README update for DELG training (#9112)

* Merged commit includes the following changes:
326369548  by Andre Araujo:

    Fix import issues.

--
326159826  by Andre Araujo:

    Changed the implementation of the cosine weights from Keras layer to tf.Variable to manually control for L2 normalization.

--
326139082  by Andre Araujo:

    Support local feature matching using ratio test.

    To allow for easily choosing which matching type to use, we rename a flag/argument and modify all related files to avoid breakages.

    Also include a small change when computing nearest neighbors for geometric matching, to parallelize computation, which saves a little bit of time during execution (argument "n_jobs=-1").

--
326119848  by Andre Araujo:

    Option to measure DELG latency taking binarization into account.

--
324316608  by Andre Araujo:

    DELG global features training.

--
323693131  by Andre Araujo:

    PY3 conversion for delf public lib.

--
321046157  by Andre Araujo:

    Purely Google refactor

--

PiperOrigin-RevId: 326369548

* Added export of delg_model module.

* README update to explain training DELG global features head
Co-authored-by: Andre Araujo <andrearaujo@google.com>

research/delf/delf/python/training/README.md

@@ -164,6 +164,22 @@ the batch size to `256`:
 --batch_size=256
 ```
 
+It is also possible to train the model with an improved global features head as
+introduced in the [DELG paper](https://arxiv.org/abs/2001.05027). To do this,
+specify the additional parameter `--delg_global_features` when launching the 
+training, like in the following example:
+
+```
+python3 train.py \
+  --train_file_pattern=gldv2_dataset/tfrecord/train* \
+  --validation_file_pattern=gldv2_dataset/tfrecord/validation* \
+  --imagenet_checkpoint=resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5 \
+  --dataset_version=gld_v2_clean \
+  --logdir=gldv2_training/ \
+  --delg_global_features
+```
+
+
 ## Exporting the Trained Model
 
 Assuming the training output, the TensorFlow checkpoint, is in the