fix py script

Author: JyotinderSingh

guides/quantization/overview.py

@@ -21,11 +21,10 @@
 * Joint weight + activation PTQ in `int4`, `int8`, and `float8`.
 * Weight-only PTQ via **GPTQ** (2/3/4/8-bit) to maximize compression with minimal accuracy impact, especially for large language models (LLMs).
 
-> **Terminology**
->
-> * *Scale / zero-point:* Quantization maps real values `x` to integers `q` using a scale (and optionally a zero-point). Symmetric schemes use only a scale.
-> * *Per-channel vs per-tensor:* A separate scale per output channel (e.g., per hidden unit) usually preserves accuracy better than a single scale for the whole tensor.
-> * *Calibration:* A short pass over sample data to estimate activation ranges (e.g., max absolute value).
+**Terminology**
+* *Scale / zero-point:* Quantization maps real values `x` to integers `q` using a scale (and optionally a zero-point). Symmetric schemes use only a scale.
+* *Per-channel vs per-tensor:* A separate scale per output channel (e.g., per hidden unit) usually preserves accuracy better than a single scale for the whole tensor.
+* *Calibration:* A short pass over sample data to estimate activation ranges (e.g., max absolute value).
 
 
 ## Quantization Modes
@@ -56,11 +55,11 @@
   * **Why use it:** Strong accuracy retention at very low bit-widths without retraining; ideal for rapid LLM compression.
   * **What to expect:** Large storage/VRAM savings with small perplexity/accuracy deltas on many decoder-only models when calibrated on task-relevant samples.
 
-> **Implementation notes**
->
-> * For `int4`, Keras packs signed 4-bit values (range ≈ [−8, 7]) and stores per-channel scales such as `kernel_scale`. Dequantization happens on the fly, and matmuls use 8-bit (unpacked) kernels.
-> * Activation scaling for `int4` / `int8` / `float8` uses **AbsMax calibration** by default (range set by the maximum absolute value observed). Alternative calibration methods (e.g., percentile) may be added in future releases.
-> * Per-channel scaling is the default for weights where supported, because it materially improves accuracy at negligible overhead.
+**Implementation notes**
+
+* For `int4`, Keras packs signed 4-bit values (range ≈ [−8, 7]) and stores per-channel scales such as `kernel_scale`. Dequantization happens on the fly, and matmuls use 8-bit (unpacked) kernels.
+* Activation scaling for `int4` / `int8` / `float8` uses **AbsMax calibration** by default (range set by the maximum absolute value observed). Alternative calibration methods (e.g., percentile) may be added in future releases.
+* Per-channel scaling is the default for weights where supported, because it materially improves accuracy at negligible overhead.
 
 ## Quantizing Keras Models
 
@@ -134,8 +133,8 @@
 
 Since all KerasHub models subclass `keras.Model`, they automatically support the `model.quantize(...)` API. In practice, this means you can take a popular LLM preset, call a single function to obtain an int8/int4/GPTQ-quantized variant, and then save or serve it—without changing your training code.
 
-> **Practical guidance**
->
-> * For GPTQ, use a calibration set that matches your inference domain (a few hundred to a few thousand tokens is often enough to see strong retention).
-> * Measure both **VRAM** and **throughput/latency**: memory savings are immediate; speedups depend on the availability of fused low-precision kernels on your device.
+**Practical guidance**
+
+* For GPTQ, use a calibration set that matches your inference domain (a few hundred to a few thousand tokens is often enough to see strong retention).
+* Measure both **VRAM** and **throughput/latency**: memory savings are immediate; speedups depend on the availability of fused low-precision kernels on your device.
 """