Create concept guide section (huggingface#16369)

* ✨ create concept guide section * 🖍 make fixup * 🖍 apply feedback Co-authored-by: Steven <[email protected]>
natecain · Mar 25, 2022 · b320d87 · b320d87
1 parent ed2ee37
commit b320d87
Show file tree

Hide file tree

Showing 8 changed files with 113 additions and 815 deletions.
diff --git a/docs/source/_toctree.yml b/docs/source/_toctree.yml
@@ -5,10 +5,6 @@
     title: Quick tour
   - local: installation
     title: Installation
-  - local: philosophy
-    title: Philosophy
-  - local: glossary
-    title: Glossary
   title: Get started
 - sections:
   - local: pipeline_tutorial
@@ -17,30 +13,20 @@
     title: Load pretrained instances with an AutoClass
   - local: preprocessing
     title: Preprocess
-  - local: task_summary
-    title: Summary of the tasks
-  - local: model_summary
-    title: Summary of the models
   - local: training
     title: Fine-tune a pretrained model
   - local: accelerate
     title: Distributed training with 🤗 Accelerate
   - local: model_sharing
     title: Share a model
-  - local: tokenizer_summary
-    title: Summary of the tokenizers
-  - local: multilingual
-    title: Multi-lingual models
   title: Tutorials
 - sections:
+  - local: fast_tokenizers
+    title: "Use tokenizers from 🤗 Tokenizers"
   - local: create_a_model
-    title: Create a custom model
-  - local: multilingual
-    title: Inference for multilingual models
-  - local: troubleshooting
-    title: Troubleshoot
-  - local: custom_datasets
-    title: Fine-tuning with custom datasets
+    title: Create a custom architecture
+  - local: custom_models
+    title: Sharing custom models
   - sections:
     - local: tasks/sequence_classification
       title: Text classification
@@ -65,47 +51,59 @@
     title: Fine-tune for downstream tasks
   - local: run_scripts
     title: Train with a script
-  - local: notebooks
-    title: "🤗 Transformers Notebooks"
   - local: sagemaker
     title: Run training on Amazon SageMaker
-  - local: community
-    title: Community
+  - local: multilingual
+    title: Inference for multilingual models
   - local: converting_tensorflow_models
-    title: Converting Tensorflow Checkpoints
+    title: Converting TensorFlow Checkpoints
+  - local: serialization
+    title: Export 🤗 Transformers models
+  - local: performance
+    title: 'Performance and Scalability: How To Fit a Bigger Model and Train It Faster'
+  - local: parallelism
+    title: Model Parallelism
+  - local: benchmarks
+    title: Benchmarks
   - local: migration
     title: Migrating from previous packages
+  - local: troubleshooting
+    title: Troubleshoot
+  - local: debugging
+    title: Debugging
+  - local: notebooks
+    title: "🤗 Transformers Notebooks"
+  - local: community
+    title: Community
   - local: contributing
     title: How to contribute to transformers?
   - local: add_new_model
     title: "How to add a model to 🤗 Transformers?"
   - local: add_new_pipeline
     title: "How to add a pipeline to 🤗 Transformers?"
-  - local: fast_tokenizers
-    title: "Using tokenizers from 🤗 Tokenizers"
-  - local: performance
-    title: 'Performance and Scalability: How To Fit a Bigger Model and Train It Faster'
-  - local: parallelism
-    title: Model Parallelism
   - local: testing
     title: Testing
-  - local: debugging
-    title: Debugging
-  - local: serialization
-    title: Exporting 🤗 Transformers models
-  - local: custom_models
-    title: Sharing custom models
   - local: pr_checks
     title: Checks on a Pull Request
   title: How-to guides
 - sections:
+  - local: philosophy
+    title: Philosophy
+  - local: glossary
+    title: Glossary
+  - local: task_summary
+    title: Summary of the tasks
+  - local: model_summary
+    title: Summary of the models
+  - local: tokenizer_summary
+    title: Summary of the tokenizers
+  - local: pad_truncation
+    title: Padding and truncation
   - local: bertology
     title: BERTology
   - local: perplexity
     title: Perplexity of fixed-length models
-  - local: benchmarks
-    title: Benchmarks
-  title: Research
+  title: Conceptual guides
 - sections:
   - sections:
     - local: main_classes/callback

diff --git a/docs/source/create_a_model.mdx b/docs/source/create_a_model.mdx
@@ -10,7 +10,7 @@ an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express o
 specific language governing permissions and limitations under the License.
 -->
 
-# Create a custom model
+# Create a custom architecture
 
 An [`AutoClass`](model_doc/auto) automatically infers the model architecture and downloads pretrained configuration and weights. Generally, we recommend using an `AutoClass` to produce checkpoint-agnostic code. But users who want more control over specific model parameters can create a custom 🤗 Transformers model from just a few base classes. This could be particularly useful for anyone who is interested in studying, training or experimenting with a 🤗 Transformers model. In this guide, dive deeper into creating a custom model without an `AutoClass`. Learn how to: