Update README.md with new model card content

README.md

@@ -1,20 +1,182 @@
 ---
 library_name: keras-hub
 ---
-This is a [`Mistral` model](https://keras.io/api/keras_hub/models/mistral) uploaded using the KerasHub library and can be used with JAX, TensorFlow, and PyTorch backends.
-Model config:
-* **name:** mistral_backbone_1
-* **trainable:** True
-* **vocabulary_size:** 32000
-* **num_layers:** 32
-* **num_query_heads:** 32
-* **hidden_dim:** 4096
-* **intermediate_dim:** 14336
-* **rope_max_wavelength:** 1000000.0
-* **rope_scaling_factor:** 1.0
-* **num_key_value_heads:** 8
-* **sliding_window:** None
-* **layer_norm_epsilon:** 1e-05
-* **dropout:** 0
-
-This model card has been generated automatically and should be completed by the model author. See [Model Cards documentation](https://huggingface.co/docs/hub/model-cards) for more information.
+### Model Overview
+Mistral is a set of large language models published by the Mistral AI team. Both pretrained and instruction tuned models are available with 7 billion parameters. See the model card below for benchmarks, data sources, and intended use cases.
+
+Both weights and Keras model code is released under the [Apache 2 License](https://github.com/keras-team/keras-hub/blob/master/LICENSE).
+
+## Links
+
+* [Mistral 2 Quickstart Notebook](https://www.kaggle.com/code/matthewdwatson/mistral-quickstart)
+* [Mistral 2 API Documentation](https://keras.io/api/keras_hub/models/mistral/)
+* [Mistral 2 Model Card](https://huggingface.co/mistralai/Mistral-7B-v0.1)
+* [KerasHub Beginner Guide](https://keras.io/guides/keras_hub/getting_started/)
+* [KerasHub Model Publishing Guide](https://keras.io/guides/keras_hub/upload/)
+
+## Installation
+
+Keras and KerasHub can be installed with:
+
+```
+pip install -U -q keras-hub
+pip install -U -q keras>=3
+```
+
+Jax, TensorFlow, and Torch come preinstalled in Kaggle Notebooks. For instructions on installing them in another environment see the [Keras Getting Started](https://keras.io/getting_started/) page.
+
+## Presets
+
+The following model checkpoints are provided by the Keras team. Full code examples for each are available below.
+
+| Preset name           | Parameters | Description   |
+|-----------------------|------------|---------------|
+|` mistral_7b_en`          | 7.24B      | 7B base model |
+| `mistral_instruct_7b_en ` | 	7.24B      | 7B instruction-tuned model |
+| `mistral_0.2_instruct_7b_en ` | 	7.24B      | 7B instruction-tuned model version 0.2 |
+
+## Prompts
+
+Mistral "instruct" models are instruction tuned on turn by turn conversations and should be prompted with examples that precisely match the training data. Specifically, you must alternate user and assistant turns that begin and end with special tokens. See the following for an example:
+
+```python
+prompt = """[INST] Hello! [/INST] Hello! How are you? [INST] I'm great. Could you help me with a task? [/INST]
+"""
+```
+
+Base models (without instruct in the name) have no specific prompting structure, and should usually be fine-tuned for a specific task.
+
+### Example Usage
+```python
+import keras
+import keras_hub
+import numpy as np
+```
+
+Use `generate()` to do text generation.
+```python
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset("mistral_0.2_instruct_7b_en", dtype="bfloat16")
+mistral_lm.generate("[INST] What is Keras? [/INST]", max_length=500)
+
+# Generate with batched prompts.
+mistral_lm.generate(["[INST] What is Keras? [/INST]", "[INST] Give me your best brownie recipe. [/INST]"], max_length=500)
+```
+
+Compile the `generate()` function with a custom sampler.
+```python
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset("mistral_0.2_instruct_7b_en", dtype="bfloat16")
+mistral_lm.compile(sampler="greedy")
+mistral_lm.generate("I want to say", max_length=30)
+
+mistral_lm.compile(sampler=keras_hub.samplers.BeamSampler(num_beams=2))
+mistral_lm.generate("I want to say", max_length=30)
+```
+
+Use `generate()` without preprocessing.
+```python
+prompt = {
+    # `1` maps to the start token followed by "I want to say".
+    "token_ids": np.array([[1, 315, 947, 298, 1315, 0, 0, 0, 0, 0]] * 2),
+    # Use `"padding_mask"` to indicate values that should not be overridden.
+    "padding_mask": np.array([[1, 1, 1, 1, 1, 0, 0, 0, 0, 0]] * 2),
+}
+
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset(
+    "mistral_0.2_instruct_7b_en",
+    preprocessor=None,
+    dtype="bfloat16"
+)
+mistral_lm.generate(prompt)
+```
+
+Call `fit()` on a single batch.
+```python
+features = ["The quick brown fox jumped.", "I forgot my homework."]
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset("mistral_0.2_instruct_7b_en", dtype="bfloat16")
+mistral_lm.fit(x=features, batch_size=2)
+```
+
+Call `fit()` without preprocessing.
+```python
+x = {
+    "token_ids": np.array([[1, 315, 947, 298, 1315, 369, 315, 837, 0, 0]] * 2),
+    "padding_mask": np.array([[1, 1, 1, 1, 1, 1, 1, 1, 0, 0]] * 2),
+}
+y = np.array([[315, 947, 298, 1315, 369, 315, 837, 0, 0, 0]] * 2)
+sw = np.array([[1, 1, 1, 1, 1, 1, 1, 0, 0, 0]] * 2)
+
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset(
+    "mistral_0.2_instruct_7b_en",
+    preprocessor=None,
+    dtype="bfloat16"
+)
+mistral_lm.fit(x=x, y=y, sample_weight=sw, batch_size=2)
+```
+
+## Example Usage with Hugging Face URI
+
+```python
+import keras
+import keras_hub
+import numpy as np
+```
+
+Use `generate()` to do text generation.
+```python
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset("hf://keras/mistral_0.2_instruct_7b_en", dtype="bfloat16")
+mistral_lm.generate("[INST] What is Keras? [/INST]", max_length=500)
+
+# Generate with batched prompts.
+mistral_lm.generate(["[INST] What is Keras? [/INST]", "[INST] Give me your best brownie recipe. [/INST]"], max_length=500)
+```
+
+Compile the `generate()` function with a custom sampler.
+```python
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset("hf://keras/mistral_0.2_instruct_7b_en", dtype="bfloat16")
+mistral_lm.compile(sampler="greedy")
+mistral_lm.generate("I want to say", max_length=30)
+
+mistral_lm.compile(sampler=keras_hub.samplers.BeamSampler(num_beams=2))
+mistral_lm.generate("I want to say", max_length=30)
+```
+
+Use `generate()` without preprocessing.
+```python
+prompt = {
+    # `1` maps to the start token followed by "I want to say".
+    "token_ids": np.array([[1, 315, 947, 298, 1315, 0, 0, 0, 0, 0]] * 2),
+    # Use `"padding_mask"` to indicate values that should not be overridden.
+    "padding_mask": np.array([[1, 1, 1, 1, 1, 0, 0, 0, 0, 0]] * 2),
+}
+
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset(
+    "hf://keras/mistral_0.2_instruct_7b_en",
+    preprocessor=None,
+    dtype="bfloat16"
+)
+mistral_lm.generate(prompt)
+```
+
+Call `fit()` on a single batch.
+```python
+features = ["The quick brown fox jumped.", "I forgot my homework."]
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset("hf://keras/mistral_0.2_instruct_7b_en", dtype="bfloat16")
+mistral_lm.fit(x=features, batch_size=2)
+```
+
+Call `fit()` without preprocessing.
+```python
+x = {
+    "token_ids": np.array([[1, 315, 947, 298, 1315, 369, 315, 837, 0, 0]] * 2),
+    "padding_mask": np.array([[1, 1, 1, 1, 1, 1, 1, 1, 0, 0]] * 2),
+}
+y = np.array([[315, 947, 298, 1315, 369, 315, 837, 0, 0, 0]] * 2)
+sw = np.array([[1, 1, 1, 1, 1, 1, 1, 0, 0, 0]] * 2)
+
+mistral_lm = keras_hub.models.MistralCausalLM.from_preset(
+    "hf://keras/mistral_0.2_instruct_7b_en",
+    preprocessor=None,
+    dtype="bfloat16"
+)
+mistral_lm.fit(x=x, y=y, sample_weight=sw, batch_size=2)
+```