", "onnxruntime-tools>=1.4.2"]
-extras["modelcreation"] = ["cookiecutter==1.7.2"]
 
 extras["serving"] = ["pydantic", "uvicorn", "fastapi", "starlette"]
 
 extras["sentencepiece"] = ["sentencepiece==0.1.91"]
 extras["retrieval"] = ["faiss-cpu", "datasets"]
-extras["testing"] = ["pytest", "pytest-xdist", "timeout-decorator", "parameterized", "psutil"] + extras["retrieval"] + extras["modelcreation"]
+extras["testing"] = ["pytest", "pytest-xdist", "timeout-decorator", "parameterized", "psutil"] + extras["retrieval"]
 # sphinx-rtd-theme==0.5.0 introduced big changes in the style.
 extras["docs"] = ["recommonmark", "sphinx==3.2.1", "sphinx-markdown-tables", "sphinx-rtd-theme==0.4.3", "sphinx-copybutton"]
 extras["quality"] = ["black >= 20.8b1", "isort >= 5.5.4", "flake8 >= 3.8.3"]
 
 
 extras["all"] = extras["tf"] + extras["torch"] + extras["flax"] + extras["sentencepiece"] + extras["tokenizers"]
 
-extras["dev"] = extras["all"] + extras["testing"] + extras["quality"] + extras["ja"] + extras["docs"] + extras["sklearn"] + extras["modelcreation"]
+extras["dev"] = extras["all"] + extras["testing"] + extras["quality"] + extras["ja"] + extras["docs"] + extras["sklearn"]
 
 
 setup(

src/transformers/commands/transformers_cli.py

@@ -1,7 +1,6 @@
 #!/usr/bin/env python
 from argparse import ArgumentParser
 
-from transformers.commands.add_new_model import AddNewModelCommand
 from transformers.commands.convert import ConvertCommand
 from transformers.commands.download import DownloadCommand
 from transformers.commands.env import EnvironmentCommand
@@ -21,7 +20,6 @@ def main():
     RunCommand.register_subcommand(commands_parser)
     ServeCommand.register_subcommand(commands_parser)
     UserCommands.register_subcommand(commands_parser)
-    AddNewModelCommand.register_subcommand(commands_parser)
 
     # Let's go
     args = parser.parse_args()

src/transformers/configuration_auto.py

@@ -59,7 +59,6 @@
 ALL_PRETRAINED_CONFIG_ARCHIVE_MAP = dict(
     (key, value)
     for pretrained_map in [
-        # Add archive maps here
         BERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
         BART_PRETRAINED_CONFIG_ARCHIVE_MAP,
         BLENDERBOT_PRETRAINED_CONFIG_ARCHIVE_MAP,
@@ -96,7 +95,6 @@
 
 CONFIG_MAPPING = OrderedDict(
     [
-        # Add configs here
         ("retribert", RetriBertConfig),
         ("t5", T5Config),
         ("mobilebert", MobileBertConfig),
@@ -138,7 +136,6 @@
 
 MODEL_NAMES_MAPPING = OrderedDict(
     [
-        # Add full (and cased) model names here
         ("retribert", "RetriBERT"),
         ("t5", "T5"),
         ("mobilebert", "MobileBERT"),

src/transformers/modeling_auto.py

@@ -226,14 +226,11 @@
 from .utils import logging
 
 
-# Add modeling imports here
-
 logger = logging.get_logger(__name__)
 
 
 MODEL_MAPPING = OrderedDict(
     [
-        # Base model mapping
         (RetriBertConfig, RetriBertModel),
         (T5Config, T5Model),
         (DistilBertConfig, DistilBertModel),
@@ -269,7 +266,6 @@
 
 MODEL_FOR_PRETRAINING_MAPPING = OrderedDict(
     [
-        # Model for pre-training mapping
         (LayoutLMConfig, LayoutLMForMaskedLM),
         (RetriBertConfig, RetriBertModel),
         (T5Config, T5ForConditionalGeneration),
@@ -299,7 +295,6 @@
 
 MODEL_WITH_LM_HEAD_MAPPING = OrderedDict(
     [
-        # Model with LM heads mapping
         (LayoutLMConfig, LayoutLMForMaskedLM),
         (T5Config, T5ForConditionalGeneration),
         (DistilBertConfig, DistilBertForMaskedLM),
@@ -330,7 +325,6 @@
 
 MODEL_FOR_CAUSAL_LM_MAPPING = OrderedDict(
     [
-        # Model for Causal LM mapping
         (CamembertConfig, CamembertForCausalLM),
         (XLMRobertaConfig, XLMRobertaForCausalLM),
         (RobertaConfig, RobertaForCausalLM),
@@ -353,7 +347,6 @@
 
 MODEL_FOR_MASKED_LM_MAPPING = OrderedDict(
     [
-        # Model for Masked LM mapping
         (LayoutLMConfig, LayoutLMForMaskedLM),
         (DistilBertConfig, DistilBertForMaskedLM),
         (AlbertConfig, AlbertForMaskedLM),
@@ -375,7 +368,6 @@
 
 MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING = OrderedDict(
     [
-        # Model for Seq2Seq Causal LM mapping
         (T5Config, T5ForConditionalGeneration),
         (PegasusConfig, PegasusForConditionalGeneration),
         (MarianConfig, MarianMTModel),
@@ -391,7 +383,6 @@
 
 MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING = OrderedDict(
     [
-        # Model for Sequence Classification mapping
         (DistilBertConfig, DistilBertForSequenceClassification),
         (AlbertConfig, AlbertForSequenceClassification),
         (CamembertConfig, CamembertForSequenceClassification),
@@ -416,7 +407,6 @@
 
 MODEL_FOR_QUESTION_ANSWERING_MAPPING = OrderedDict(
     [
-        # Model for Question Answering mapping
         (DistilBertConfig, DistilBertForQuestionAnswering),
         (AlbertConfig, AlbertForQuestionAnswering),
         (CamembertConfig, CamembertForQuestionAnswering),
@@ -439,7 +429,6 @@
 
 MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING = OrderedDict(
     [
-        # Model for Token Classification mapping
         (LayoutLMConfig, LayoutLMForTokenClassification),
         (DistilBertConfig, DistilBertForTokenClassification),
         (CamembertConfig, CamembertForTokenClassification),
@@ -461,7 +450,6 @@
 
 MODEL_FOR_MULTIPLE_CHOICE_MAPPING = OrderedDict(
     [
-        # Model for Multiple Choice mapping
         (CamembertConfig, CamembertForMultipleChoice),
         (ElectraConfig, ElectraForMultipleChoice),
         (XLMRobertaConfig, XLMRobertaForMultipleChoice),