diff --git a/part2/README.md b/part2/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..30d9e44f56bec382c73dba8285607f230c84a41a
--- /dev/null
+++ b/part2/README.md
@@ -0,0 +1,8 @@
+# Efficient inference: Model compression and hls4ml
+
+In this part, we will start from the data and model you trained in Part 1. We will train a new model quantization aware using QKeras, compare the model performance to that of Part 1 and build the FPGA firmware of this quantized, sparse model using hls4ml.
+
+We assume that you have already participated in Part 1, but if you have not, you can copy the neccessary files from
+```/eos/home-t/thaarres/cms_mlatl1t_tutorial/part1/```
+
+Assuming you have already followed the setup instructions (```bash start_notebooks.sh```), go ahead and run through ```part2_compression.ipynb```!
diff --git a/part2/part2_compression.ipynb b/part2/part2_compression.ipynb
index 31388f21496de3565f7faab81b79cf8d3de7961f..8a1776c9603db79852deddef3a00d688fa12f400 100644
--- a/part2/part2_compression.ipynb
+++ b/part2/part2_compression.ipynb
@@ -70,8 +70,11 @@
    "outputs": [],
    "source": [
     "from tensorflow.keras.models import load_model\n",
+    "import os\n",
     "\n",
-    "model_path = '/eos/home-t/thaarres/cms_mlatl1t_tutorial/full_model.h5'\n",
+    "part1_output_dir = os.environ['MLATL1T_DIR']+'/part1/part1_outputs/'\n",
+    "\n",
+    "model_path =  part1_output_dir + '/model.h5'\n",
     "baseline_model = load_model(model_path)\n",
     "\n",
     "baseline_model.summary()"
@@ -84,6 +87,49 @@
    "source": [
     "So we have 3 hidden layers with [64,32,32] neurons. We don't see it here, but they are all followed by an \"elu\" activation. The output is one node activated by a sigmoid activation function.\n",
     "\n",
+    "# Load the data from Part 1\n",
+    "\n",
+    "Let's also load the data from part one already now so we know what the input shape is for defining our quantized model. Afterwards we'll also further process this input before training it."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c7627ae9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import awkward as ak\n",
+    "import pickle\n",
+    "\n",
+    "X_train = ak.from_parquet(part1_output_dir + \"/X_train_scaled.parquet\").to_numpy() \n",
+    "X_test  = ak.from_parquet(part1_output_dir + \"/X_test_scaled.parquet\").to_numpy() \n",
+    "\n",
+    "y_train = ak.from_parquet(part1_output_dir + \"/y_train_scaled.parquet\").to_numpy()\n",
+    "y_test  = ak.from_parquet(part1_output_dir + \"/y_test_scaled.parquet\").to_numpy()\n",
+    "\n",
+    "# In this case the test and train data is already scaled, but this is how you would laod and apply it:\n",
+    "#Load the scaler and parameters and apply to the data\n",
+    "scale = False\n",
+    "if scale:\n",
+    "    file_path = part1_output_dir+'/scaler.pkl'\n",
+    "\n",
+    "    with open(file_path, 'rb') as file:\n",
+    "        scaler = pickle.load(file)\n",
+    "\n",
+    "    X_train = scaler.transform(X_train)\n",
+    "    X_test  = scaler.transform(X_test);\n",
+    "\n",
+    "\n",
+    "print(f\"Training on {X_train.shape[0]} events, represented by {X_train.shape[1]} input features\")\n",
+    "print(f\"Testing on {X_test.shape[0]} events, represented by {X_test.shape[1]} input features\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "808a79e1",
+   "metadata": {},
+   "source": [
     "## Translating to a QKeras QAT model\n",
     "There are two ways to translate this into a QKeras model that can be trained quantization aware, lets first do it manually:\n",
     "\n",
@@ -106,7 +152,7 @@
     "from qkeras.qlayers import QDense, QActivation\n",
     "from qkeras.quantizers import quantized_bits, quantized_relu\n",
     "\n",
-    "input_size=26\n",
+    "input_size=X_train.shape[1]\n",
     "\n",
     "# Define the input layer\n",
     "inputs = Input(shape=(input_size,))\n",
@@ -167,7 +213,7 @@
     "- ```bits```: The bitwidth, allowing you to have $2^{bits}$ unique values of each weight parameter\n",
     "- ```integers```: How many are integer bits, in this case zero. All 8 bits are used to represent the fractional part of the weight parameter, with no bits dedicated to representing whole numbers. This forces the value to be between -1 and 1. For DNNs this can be useful because the focus is entirely on the precision of the fraction rather than the magnitude of the number. Question: Would this also work on the output node if your algorithm is a regression of the jet mass?\n",
     "- ```symmetric```: should the values be symmetric around 0? In this case it doesnt have to be.\n",
-    "- ```alpha```: with $2^W$ unique values available, we could let them go from [-2^W, 2^W-1] like above, but we can also let them go from $[-2^W*\\alpha, (2^W-1)*\\alpha]$. ```alpha``` is a scaling of the weights. Enabling this often leads to improved performance, but it doesnt talk so nicely to hls4ml, so we recommend leaving it at 1 (or get ready for having to debug)\n",
+    "- ```alpha```: with $2^W$ unique values available, we could let them go from $[-2^W, 2^W-1]$ like above, but we can also let them go from $[-2^W*\\alpha, (2^W-1)*\\alpha]$. ```alpha``` is a scaling of the weights. Enabling this often leads to improved performance, but it doesnt talk so nicely to hls4ml, so we recommend leaving it at 1 (or get ready for having to debug)\n",
     "\n",
     "Having added this, QKeras will automatically apply fake quantization for us during the forward pass, accounting for the quantization error and returning a network that is optimized for the precision you plan on using in hardware.\n",
     "\n",
@@ -188,26 +234,6 @@
     "autoQuant = False\n",
     "\n",
     "if autoQuant:\n",
-    "    # Fine grained, per-layer control\n",
-    "    #  config = {\n",
-    "    #  \"example_model_topo_fc1\": {\n",
-    "    #      \"kernel_quantizer\": \"quantized_bits(8,0,1)\",\n",
-    "    #      \"bias_quantizer\": \"quantized_bits(8,0,1)\",\n",
-    "    #   },                                                           \n",
-    "    #  \"example_model_topo_activation1\": \"quantized_relu(8)\",                                         \n",
-    "\n",
-    "    #  \"example_model_topo_fc2\": {\n",
-    "    #       \"kernel_quantizer\": \"quantized_bits(8,0,1)\",\n",
-    "    #       \"bias_quantizer\": \"quantized_bits(8,0,1)\",\n",
-    "    #   },                                                               \n",
-    "    #  \"example_model_topo_activation2\": \"quantized_relu(8)\",                                                                                                     \n",
-    "    #  \"example_model_topo_fc3\": {\n",
-    "    #       \"kernel_quantizer\": \"quantized_bits(8,0,1)\",\n",
-    "    #       \"bias_quantizer\": \"quantized_bits(8,0,1)\",\n",
-    "    #   },                                                                                                     \n",
-    "    #  example_model_topo_activation3: \"quantized_relu(8)\", \n",
-    "    #  }      \n",
-    "    # Coarse grained, per-layertype quantization\n",
     "    config = {\n",
     "      \"QDense\": {\n",
     "          \"kernel_quantizer\": \"quantized_bits(bits=8, integer=0, symmetric=0, alpha=1)\",\n",
@@ -296,56 +322,21 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "ce0355af",
+   "id": "9092b6db",
    "metadata": {},
    "outputs": [],
    "source": [
-    "import awkward as ak\n",
-    "import pickle\n",
-    "\n",
-    "path = '/eos/home-t/thaarres/cms_mlatl1t_tutorial/'\n",
-    "\n",
-    "X_train = ak.from_parquet(path + \"/X_train.parquet\").to_numpy() \n",
-    "X_test  = ak.from_parquet(path + \"/X_test.parquet\").to_numpy() \n",
-    "\n",
-    "y_train = ak.from_parquet(path + \"/y_train.parquet\").to_numpy()\n",
-    "y_test  = ak.from_parquet(path + \"/y_test.parquet\").to_numpy()\n",
-    "\n",
-    "# In this case the test and train data is already scaled, but this is how you would laod and apply it:\n",
-    "#Load the scaler and parameters and apply to the data\n",
-    "scale = False\n",
-    "if scale:\n",
-    "    file_path = path+'/scaler.pkl'\n",
-    "\n",
-    "    with open(file_path, 'rb') as file:\n",
-    "        scaler = pickle.load(file)\n",
     "\n",
-    "    X_train = scaler.transform(X_train)\n",
-    "    X_test  = scaler.transform(X_test);\n",
-    "\n",
-    "\n",
-    "print(f\"Training on {X_train.shape[0]} events, represented by {X_train.shape[1]} input features\")\n",
-    "print(f\"Testing on {X_test.shape[0]} events, represented by {X_test.shape[1]} input features\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "94d20cfb",
-   "metadata": {},
-   "outputs": [],
-   "source": [
     "import matplotlib.pyplot as plt\n",
     "\n",
-    "bins = 1024\n",
+    "bins = 4096\n",
     "\n",
     "plt.figure(figsize=(10, 6))\n",
-    "\n",
-    "plt.hist(X_train[:1000, :], bins=bins, stacked=True, label=[f'Input {i+1}' for i in range(X_train.shape[1])])\n",
-    "\n",
+    "#Input distribution, stacked per feature. This is very slow to plot, so lets look at all the features flattened later on\n",
+    "plt.hist(X_train, bins=bins, stacked=True, label=[f'Input {i+1}' for i in range(X_train.shape[1])]) \n",
+    "# plt.hist(X_train.flatten(), bins=bins, color='orangered', label='Floating point')\n",
     "plt.xlabel('Feature Value (standardized)')\n",
     "plt.ylabel('Frequency')\n",
-    "plt.title('Stacked Histogram of all features')\n",
     "plt.legend(loc='upper right', ncol=2)\n",
     "plt.semilogy()\n",
     "plt"
@@ -356,7 +347,11 @@
    "id": "a275844b",
    "metadata": {},
    "source": [
-    "In this case, the values seem to be <50 so lets assume 6 integer bits ($2^6=64$). The number of fractional bits will define our precision, and will affect the network performance. Let's assume 10 is sufficient (the smallest increment we can represent is $2^{-10}=0.0009765625$). To make our network adapt to this input precision, we need to \"treat\" our training and testing set with a quantizer to go from FP32 $\\rightarrow <16,6>$:"
+    "In this case, the values seem to be mostly <50, with a few outliers so lets assume 6 integer bits ($2^6=64$) is sufficient (the rest will get clipped). The number of fractional bits will define our precision, and will affect the network performance. Let's assume 10 is sufficient (the smallest increment we can represent is $2^{-10}=0.0009765625$).\n",
+    "\n",
+    "We can evaluate these choices by comparing the accuracy of the network to that in the previous part. \n",
+    "\n",
+    "To make our network adapt to this input precision, we need to \"treat\" our training and testing set with a quantizer to go from FP32 $\\rightarrow <16,6>$:"
    ]
   },
   {
@@ -385,11 +380,11 @@
    "source": [
     "plt.figure(figsize=(10, 6))\n",
     "\n",
-    "plt.hist(qX_train[:1000, :], bins=bins, stacked=True, label=[f'Input {i+1}' for i in range(X_train.shape[1])])\n",
-    "\n",
-    "plt.xlabel('Quantized Feature Value (standardized)')\n",
+    "# plt.hist(qX_train, bins=bins, stacked=True, label=[f'Input {i+1}' for i in range(X_train.shape[1])])\n",
+    "plt.hist(X_train.flatten(), bins=bins, color='orangered', label='Floating point')\n",
+    "plt.hist(qX_train.flatten(), bins=bins, color='royalblue', label='Quantized')\n",
+    "plt.xlabel('Feature Value (standardized)')\n",
     "plt.ylabel('Frequency')\n",
-    "plt.title('Stacked Histogram of all features')\n",
     "plt.legend(loc='upper right', ncol=2)\n",
     "plt.semilogy()\n",
     "plt"
@@ -400,9 +395,12 @@
    "id": "3e0e7438",
    "metadata": {},
    "source": [
-    "The weight distribution looks similar, but we can not really say how much we loose in performance before training with different input precisions.\n",
+    "The weight distribution looks similar, but we can not really say how much we lose in performance before training with different input precisions.\n",
+    "\n",
+    "## Train the network quantization aware\n",
+    "Phew, okay, finally time to train. For this part there are 2 things to note: you need to add a pruning callback and also you might need to adjust the learning rate (like add a learning rate decay). Also, most likely you need to increase the number of epochs.\n",
     "\n",
-    "Phew, okay, finally time to train. For this part there are 2 things to note: you need to add a pruning callback and also you might need to adjust the learning rate (like add a learning rate decay). Let's train!"
+    "Let's train!"
    ]
   },
   {
@@ -425,10 +423,10 @@
     "early_stopping = EarlyStopping(monitor='val_loss', patience=5)\n",
     "callbacks=[early_stopping, reduce_lr, model_checkpoint, pruning_callbacks.UpdatePruningStep()]\n",
     "\n",
-    "adam = Adam(learning_rate=0.0001)\n",
+    "adam = Adam(learning_rate=0.001)\n",
     "qmodel.compile(optimizer=adam, loss=['binary_crossentropy'], metrics=['accuracy'])\n",
     "\n",
-    "qmodel.fit(qX_train, y_train, batch_size=2048, epochs=50,validation_split=0.20, shuffle=True,callbacks=callbacks,verbose=1) \n",
+    "qmodel.fit(qX_train, y_train, batch_size=4096, epochs=60,validation_split=0.20, shuffle=True,callbacks=callbacks,verbose=1) \n",
     "qmodel = strip_pruning(qmodel)\n",
     "qmodel.save('qtopo_model.h5')"
    ]
@@ -438,6 +436,8 @@
    "id": "75409ec1",
    "metadata": {},
    "source": [
+    "## Comparing to he floating point model\n",
+    "\n",
     "Before checking and comparing the accuracy, lets look at the weights and see if they look quantized and pruned:"
    ]
   },
@@ -484,7 +484,11 @@
    "id": "cdd44876",
    "metadata": {},
    "source": [
-    "This looks like expected! Now, lets compare the performance to that of the floating point model:"
+    "This looks quantized and pruned indeed! Now, lets compare the performance to that of the floating point model. \n",
+    "\n",
+    "We are not so interested in false positive rate (FPR) and more interested in the absolute L1 rate, so lets convert it. We will Zoom into the region $<100$ kHz for obvious reasons, which means we are working at a very low FPR. \n",
+    "\n",
+    "Ealuating the performane at such high thresholds will require a lot of stiatistics, which luckily we have:"
    ]
   },
   {
@@ -568,7 +572,7 @@
     "\n",
     "<img src=\"https://gitlab.cern.ch/fastmachinelearning/cms_mlatl1t_tutorial/-/raw/master/part2/images/hls4ml_logo.png?ref_type=heads\" width=\"400\"/>\n",
     "\n",
-    "Time to translate this model into HLS (which we will integrate in the emulator) and use to generate the vhdl to be integrated in the trigger firmware.\n",
+    "Time to translate this model into HLS (which we will integrate in the emulator) and use to generate the vhdl to be integrated in the trigger firmware. We will use the Python library hls4ml for that ([here](https://github.com/fastmachinelearning/hls4ml-tutorial/tree/main) is the hls4ml tutorial).\n",
     "hls4ml seamlessly talks to QKeras, making our jobs way easier for us, but there is still some work for us to do to make sure we get good hardware model accuracy. Lets start!\n",
     "There are a few things I already know in advance and would like my model to include:\n",
     "- Be execuded fully parallel (=unrolled) to reach the lowest possible latency. We set the ReuseFactor=1 and Strategy=Latency\n",
@@ -580,7 +584,7 @@
     "\n",
     "<img src=\"https://gitlab.cern.ch/fastmachinelearning/cms_mlatl1t_tutorial/-/raw/master/part2/images/hls4ml_precisions.png?ref_type=heads\" width=\"400\"/>\n",
     "\n",
-    "Whereas the $weight$ and $bias$ is set to its optimal value from the QKeras model, the accumulator $accum$ and $result$ is set to some default value that might not be optimal for a given model and might need tuning. Let's do a first attemt and compare the ROC curves:"
+    "Whereas the $weight$ and $bias$ is set to its optimal value from the QKeras model, the accumulator $accum$ and $result$ is set to some default value that might not be optimal for a given model and might need tuning. Let's do a first attemt:"
    ]
   },
   {
@@ -629,8 +633,8 @@
     "                                                       project_name='L1TMLDemo_v1', \n",
     "                                                       part='xcu250-figd2104-2L-e', #Target FPGA, ideally you would use VU9P and VU13P that we use in L1T but they are not installed at lxplus, this one is close enought for this\n",
     "                                                       clock_period=2.5, # Target frequency 1/2.5ns = 400 MHz\n",
-    "                                                       input_data_tb='qX_test.npy', # For co-simulation\n",
-    "                                                       output_data_tb='qy_test.npy',# For co-simulation\n",
+    "#                                                        input_data_tb='qX_test.npy', # For co-simulation\n",
+    "#                                                        output_data_tb='qy_test.npy',# For co-simulation\n",
     ")\n",
     "hls_model.compile()"
    ]
@@ -661,7 +665,12 @@
     "Here you can see that the precision is what we set it to be in QKeras as well as what we set manually in the config. One thing to note is the different definitions used in QKeras and in ap_fixed:\n",
     "- ```quantized_bits(8,0) -> ap_fixed<8,1>```\n",
     "- ```quantized_relu(8,0) -> ap_ufixed<8,0>```\n",
-    "Also you can see that the defualt value for result/accu is set to $16,6$. This can also be tuned to more optimal values."
+    "Also you can see that the defualt value for result/accu is set to $16,6$. This can also be tuned to more optimal values.\n",
+    "\n",
+    "## Validate the firmware model accuracy\n",
+    "\n",
+    "#et's also run predict on the C++ implementation of our model and make sure it's the ~same as for the QKeras model.\n",
+    "This is very slow for the C++ implementation of our model, but we need a lot of statistics to probe the low rate region. Keep reading while you wait :)!\n"
    ]
   },
   {
@@ -671,13 +680,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Let's also run predict on the C++ implementation of our model and make sure it's the ~same as for the QKeras model:\n",
-    "# This is very slow for the C++ implementation of our model, so lets only do 1000 events for this\n",
     "y_hls = hls_model.predict(np.ascontiguousarray(qX_test))\n",
     "\n",
-    "print(f\"Truth labels:\\n {y_test[17:27]}\")\n",
-    "print(f\"Qkeras prediction:\\n {qy_pred[17:27]}\")\n",
-    "print(f\"HLS prediction:\\n {y_hls[17:27]}\")"
+    "print(f\"Truth labels:\\n {y_test[17:27]}\\n\")\n",
+    "print(f\"Qkeras prediction:\\n {qy_pred[17:27]}\\n\")\n",
+    "print(f\"HLS prediction:\\n {y_hls[17:27]}\\n\")"
    ]
   },
   {
@@ -688,7 +695,7 @@
    "outputs": [],
    "source": [
     "# Lets plot it!\n",
-    "hlsfpr, hlstpr, hlsthr = roc_curve(y_test, y_hls, pos_label=None, sample_weight=None, drop_intermediate=True)\n",
+    "hlsfpr, hlstpr, hlsthr = roc_curve(y_test, y_hls, pos_label=1, sample_weight=None, drop_intermediate=True)\n",
     "hlsfpr *= totalMinBiasRate()\n",
     "hlsroc_auc = roc_auc_score(y_test, y_hls)\n",
     "\n",
@@ -713,7 +720,13 @@
    "id": "00985383",
    "metadata": {},
    "source": [
-    "Oh! That was easier than expected. If you see the accuracies differing significantly, it's a good idea to look into accumulator and reult precisions. Also with tools like $Trace$ and $Profiling$ that you can learn from in the [official hls4ml tutorial](https://github.com/fastmachinelearning/hls4ml-tutorial/blob/main/part2_advanced_config.ipynb) can be helpful! In this case, it doesnt seem like it's necessary. Now let's build it! Lets run C-synthesis (C++ to register-transfer level), Vivado logic synthesis (gate level representation) and co-simulation (send test vectors, do an exhaustive functional test of the implemented logic)"
+    "Oh! That was easier than expected. If you see the accuracies differing significantly, it's a good idea to look into accumulator and reult precisions. Also with tools like $Trace$ and $Profiling$ that you can learn from in the [official hls4ml tutorial](https://github.com/fastmachinelearning/hls4ml-tutorial/blob/main/part2_advanced_config.ipynb) can be helpful! In this case, it doesnt seem like it's necessary. \n",
+    "\n",
+    "## Synthesise!\n",
+    "\n",
+    "Now let's build it! Lets run C-synthesis (C++ to register-transfer level) and Vivado logic synthesis (gate level representation). We will not do co-simulation (send test vectors, do an exhaustive functional test of the implemented logic), but this can be a good idea if you are using CNNs and the $io_stream$ io. \n",
+    "\n",
+    "Let's run!"
    ]
   },
   {
@@ -723,7 +736,6 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "print(\"Running synthesis!\")\n",
     "report = hls_model.build(csim=False, synth=True, vsynth=True, cosim=False)"
    ]
   },
@@ -732,7 +744,11 @@
    "id": "3893c6cd",
    "metadata": {},
    "source": [
-    "Now, lets, look at the reports! The latency can be extracted from the C-synthesis report, and validated from the co-simulation report (where actual data is sent through the logic. The resource consumption can be extracted from the implementation report (Vivado logic synthesis) and is more accurate then what is quoted in the C-synthesis report:"
+    "Now, lets, look at the reports! The latency can be extracted from the C-synthesis report, and validated from the co-simulation report (where actual data is sent through the logic. \n",
+    "\n",
+    "The resource consumption can be extracted from the implementation report (Vivado logic synthesis) and is more accurate then what is quoted in the C-synthesis report. \n",
+    "\n",
+    "In this case we did not run co-simulation (this mainly because important when using CNNs and io_stream), but lets print the rest:"
    ]
   },
   {
@@ -744,7 +760,7 @@
    "source": [
     "print(\"\\nC synthesis report (latency estimate):\")\n",
     "print_dict(report[\"CSynthesisReport\"])\n",
-    "#print_dict(report[\"CosimReport\"]) Not working due to missing libc header sys/cdefs.h :(?\n",
+    "#print_dict(report[\"CosimReport\"]) # If also running co-sim\n",
     "print(\"\\nVivado synthesis report (resource estimates):\")\n",
     "print_dict(report[\"VivadoSynthReport\"])"
    ]
@@ -754,8 +770,31 @@
    "id": "72c1723a",
    "metadata": {},
    "source": [
-    "A latency of $2.5\\cdot16=40$ ns, that is not bad! Also, the network is using very little resources: 8k out of 1728k LUTs, 15 out of 12k DSPs. This is <1% of the total available resources.  We have a set of HLS files that will be integrated into the CMSSW emulator (```L1TMLDemo_v1/firmware/```) and VHDL that will be integrated into the mGT firmware (```L1TMLDemo_v1/myproject_prj/solution1/impl/vhdl/```). That's next!"
+    "A latency of $2.5\\cdot15=37.5$ ns, that is not bad! \n",
+    "\n",
+    "Also, the network is using very little resources: 5k out of 1728k LUTs, 26 out of 12k DSPs. This is <1% of the total available resources.  We have a set of HLS files that will be integrated into the CMSSW emulator (```L1TMLDemo_v1/firmware/```) and VHDL that will be integrated into the mGT firmware (```L1TMLDemo_v1/myproject_prj/solution1/impl/vhdl/```). That's next!\n",
+    "\n",
+    "If you did not finish synthesising before the start of the next exercise, you can copy an already synthesised project from here:"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "80d4fe1d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# ! cp /eos/home-t/thaarres/cms_mlatl1t_tutorial/L1TMLDemo_v1.tar.gz\n",
+    "# ! tar -xzvf "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "5d954d26",
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
diff --git a/solutions/part2_solutions.ipynb b/solutions/part2_solutions.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..27600bd932d909422e0126830f45cad138c570b1
--- /dev/null
+++ b/solutions/part2_solutions.ipynb
@@ -0,0 +1,1352 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d60b2756",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Install some dependencies\n",
+    "%pip install pyarrow hls4ml pyparser"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "bd9c5bc4",
+   "metadata": {},
+   "source": [
+    "\n",
+    "# Quantization aware training with QKeras\n",
+    "\n",
+    "Quantization is a powerful way to reduce model memory and resource consumption. In this tutorial, we will use the libary QKeras to perform quantization aware training (QAT).\n",
+    "\n",
+    "In contrast to in Keras, where models are trained using floating point precision, QKeras quantizes each of the model weights and activation functions during training, allowing the network to adapt to the numerical precision that will eventually be used on hardware.\n",
+    "\n",
+    "During the forward pass of the network, each floating point weight is put into one of $2^{bitwidth}$ buckets. Which one it goes into is defined through rounding and clipping schemes.\n",
+    "\n",
+    "Below you can see an example of a tensor with a (symmetric) dynamic range of $x_{f}$ $[-amax, amax]$ mapped through quantization to a an 8 bit integer, $2^8=256$ discrete values in the interval $[-128, 127]$ (32-bit floating-point can represent ~4B numbers in the interval $[-3.4e38, 3.40e38]$).\n",
+    "\n",
+    "<img src=\"https://gitlab.cern.ch/fastmachinelearning/cms_mlatl1t_tutorial/-/raw/master/part2/images/8-bit-signed-integer-quantization.png?ref_type=heads\" width=\"800\"/>\n",
+    "\n",
+    "Quantization of floating point numbers can be achieved using the quantization operation\n",
+    "\n",
+    "$$x_{q} = Clip(Round(x_{f}/scale))$$\n",
+    "\n",
+    "where $x_{q}$ is the quantized digit and $x_{f}$ is the floating point digit. $Round$ is a function that applies some rounding scheme to each number and $Clip$ is a function that clips outliers that fall outside the $[-128, 127]$ interval. The $scale$ parameter is obtained by dividing the float-point dynamic-range into 256 equal parts.\n",
+    "\n",
+    "On FPGA, we do not use int8 quantization, but fixed-point quantization, bu the idea is similar. Fixed-point representation is a way to express fractions with integers and offers more control over precision and range. We can split the $W$-bits making up an integer (in our case $W=8$) to represent the integer part of a number and the fractional part of the number. We usually reserve 1-bit representing the sign of the digit. The radix splits the remaining $W-1$ bits to $I$ most significant bits representing the integer value and $F$ least significant bits representing the fraction. We write this as $<W,I>$, where $F=W-1-I$.  Here is an example for an unsigned $<8,3>$:\n",
+    "\n",
+    "<img src=\"https://gitlab.cern.ch/fastmachinelearning/cms_mlatl1t_tutorial/-/raw/master/part2/images/fixedpoint.png?ref_type=heads\" width=\"400\"/>\n",
+    "\n",
+    "\n",
+    "This fixed point number corresponds to $2^4\\cdot0+2^3\\cdot0+2^2\\cdot0+2^1\\cdot1+2^0\\cdot0+2^{-1}\\cdot1+2^{-2}\\cdot1+2^{-3}\\cdot0=2.75$.\n",
+    "\n",
+    "The choice of $I$ and $F$ has to be derived as a trade-off between representation range and precision, where $I$ controls the range and $F$ the precision.\n",
+    "\n",
+    "In the following we will use a bitwidth of 8 and 0 integer bits. Not considering the sign bit, this means that the smallest number you can represent (the precision) and the largest number (the range) is:\n",
+    "\n",
+    "$$ \\rm{Precision}= \\frac{1}{2^{F}}= \\frac{1}{2^8} = 0.00390625$$\n",
+    "$$\\rm{Range}= [-2^0,-2^0-1]=[-1,0] $$\n",
+    "With zero integer bits the largest number you can represent is just below (but not including) 1. For weights in deep neural networks, being constrained to be less than 1 is often a reasonable assumtion.\n",
+    "\n",
+    "\n",
+    "\n",
+    "What QKeras (and other QAT libraries) do, is to include the quantization error during the training, in the following way:\n",
+    "- \"Fake quantize\" the floating-point weights and activations during the forward pass: quantize the weights and use them for the layer operations\n",
+    "- Immediately un-quantize the parameters so the rest of the computations take place in floating-point\n",
+    "- During the backward pass, the gradient of the weights is used to update the floating point weight\n",
+    "- The quantization operation gradient (zero or undefined) is handled by passing the gradient through as is (\"straight through estimator\")\n",
+    "\n",
+    "## Inspect the original model\n",
+    "In the following we will use the QKeras library to add quantizers to our model. First, let's load the baseline model and remind ourselves what the architecture looks like:\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "0e6c684c",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2023-12-07 16:03:13.291675: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
+      "To enable the following instructions: SSE4.1 SSE4.2 AVX AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
+      "/afs/cern.ch/user/t/thaarres/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:500: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.\n",
+      "  setattr(self, word, getattr(machar, word).flat[0])\n",
+      "/afs/cern.ch/user/t/thaarres/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.\n",
+      "  return self._float_to_str(self.smallest_subnormal)\n",
+      "/afs/cern.ch/user/t/thaarres/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:500: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.\n",
+      "  setattr(self, word, getattr(machar, word).flat[0])\n",
+      "/afs/cern.ch/user/t/thaarres/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.\n",
+      "  return self._float_to_str(self.smallest_subnormal)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential_1\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " fc1 (Dense)                 (None, 64)                3648      \n",
+      "                                                                 \n",
+      " relu1 (Activation)          (None, 64)                0         \n",
+      "                                                                 \n",
+      " fc2 (Dense)                 (None, 32)                2080      \n",
+      "                                                                 \n",
+      " relu2 (Activation)          (None, 32)                0         \n",
+      "                                                                 \n",
+      " fc3 (Dense)                 (None, 32)                1056      \n",
+      "                                                                 \n",
+      " relu3 (Activation)          (None, 32)                0         \n",
+      "                                                                 \n",
+      " output (Dense)              (None, 1)                 33        \n",
+      "                                                                 \n",
+      " sigmoid (Activation)        (None, 1)                 0         \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 6,817\n",
+      "Trainable params: 6,817\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2023-12-07 16:03:33.284567: E tensorflow/compiler/xla/stream_executor/cuda/cuda_driver.cc:266] failed call to cuInit: UNKNOWN ERROR (34)\n"
+     ]
+    }
+   ],
+   "source": [
+    "from tensorflow.keras.models import load_model\n",
+    "import os\n",
+    "\n",
+    "part1_output_dir = os.environ['MLATL1T_DIR']+'/part1/part1_outputs/'\n",
+    "\n",
+    "model_path =  part1_output_dir + '/model.h5'\n",
+    "baseline_model = load_model(model_path)\n",
+    "\n",
+    "baseline_model.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "eb84f91a",
+   "metadata": {},
+   "source": [
+    "So we have 3 hidden layers with [64,32,32] neurons. We don't see it here, but they are all followed by an \"elu\" activation. The output is one node activated by a sigmoid activation function.\n",
+    "\n",
+    "# Load the data from Part 1\n",
+    "\n",
+    "Let's also load the data from part one already now so we know what the input shape is for defining our quantized model. Afterwards we'll also further process this input before training it."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "c7627ae9",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Training on 523217 events, represented by 56 input features\n",
+      "Testing on 523218 events, represented by 56 input features\n"
+     ]
+    }
+   ],
+   "source": [
+    "import awkward as ak\n",
+    "import pickle\n",
+    "\n",
+    "X_train = ak.from_parquet(part1_output_dir + \"/X_train_scaled.parquet\").to_numpy() \n",
+    "X_test  = ak.from_parquet(part1_output_dir + \"/X_test_scaled.parquet\").to_numpy() \n",
+    "\n",
+    "y_train = ak.from_parquet(part1_output_dir + \"/y_train_scaled.parquet\").to_numpy()\n",
+    "y_test  = ak.from_parquet(part1_output_dir + \"/y_test_scaled.parquet\").to_numpy()\n",
+    "\n",
+    "# In this case the test and train data is already scaled, but this is how you would laod and apply it:\n",
+    "#Load the scaler and parameters and apply to the data\n",
+    "scale = False\n",
+    "if scale:\n",
+    "    file_path = part1_output_dir+'/scaler.pkl'\n",
+    "\n",
+    "    with open(file_path, 'rb') as file:\n",
+    "        scaler = pickle.load(file)\n",
+    "\n",
+    "    X_train = scaler.transform(X_train)\n",
+    "    X_test  = scaler.transform(X_test);\n",
+    "\n",
+    "\n",
+    "print(f\"Training on {X_train.shape[0]} events, represented by {X_train.shape[1]} input features\")\n",
+    "print(f\"Testing on {X_test.shape[0]} events, represented by {X_test.shape[1]} input features\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "125bda95",
+   "metadata": {},
+   "source": [
+    "## Translating to a QKeras QAT model\n",
+    "There are two ways to translate this into a QKeras model that can be trained quantization aware, lets first do it manually:\n",
+    "\n",
+    "### Manual QKeras model definition:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "d5073f61",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"model\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " input_1 (InputLayer)        [(None, 56)]              0         \n",
+      "                                                                 \n",
+      " qd1 (QDense)                (None, 64)                3648      \n",
+      "                                                                 \n",
+      " qrelu1 (QActivation)        (None, 64)                0         \n",
+      "                                                                 \n",
+      " qd2 (QDense)                (None, 32)                2080      \n",
+      "                                                                 \n",
+      " qrelu2 (QActivation)        (None, 32)                0         \n",
+      "                                                                 \n",
+      " qd3 (QDense)                (None, 32)                1056      \n",
+      "                                                                 \n",
+      " qrelu3 (QActivation)        (None, 32)                0         \n",
+      "                                                                 \n",
+      " logits (QDense)             (None, 1)                 33        \n",
+      "                                                                 \n",
+      " output (Activation)         (None, 1)                 0         \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 6,817\n",
+      "Trainable params: 6,817\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "import tensorflow as tf\n",
+    "from tensorflow.keras.layers import Input, Dense\n",
+    "from tensorflow.keras.models import Model\n",
+    "from tensorflow.keras.regularizers import l1\n",
+    "\n",
+    "from tensorflow.keras.layers import Activation\n",
+    "from qkeras.qlayers import QDense, QActivation\n",
+    "from qkeras.quantizers import quantized_bits, quantized_relu\n",
+    "\n",
+    "input_size=X_train.shape[1]\n",
+    "\n",
+    "# Define the input layer\n",
+    "inputs = Input(shape=(input_size,))\n",
+    "\n",
+    "# Define the three hidden layers and output layer\n",
+    "hidden1 = QDense(\n",
+    "        64,\n",
+    "        name='qd1',\n",
+    "        kernel_quantizer=quantized_bits(bits=8, integer=0, symmetric=0, alpha=1),\n",
+    "        bias_quantizer=quantized_bits(bits=8, integer=0, symmetric=0, alpha=1),\n",
+    "        kernel_initializer='lecun_uniform',\n",
+    "        kernel_regularizer=l1(0.0001),\n",
+    "        ) (inputs)\n",
+    "hidden1 = QActivation(activation=quantized_relu(8), name='qrelu1')(hidden1)\n",
+    "hidden2 = QDense(\n",
+    "        32,\n",
+    "        name='qd2',\n",
+    "        kernel_quantizer=quantized_bits(bits=8, integer=0, symmetric=0, alpha=1),\n",
+    "        bias_quantizer=quantized_bits(bits=8, integer=0, symmetric=0, alpha=1),\n",
+    "        kernel_initializer='lecun_uniform',\n",
+    "        kernel_regularizer=l1(0.0001),\n",
+    "        ) (hidden1)\n",
+    "hidden2 = QActivation(activation=quantized_relu(8), name='qrelu2')(hidden2)\n",
+    "hidden3 = QDense(\n",
+    "        32,\n",
+    "        name='qd3',\n",
+    "        kernel_quantizer=quantized_bits(bits=8, integer=0, symmetric=0, alpha=1),\n",
+    "        bias_quantizer=quantized_bits(bits=8, integer=0, symmetric=0, alpha=1),\n",
+    "        kernel_initializer='lecun_uniform',\n",
+    "        kernel_regularizer=l1(0.0001),\n",
+    "        ) (hidden2)\n",
+    "hidden3 = QActivation(activation=quantized_relu(8), name='qrelu3')(hidden3)\n",
+    "# Define the output layer with a single node, let's be careful with quantizing this one and be a bit more generous\n",
+    "# Some prefer to leave this a Keras Dense layer, but then it requires more manual tuning in the hs4ml part\n",
+    "logits = QDense(1, \n",
+    "        name='logits',\n",
+    "        kernel_quantizer=quantized_bits(bits=13, integer=0, symmetric=0, alpha=1),\n",
+    "        bias_quantizer=quantized_bits(bits=13, integer=0, symmetric=0, alpha=1),\n",
+    "        kernel_initializer='lecun_uniform',\n",
+    "        kernel_regularizer=l1(0.0001),\n",
+    "        ) (hidden3)\n",
+    "\n",
+    "output = Activation(activation='sigmoid', name='output')(logits)\n",
+    "# Create the model\n",
+    "qmodel = Model(inputs=inputs, outputs=output)\n",
+    "\n",
+    "# Model summary\n",
+    "qmodel.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2c429c55",
+   "metadata": {},
+   "source": [
+    "Wait! What is going on here?\n",
+    "The magic happens in ```quantized_bits``` (see implementation [here](https://github.com/google/qkeras/blob/master/qkeras/quantizers.py#L1245)), where the parameters are the following:\n",
+    "- ```bits```: The bitwidth, allowing you to have $2^{bits}$ unique values of each weight parameter\n",
+    "- ```integers```: How many are integer bits, in this case zero. All 8 bits are used to represent the fractional part of the weight parameter, with no bits dedicated to representing whole numbers. This forces the value to be between -1 and 1. For DNNs this can be useful because the focus is entirely on the precision of the fraction rather than the magnitude of the number. Question: Would this also work on the output node if your algorithm is a regression of the jet mass?\n",
+    "- ```symmetric```: should the values be symmetric around 0? In this case it doesnt have to be.\n",
+    "- ```alpha```: with $2^W$ unique values available, we could let them go from $[-2^W, 2^W-1]$ like above, but we can also let them go from $[-2^W*\\alpha, (2^W-1)*\\alpha]$. ```alpha``` is a scaling of the weights. Enabling this often leads to improved performance, but it doesnt talk so nicely to hls4ml, so we recommend leaving it at 1 (or get ready for having to debug)\n",
+    "\n",
+    "Having added this, QKeras will automatically apply fake quantization for us during the forward pass, accounting for the quantization error and returning a network that is optimized for the precision you plan on using in hardware.\n",
+    "\n",
+    "Another thing to notice is that we leave the sigmoid and the final output logit unquantized. This is because this is were we want the values to be very accurate, and it is not going to save us a lot of resources quantizing it.\n",
+    "\n",
+    "\n",
+    "### Automatic model quantization through config\n",
+    " You can also set the quantization for the full model using a model configuration. Sometimes this can be sater if you're using the same quantizer for all layers of the same type. You don't have to use this for this tutorial, we already have a model, but we will leave it here as an example:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "1b138d1f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "autoQuant = False\n",
+    "\n",
+    "if autoQuant:\n",
+    "    config = {\n",
+    "      \"QDense\": {\n",
+    "          \"kernel_quantizer\": \"quantized_bits(bits=8, integer=0, symmetric=0, alpha=1)\",\n",
+    "          \"bias_quantizer\": \"quantized_bits(bits=8, integer=0, symmetric=0, alpha=1)\",\n",
+    "      },\n",
+    "      \"QActivation\": { \"relu\": \"quantized_relu(8)\" }\n",
+    "    }\n",
+    "    from qkeras.utils import model_quantize\n",
+    "\n",
+    "    qmodel = model_quantize(model, config, 4, transfer_weights=True)\n",
+    "\n",
+    "    for layer in qmodel.layers:\n",
+    "        if hasattr(layer, \"kernel_quantizer\"):\n",
+    "            print(layer.name, \"kernel:\", str(layer.kernel_quantizer_internal), \"bias:\", str(layer.bias_quantizer_internal))\n",
+    "        elif hasattr(layer, \"quantizer\"):\n",
+    "            print(layer.name, \"quantizer:\", str(layer.quantizer))\n",
+    "\n",
+    "    print()\n",
+    "    qmodel.summary()\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6947eda4",
+   "metadata": {},
+   "source": [
+    "But be careful that activation functions like softmax/sigmoid and perhaps logit layers you want to keep at full presision doesn't get quantized!"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e59eea22",
+   "metadata": {},
+   "source": [
+    "## But how many bits?\n",
+    "\n",
+    "So now we know how to quantize our models, but how do we know wich precision to choose?\n",
+    "Finding the best number of bits and integer bits to use is non-trivial, and there are two ways we recommend:\n",
+    "- The easiest strategy is to scan over the possible bit widths from binary up to some maximum value and choose the smallest one that still has acceptable accuracy, and this is what we often do. \n",
+    "Code for how to do this can be found [here](https://github.com/thesps/keras-training/blob/qkeras/train/train_scan_models.py#L16), and is illustrated below.\n",
+    "For binary and ternary quantization, we use the special ```binary(alpha=1.0)(x)``` and ```ternary(alpha=1.0)(x)``` quantizers. \n",
+    "\n",
+    "<img src=\"https://gitlab.cern.ch/fastmachinelearning/cms_mlatl1t_tutorial/-/raw/master/part2/images/quant_scan.png?ref_type=heads\" width=\"400\"/>\n",
+    "\n",
+    "- Another thing you can do is to use our library for automatic quantization, [AutoQKeras](https://github.com/google/qkeras/blob/master/notebook/AutoQKeras.ipynb), to find the optimal quantization for each layer. This runs hyperparameter optimisation over quantizers/nodes/filters simultenously. An example can be found at the end of [this notebook](https://github.com/fastmachinelearning/hls4ml-tutorial/blob/main/part6_cnns.ipynb) \"Bonus exercise: Automatic quantization with AutoQKeras\"."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "17da0954",
+   "metadata": {},
+   "source": [
+    "## Pruning\n",
+    "\n",
+    "Besides reducing the numerical precision of all the weights, biases and activations, I also want to remove neurons and synapses that do not contribute much to the network overall decision. We do that by pruning, let's remove 50\\% of the weights (spasity=0.5):"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "195fe6ae",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from tensorflow_model_optimization.python.core.sparsity.keras import prune, pruning_schedule\n",
+    "from tensorflow_model_optimization.sparsity.keras import strip_pruning\n",
+    "\n",
+    "# The training step is one gradient update, or epochs*N_samples/batchsize\n",
+    "pruning_params = {\"pruning_schedule\": pruning_schedule.ConstantSparsity(0.5, begin_step=6000, frequency=10)}\n",
+    "qmodel = prune.prune_low_magnitude(qmodel, **pruning_params)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "35b1494e",
+   "metadata": {},
+   "source": [
+    "## Defining the data input type\n",
+    "Great, we now have our model ready to be trained! There is one last important thing we have to think about and that is the *precision of the input*.  In the L1T, all of the inputs are quantized. For instance, the precision used for the GT is listed [here](https://github.com/cms-l1-globaltrigger/mp7_ugt_legacy/blob/master/doc/scales_inputs_2_ugt/pdf/scales_inputs_2_ugt.pdf).\n",
+    "\n",
+    "Ideally, when you train your network, you use the hardware values that the algorithm will actually receive when running inference in the trigger.\n",
+    "\n",
+    "We saw, however, that the inputs were all scaled to have a mean of zero and variance of one in the previous exercise. That means that the new optimal precision for the inputs have changes and you need to define what the precision will be. Here we will do it by inspection and intuition, and use the same precision for all of the input features. Let's now load, scale the data and look at the input value distribution:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "9092b6db",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<module 'matplotlib.pyplot' from '/cvmfs/cms.cern.ch/slc7_amd64_gcc11/external/py3-matplotlib/3.7.1-437a2eea83d29aac3bc5f3984f238002/lib/python3.9/site-packages/matplotlib/pyplot.py'>"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "bins = 4096\n",
+    "\n",
+    "plt.figure(figsize=(10, 6))\n",
+    "#Input distribution, stacked per feature. This is very slow to plot, so lets look at all the features flattened later on\n",
+    "plt.hist(X_train, bins=bins, stacked=True, label=[f'Input {i+1}' for i in range(X_train.shape[1])]) \n",
+    "# plt.hist(X_train.flatten(), bins=bins, color='orangered', label='Floating point')\n",
+    "plt.xlabel('Feature Value (standardized)')\n",
+    "plt.ylabel('Frequency')\n",
+    "plt.legend(loc='upper right', ncol=2)\n",
+    "plt.semilogy()\n",
+    "plt"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a275844b",
+   "metadata": {},
+   "source": [
+    "In this case, the values seem to be mostly <50, with a few outliers so lets assume 6 integer bits ($2^6=64$) is sufficient (the rest will get clipped). The number of fractional bits will define our precision, and will affect the network performance. Let's assume 10 is sufficient (the smallest increment we can represent is $2^{-10}=0.0009765625$).\n",
+    "\n",
+    "We can evaluate these choices by comparing the accuracy of the network to that in the previous part. \n",
+    "\n",
+    "To make our network adapt to this input precision, we need to \"treat\" our training and testing set with a quantizer to go from FP32 $\\rightarrow <16,6>$:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "dc7be5f8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "input_quantizer = quantized_bits(bits=16, integer=6, symmetric=0, alpha=1)\n",
+    "qX_train = input_quantizer(X_train.astype(np.float32)).numpy()\n",
+    "qX_test = input_quantizer(X_test.astype(np.float32)).numpy()\n",
+    "\n",
+    "# Save the quantized test data and labels to a numpy file, such that it can be used to test the firmware\n",
+    "np.save('qX_test.npy', qX_test)\n",
+    "np.save('qy_test.npy', y_test)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "3de19ae6",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<module 'matplotlib.pyplot' from '/cvmfs/cms.cern.ch/slc7_amd64_gcc11/external/py3-matplotlib/3.7.1-437a2eea83d29aac3bc5f3984f238002/lib/python3.9/site-packages/matplotlib/pyplot.py'>"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.figure(figsize=(10, 6))\n",
+    "\n",
+    "# plt.hist(qX_train, bins=bins, stacked=True, label=[f'Input {i+1}' for i in range(X_train.shape[1])])\n",
+    "plt.hist(X_train.flatten(), bins=bins, color='orangered', label='Floating point')\n",
+    "plt.hist(qX_train.flatten(), bins=bins, color='royalblue', label='Quantized')\n",
+    "plt.xlabel('Feature Value (standardized)')\n",
+    "plt.ylabel('Frequency')\n",
+    "plt.legend(loc='upper right', ncol=2)\n",
+    "plt.semilogy()\n",
+    "plt"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3e0e7438",
+   "metadata": {},
+   "source": [
+    "The weight distribution looks similar, but we can not really say how much we lose in performance before training with different input precisions.\n",
+    "\n",
+    "## Train the network quantization aware\n",
+    "Phew, okay, finally time to train. For this part there are 2 things to note: you need to add a pruning callback and also you might need to adjust the learning rate (like add a learning rate decay). Also, most likely you need to increase the number of epochs.\n",
+    "\n",
+    "Let's train!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "9556e6bf",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/60\n",
+      "103/103 [==============================] - 5s 14ms/step - loss: 0.3274 - accuracy: 0.9273 - val_loss: 0.1402 - val_accuracy: 0.9870 - lr: 0.0010\n",
+      "Epoch 2/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.1138 - accuracy: 0.9913 - val_loss: 0.0988 - val_accuracy: 0.9923 - lr: 0.0010\n",
+      "Epoch 3/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0905 - accuracy: 0.9927 - val_loss: 0.0828 - val_accuracy: 0.9928 - lr: 0.0010\n",
+      "Epoch 4/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0768 - accuracy: 0.9933 - val_loss: 0.0707 - val_accuracy: 0.9932 - lr: 0.0010\n",
+      "Epoch 5/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0666 - accuracy: 0.9934 - val_loss: 0.0622 - val_accuracy: 0.9933 - lr: 0.0010\n",
+      "Epoch 6/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0586 - accuracy: 0.9936 - val_loss: 0.0551 - val_accuracy: 0.9935 - lr: 0.0010\n",
+      "Epoch 7/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0525 - accuracy: 0.9938 - val_loss: 0.0493 - val_accuracy: 0.9937 - lr: 0.0010\n",
+      "Epoch 8/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0478 - accuracy: 0.9938 - val_loss: 0.0453 - val_accuracy: 0.9937 - lr: 0.0010\n",
+      "Epoch 9/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0442 - accuracy: 0.9938 - val_loss: 0.0423 - val_accuracy: 0.9937 - lr: 0.0010\n",
+      "Epoch 10/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0413 - accuracy: 0.9939 - val_loss: 0.0395 - val_accuracy: 0.9938 - lr: 0.0010\n",
+      "Epoch 11/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0388 - accuracy: 0.9939 - val_loss: 0.0371 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 12/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0368 - accuracy: 0.9939 - val_loss: 0.0353 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 13/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0352 - accuracy: 0.9939 - val_loss: 0.0338 - val_accuracy: 0.9939 - lr: 0.0010\n",
+      "Epoch 14/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0338 - accuracy: 0.9939 - val_loss: 0.0328 - val_accuracy: 0.9939 - lr: 0.0010\n",
+      "Epoch 15/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0325 - accuracy: 0.9940 - val_loss: 0.0311 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 16/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0314 - accuracy: 0.9939 - val_loss: 0.0301 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 17/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0305 - accuracy: 0.9940 - val_loss: 0.0295 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 18/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0297 - accuracy: 0.9940 - val_loss: 0.0287 - val_accuracy: 0.9941 - lr: 0.0010\n",
+      "Epoch 19/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0290 - accuracy: 0.9939 - val_loss: 0.0279 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 20/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0284 - accuracy: 0.9940 - val_loss: 0.0273 - val_accuracy: 0.9941 - lr: 0.0010\n",
+      "Epoch 21/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0279 - accuracy: 0.9940 - val_loss: 0.0271 - val_accuracy: 0.9939 - lr: 0.0010\n",
+      "Epoch 22/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0275 - accuracy: 0.9939 - val_loss: 0.0265 - val_accuracy: 0.9941 - lr: 0.0010\n",
+      "Epoch 23/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0271 - accuracy: 0.9939 - val_loss: 0.0262 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 24/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0268 - accuracy: 0.9939 - val_loss: 0.0257 - val_accuracy: 0.9941 - lr: 0.0010\n",
+      "Epoch 25/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0264 - accuracy: 0.9939 - val_loss: 0.0257 - val_accuracy: 0.9939 - lr: 0.0010\n",
+      "Epoch 26/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0261 - accuracy: 0.9940 - val_loss: 0.0253 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 27/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0259 - accuracy: 0.9940 - val_loss: 0.0252 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 28/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0258 - accuracy: 0.9940 - val_loss: 0.0251 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 29/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0256 - accuracy: 0.9940 - val_loss: 0.0246 - val_accuracy: 0.9940 - lr: 0.0010\n",
+      "Epoch 30/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0253 - accuracy: 0.9940 - val_loss: 0.0244 - val_accuracy: 0.9941 - lr: 0.0010\n",
+      "Epoch 31/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0251 - accuracy: 0.9939 - val_loss: 0.0244 - val_accuracy: 0.9941 - lr: 0.0010\n",
+      "Epoch 32/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0250 - accuracy: 0.9941 - val_loss: 0.0248 - val_accuracy: 0.9938 - lr: 0.0010\n",
+      "Epoch 33/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0249 - accuracy: 0.9940 - val_loss: 0.0248 - val_accuracy: 0.9936 - lr: 0.0010\n",
+      "Epoch 34/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0246 - accuracy: 0.9941 - val_loss: 0.0239 - val_accuracy: 0.9942 - lr: 1.0000e-04\n",
+      "Epoch 35/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0244 - accuracy: 0.9941 - val_loss: 0.0239 - val_accuracy: 0.9941 - lr: 1.0000e-04\n",
+      "Epoch 36/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0244 - accuracy: 0.9941 - val_loss: 0.0239 - val_accuracy: 0.9942 - lr: 1.0000e-04\n",
+      "Epoch 37/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0244 - accuracy: 0.9941 - val_loss: 0.0239 - val_accuracy: 0.9941 - lr: 1.0000e-04\n",
+      "Epoch 38/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0244 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-05\n",
+      "Epoch 39/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0244 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-05\n",
+      "Epoch 40/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0244 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-05\n",
+      "Epoch 41/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9941 - lr: 1.0000e-06\n",
+      "Epoch 42/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9941 - lr: 1.0000e-06\n",
+      "Epoch 43/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9941 - lr: 1.0000e-06\n",
+      "Epoch 44/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-07\n",
+      "Epoch 45/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9941 - lr: 1.0000e-07\n",
+      "Epoch 46/60\n",
+      "103/103 [==============================] - 1s 9ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-07\n",
+      "Epoch 47/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-08\n",
+      "Epoch 48/60\n",
+      "103/103 [==============================] - 1s 8ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-08\n",
+      "Epoch 49/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-08\n",
+      "Epoch 50/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-09\n",
+      "Epoch 51/60\n",
+      "103/103 [==============================] - 1s 7ms/step - loss: 0.0243 - accuracy: 0.9941 - val_loss: 0.0238 - val_accuracy: 0.9942 - lr: 1.0000e-09\n",
+      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
+     ]
+    }
+   ],
+   "source": [
+    "from tensorflow.keras.optimizers import Adam\n",
+    "from tensorflow.keras.callbacks import EarlyStopping\n",
+    "from tensorflow.keras.callbacks import ReduceLROnPlateau\n",
+    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
+    "\n",
+    "from tensorflow_model_optimization.python.core.sparsity.keras import pruning_callbacks\n",
+    "\n",
+    "model_checkpoint = ModelCheckpoint('model_best_checkpoint.h5', save_best_only=True, monitor='val_loss')\n",
+    "# This might result in returning a not fully pruned model, but that's okay!\n",
+    "reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.1, patience=3)\n",
+    "early_stopping = EarlyStopping(monitor='val_loss', patience=5)\n",
+    "callbacks=[early_stopping, reduce_lr, model_checkpoint, pruning_callbacks.UpdatePruningStep()]\n",
+    "\n",
+    "adam = Adam(learning_rate=0.001)\n",
+    "qmodel.compile(optimizer=adam, loss=['binary_crossentropy'], metrics=['accuracy'])\n",
+    "\n",
+    "qmodel.fit(qX_train, y_train, batch_size=4096, epochs=60,validation_split=0.20, shuffle=True,callbacks=callbacks,verbose=1) \n",
+    "qmodel = strip_pruning(qmodel)\n",
+    "qmodel.save('qtopo_model.h5')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "75409ec1",
+   "metadata": {},
+   "source": [
+    "## Comparing to he floating point model\n",
+    "\n",
+    "Before checking and comparing the accuracy, lets look at the weights and see if they look quantized and pruned:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "402b4267",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Layer qd1: % of zeros = 0.0\n",
+      "Layer qd2: % of zeros = 0.0\n",
+      "Layer qd3: % of zeros = 0.0\n",
+      "Layer logits: % of zeros = 0.0\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<module 'matplotlib.pyplot' from '/cvmfs/cms.cern.ch/slc7_amd64_gcc11/external/py3-matplotlib/3.7.1-437a2eea83d29aac3bc5f3984f238002/lib/python3.9/site-packages/matplotlib/pyplot.py'>"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "colors  = ['#7b3294','#c2a5cf','#a6dba0','#008837']\n",
+    "# TAKE EVERY OPPORTUNITY TO ADVERTISE COLORBLIND SAFE PLOTS :)\n",
+    "\n",
+    "allWeightsByLayer = {}\n",
+    "for layer in qmodel.layers:\n",
+    "    layername = layer._name\n",
+    "    if len(layer.get_weights())<1:\n",
+    "      continue\n",
+    "    weights=layer.weights[0].numpy().flatten()  \n",
+    "    allWeightsByLayer[layername] = weights\n",
+    "    print('Layer {}: % of zeros = {}'.format(layername,np.sum(weights==0)/np.size(weights)))\n",
+    "labelsW = []\n",
+    "histosW = []\n",
+    "  \n",
+    "for key in reversed(sorted(allWeightsByLayer.keys())):\n",
+    "    labelsW.append(key)\n",
+    "    histosW.append(allWeightsByLayer[key])\n",
+    "\n",
+    "fig = plt.figure()\n",
+    "ax = fig.add_subplot()\n",
+    "plt.semilogy()\n",
+    "plt.legend(loc='upper left',fontsize=15,frameon=False)\n",
+    "bins = np.linspace(-1, 1, 1024) \n",
+    "ax.hist(histosW,bins,histtype='stepfilled',stacked=True,label=labelsW,color=colors)#, edgecolor='black')\n",
+    "ax.legend(frameon=False,loc='upper left')\n",
+    "axis = plt.gca()\n",
+    "plt.ylabel('Number of Weights')\n",
+    "plt.xlabel('Weights')\n",
+    "plt"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "cdd44876",
+   "metadata": {},
+   "source": [
+    "This looks quantized and pruned indeed! Now, lets compare the performance to that of the floating point model. \n",
+    "\n",
+    "We are not so interested in false positive rate (FPR) and more interested in the absolute L1 rate, so lets convert it. We will Zoom into the region $<100$ kHz for obvious reasons, which means we are working at a very low FPR. \n",
+    "\n",
+    "Ealuating the performane at such high thresholds will require a lot of stiatistics, which luckily we have:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "6fdf547a",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "128/128 [==============================] - 0s 1ms/step\n",
+      "128/128 [==============================] - 1s 4ms/step\n"
+     ]
+    }
+   ],
+   "source": [
+    "y_pred  = baseline_model.predict(X_test, batch_size = 4096)\n",
+    "qy_pred = qmodel.predict(qX_test, batch_size = 4096)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "49ea6534",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.metrics import roc_curve, roc_auc_score\n",
+    "\n",
+    "assert(len(y_test) == len(y_pred) == len(qy_pred)), \"Inconsistent predicted and true!\"\n",
+    "fpr, tpr, thr = roc_curve(y_test, y_pred, pos_label=None, sample_weight=None, drop_intermediate=True)\n",
+    "roc_auc = roc_auc_score(y_test, y_pred)\n",
+    "\n",
+    "qfpr, qtpr, qthr = roc_curve(y_test, qy_pred, pos_label=None, sample_weight=None, drop_intermediate=True)\n",
+    "qroc_auc = roc_auc_score(y_test, qy_pred)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "41a6d24c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Lets also convert from FPR to L1 rate:\n",
+    "\n",
+    "def totalMinBiasRate():\n",
+    "\n",
+    "    LHCfreq = 11245.6\n",
+    "    nCollBunch = 2544\n",
+    "\n",
+    "    return LHCfreq * nCollBunch / 1e3 # in kHz\n",
+    "fpr *= totalMinBiasRate()\n",
+    "qfpr *= totalMinBiasRate()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "a3d20287",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<module 'matplotlib.pyplot' from '/cvmfs/cms.cern.ch/slc7_amd64_gcc11/external/py3-matplotlib/3.7.1-437a2eea83d29aac3bc5f3984f238002/lib/python3.9/site-packages/matplotlib/pyplot.py'>"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Lets plot it!\n",
+    "f, ax  = plt.subplots(figsize=(8,6))\n",
+    "# plt.plot([0, 1], [0, 1], color='navy', lw=1, linestyle='--')\n",
+    "ax.tick_params(axis='both', which='major', labelsize=14)\n",
+    "ax.tick_params(axis='both', which='minor', labelsize=14) \n",
+    "ax.set_xlim(0,100)\n",
+    "\n",
+    "ax.plot(fpr, tpr, color='#7b3294', lw=2, ls='dashed', label=f'Baseline (AUC = {roc_auc:.5f})')\n",
+    "ax.plot(qfpr, qtpr, color='#008837', lw=2, label=f'Quantized+Pruned (AUC = {qroc_auc:.5f})')\n",
+    "ax.set_xlabel('L1 Rate (kHz)')\n",
+    "ax.set_ylabel('Signal efficiency')\n",
+    "ax.legend(loc=\"lower right\")\n",
+    "ax.grid(True)\n",
+    "plt"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2ba99f2a",
+   "metadata": {},
+   "source": [
+    "So it seems despite having reduced the numerical precision of the model and the input, as well as removing 50% of the model weights, we're doing pretty good! This can be tuned to get even better, by carefully adjusting the input precision and the model precision, especially increaseing the precision of the logit layer.\n",
+    "\n",
+    "# Generating firmware with\n",
+    "\n",
+    "<img src=\"https://gitlab.cern.ch/fastmachinelearning/cms_mlatl1t_tutorial/-/raw/master/part2/images/hls4ml_logo.png?ref_type=heads\" width=\"400\"/>\n",
+    "\n",
+    "Time to translate this model into HLS (which we will integrate in the emulator) and use to generate the vhdl to be integrated in the trigger firmware. We will use the Python library hls4ml for that ([here](https://github.com/fastmachinelearning/hls4ml-tutorial/tree/main) is the hls4ml tutorial).\n",
+    "hls4ml seamlessly talks to QKeras, making our jobs way easier for us, but there is still some work for us to do to make sure we get good hardware model accuracy. Lets start!\n",
+    "There are a few things I already know in advance and would like my model to include:\n",
+    "- Be execuded fully parallel (=unrolled) to reach the lowest possible latency. We set the ReuseFactor=1 and Strategy=Latency\n",
+    "- The correct input precision\n",
+    "- The correct model output (that's something you have to figure out yourself!)\n",
+    "- Use \"correct\" precisions to make sure the hardware model performs the same as the software one. QKeras handles weights/biases and activation functions for us, but there are a few other parameters that need to be set by hand\n",
+    "\n",
+    "For the final point, have a look at the following diagram:\n",
+    "\n",
+    "<img src=\"https://gitlab.cern.ch/fastmachinelearning/cms_mlatl1t_tutorial/-/raw/master/part2/images/hls4ml_precisions.png?ref_type=heads\" width=\"400\"/>\n",
+    "\n",
+    "Whereas the $weight$ and $bias$ is set to its optimal value from the QKeras model, the accumulator $accum$ and $result$ is set to some default value that might not be optimal for a given model and might need tuning. Let's do a first attemt:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "9075991f",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/afs/cern.ch/user/t/thaarres/.local/lib/python3.9/site-packages/hls4ml/converters/__init__.py:27: UserWarning: WARNING: Pytorch converter is not enabled!\n",
+      "  warnings.warn(\"WARNING: Pytorch converter is not enabled!\", stacklevel=1)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "WARNING: Failed to import handlers from convolution.py: No module named 'torch'.\n",
+      "WARNING: Failed to import handlers from core.py: No module named 'torch'.\n",
+      "WARNING: Failed to import handlers from merge.py: No module named 'torch'.\n",
+      "WARNING: Failed to import handlers from pooling.py: No module named 'torch'.\n",
+      "WARNING: Failed to import handlers from reshape.py: No module named 'torch'.\n",
+      "Interpreting Model\n",
+      "Topology:\n",
+      "Layer name: input_1, layer type: InputLayer, input shapes: [[None, 56]], output shape: [None, 56]\n",
+      "Layer name: qd1, layer type: QDense, input shapes: [[None, 56]], output shape: [None, 64]\n",
+      "Layer name: qrelu1, layer type: Activation, input shapes: [[None, 64]], output shape: [None, 64]\n",
+      "Layer name: qd2, layer type: QDense, input shapes: [[None, 64]], output shape: [None, 32]\n",
+      "Layer name: qrelu2, layer type: Activation, input shapes: [[None, 32]], output shape: [None, 32]\n",
+      "Layer name: qd3, layer type: QDense, input shapes: [[None, 32]], output shape: [None, 32]\n",
+      "Layer name: qrelu3, layer type: Activation, input shapes: [[None, 32]], output shape: [None, 32]\n",
+      "Layer name: logits, layer type: QDense, input shapes: [[None, 32]], output shape: [None, 1]\n",
+      "Layer name: output, layer type: Activation, input shapes: [[None, 1]], output shape: [None, 1]\n",
+      "-----------------------------------\n",
+      "Model\n",
+      "  Precision:         fixed<16,6>\n",
+      "  ReuseFactor:       1\n",
+      "  Strategy:          Latency\n",
+      "  BramFactor:        1000000000\n",
+      "  TraceOutput:       False\n",
+      "LayerName\n",
+      "  input_1\n",
+      "    Trace:           False\n",
+      "    Precision:       ap_fixed<16,7,AP_RND,AP_SAT>\n",
+      "  qd1\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "      weight:        fixed<8,1>\n",
+      "      bias:          fixed<8,1>\n",
+      "  qd1_linear\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "  qrelu1\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        ufixed<8,0,RND_CONV,SAT>\n",
+      "  qd2\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "      weight:        fixed<8,1>\n",
+      "      bias:          fixed<8,1>\n",
+      "  qd2_linear\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "  qrelu2\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        ufixed<8,0,RND_CONV,SAT>\n",
+      "  qd3\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "      weight:        fixed<8,1>\n",
+      "      bias:          fixed<8,1>\n",
+      "  qd3_linear\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "  qrelu3\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        ufixed<8,0,RND_CONV,SAT>\n",
+      "  logits\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "      weight:        fixed<13,1>\n",
+      "      bias:          fixed<13,1>\n",
+      "  logits_linear\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        fixed<16,6>\n",
+      "  output\n",
+      "    Trace:           False\n",
+      "    Precision\n",
+      "      result:        ap_fixed<13,2,AP_RND,AP_SAT>\n",
+      "-----------------------------------\n",
+      "Interpreting Model\n",
+      "Topology:\n",
+      "Layer name: input_1, layer type: InputLayer, input shapes: [[None, 56]], output shape: [None, 56]\n",
+      "Layer name: qd1, layer type: QDense, input shapes: [[None, 56]], output shape: [None, 64]\n",
+      "Layer name: qrelu1, layer type: Activation, input shapes: [[None, 64]], output shape: [None, 64]\n",
+      "Layer name: qd2, layer type: QDense, input shapes: [[None, 64]], output shape: [None, 32]\n",
+      "Layer name: qrelu2, layer type: Activation, input shapes: [[None, 32]], output shape: [None, 32]\n",
+      "Layer name: qd3, layer type: QDense, input shapes: [[None, 32]], output shape: [None, 32]\n",
+      "Layer name: qrelu3, layer type: Activation, input shapes: [[None, 32]], output shape: [None, 32]\n",
+      "Layer name: logits, layer type: QDense, input shapes: [[None, 32]], output shape: [None, 1]\n",
+      "Layer name: output, layer type: Activation, input shapes: [[None, 1]], output shape: [None, 1]\n",
+      "Creating HLS model\n",
+      "Writing HLS project\n",
+      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
+      "Done\n"
+     ]
+    }
+   ],
+   "source": [
+    "import hls4ml\n",
+    "\n",
+    "def print_dict(d, indent=0):\n",
+    "    for key, value in d.items():\n",
+    "        print('  ' * indent + str(key), end='')\n",
+    "        if isinstance(value, dict):\n",
+    "            print()\n",
+    "            print_dict(value, indent + 1)\n",
+    "        else:\n",
+    "            print(':' + ' ' * (20 - len(key) - 2 * indent) + str(value))\n",
+    "            \n",
+    "\n",
+    "config = hls4ml.utils.config_from_keras_model(qmodel, granularity='name')\n",
+    "config[\"Model\"][\"Strategy\"] = \"Latency\"\n",
+    "config[\"Model\"][\"ReuseFactor\"] = 1\n",
+    "\n",
+    "inputPrecision = \"ap_fixed<16,7,AP_RND,AP_SAT>\" #Adding one bit for the sign, different definitions QKeras/Vivado\n",
+    "for layer in qmodel.layers:\n",
+    "    if layer.__class__.__name__ in [\"InputLayer\"]:\n",
+    "        config[\"LayerName\"][layer.name][\"Precision\"] = inputPrecision\n",
+    "config[\"LayerName\"][\"output\"][\"Precision\"][\"result\"] = \"ap_fixed<13,2,AP_RND,AP_SAT>\"        \n",
+    "\n",
+    "# If the logit layer is a \"normal\" Keras kayer and has not been quantized during the training, \n",
+    "# we need to be careful setting the accuracy. This can be done in the following way:\n",
+    "# config[\"LayerName\"][\"logits\"][\"Precision\"][\"weight\"] = \"ap_fixed<13,2,AP_RND,AP_SAT>\" \n",
+    "# config[\"LayerName\"][\"logits\"][\"Precision\"][\"bias\"] = \"ap_fixed<13,2,AP_RND,AP_SAT>\" \n",
+    "# config[\"LayerName\"][\"logits\"][\"Precision\"][\"accum\"] = \"ap_fixed<13,2,AP_RND,AP_SAT>\" \n",
+    "# config[\"LayerName\"][\"logits\"][\"Precision\"][\"result\"] = \"ap_fixed<13,2,AP_RND,AP_SAT>\" \n",
+    "\n",
+    "print(\"-----------------------------------\")\n",
+    "print_dict(config)\n",
+    "print(\"-----------------------------------\")\n",
+    "hls_model = hls4ml.converters.convert_from_keras_model(qmodel, \n",
+    "                                                       hls_config=config, \n",
+    "                                                       io_type='io_parallel', #other option is io_stream\n",
+    "                                                       output_dir='L1TMLDemo_v1',\n",
+    "                                                       project_name='L1TMLDemo_v1', \n",
+    "                                                       part='xcu250-figd2104-2L-e', #Target FPGA, ideally you would use VU9P and VU13P that we use in L1T but they are not installed at lxplus, this one is close enought for this\n",
+    "                                                       clock_period=2.5, # Target frequency 1/2.5ns = 400 MHz\n",
+    "#                                                        input_data_tb='qX_test.npy', # For co-simulation\n",
+    "#                                                        output_data_tb='qy_test.npy',# For co-simulation\n",
+    ")\n",
+    "hls_model.compile()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "9e00bf3f",
+   "metadata": {},
+   "source": [
+    "First, what does our newly created model look like?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "id": "fc990262",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<IPython.core.display.Image object>"
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "hls4ml.utils.plot_model(hls_model, show_shapes=True, show_precision=True, to_file=None)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "011b95c4",
+   "metadata": {},
+   "source": [
+    "Here you can see that the precision is what we set it to be in QKeras as well as what we set manually in the config. One thing to note is the different definitions used in QKeras and in ap_fixed:\n",
+    "- ```quantized_bits(8,0) -> ap_fixed<8,1>```\n",
+    "- ```quantized_relu(8,0) -> ap_ufixed<8,0>```\n",
+    "Also you can see that the defualt value for result/accu is set to $16,6$. This can also be tuned to more optimal values.\n",
+    "\n",
+    "## Validate the firmware model accuracy\n",
+    "\n",
+    "#et's also run predict on the C++ implementation of our model and make sure it's the ~same as for the QKeras model.\n",
+    "This is very slow for the C++ implementation of our model, but we need a lot of statistics to probe the low rate region. Keep reading while you wait :)!\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "660f657e",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Truth labels:\n",
+      " [0. 1. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
+      "\n",
+      "Qkeras prediction:\n",
+      " [[2.4330037e-04]\n",
+      " [9.9666774e-01]\n",
+      " [2.4369828e-04]\n",
+      " [8.1537422e-03]\n",
+      " [2.4510574e-04]\n",
+      " [2.7655961e-03]\n",
+      " [3.4783210e-04]\n",
+      " [3.7922856e-04]\n",
+      " [3.3851471e-04]\n",
+      " [2.4369828e-04]]\n",
+      "\n",
+      "HLS prediction:\n",
+      " [[0.        ]\n",
+      " [0.99609375]\n",
+      " [0.        ]\n",
+      " [0.0078125 ]\n",
+      " [0.        ]\n",
+      " [0.00292969]\n",
+      " [0.        ]\n",
+      " [0.        ]\n",
+      " [0.        ]\n",
+      " [0.        ]]\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "y_hls = hls_model.predict(np.ascontiguousarray(qX_test))\n",
+    "\n",
+    "print(f\"Truth labels:\\n {y_test[17:27]}\\n\")\n",
+    "print(f\"Qkeras prediction:\\n {qy_pred[17:27]}\\n\")\n",
+    "print(f\"HLS prediction:\\n {y_hls[17:27]}\\n\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "ac7480a6",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<module 'matplotlib.pyplot' from '/cvmfs/cms.cern.ch/slc7_amd64_gcc11/external/py3-matplotlib/3.7.1-437a2eea83d29aac3bc5f3984f238002/lib/python3.9/site-packages/matplotlib/pyplot.py'>"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Lets plot it!\n",
+    "hlsfpr, hlstpr, hlsthr = roc_curve(y_test, y_hls, pos_label=1, sample_weight=None, drop_intermediate=True)\n",
+    "hlsfpr *= totalMinBiasRate()\n",
+    "hlsroc_auc = roc_auc_score(y_test, y_hls)\n",
+    "\n",
+    "f, ax  = plt.subplots(figsize=(8,6))\n",
+    "# plt.plot([0, 1], [0, 1], color='navy', lw=1, linestyle='--')\n",
+    "ax.tick_params(axis='both', which='major', labelsize=14)\n",
+    "ax.tick_params(axis='both', which='minor', labelsize=14) \n",
+    "ax.set_xlim(0,100)\n",
+    "\n",
+    "ax.plot(fpr, tpr, color='#7b3294', lw=2, ls='dashed', label=f'Baseline (AUC = {roc_auc:.5f})')\n",
+    "ax.plot(qfpr, qtpr, color='#008837', lw=2, label=f'Quantized+Pruned (AUC = {qroc_auc:.5f})')\n",
+    "ax.plot(hlsfpr, hlstpr, color='#a6dba0', lw=2, ls='dotted', label=f'HLS Quantized+Pruned (AUC = {hlsroc_auc:.5f})')\n",
+    "ax.set_xlabel('L1 Rate (kHz)')\n",
+    "ax.set_ylabel('Signal efficiency')\n",
+    "ax.legend(loc=\"lower right\")\n",
+    "ax.grid(True)\n",
+    "plt"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "00985383",
+   "metadata": {},
+   "source": [
+    "Oh! That was easier than expected. If you see the accuracies differing significantly, it's a good idea to look into accumulator and reult precisions. Also with tools like $Trace$ and $Profiling$ that you can learn from in the [official hls4ml tutorial](https://github.com/fastmachinelearning/hls4ml-tutorial/blob/main/part2_advanced_config.ipynb) can be helpful! In this case, it doesnt seem like it's necessary. \n",
+    "\n",
+    "## Synthesise!\n",
+    "\n",
+    "Now let's build it! Lets run C-synthesis (C++ to register-transfer level) and Vivado logic synthesis (gate level representation). We will not do co-simulation (send test vectors, do an exhaustive functional test of the implemented logic), but this can be a good idea if you are using CNNs and the $io_stream$ io. \n",
+    "\n",
+    "Let's run!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3d73b5aa",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "report = hls_model.build(csim=False, synth=True, vsynth=True, cosim=False)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3893c6cd",
+   "metadata": {},
+   "source": [
+    "Now, lets, look at the reports! The latency can be extracted from the C-synthesis report, and validated from the co-simulation report (where actual data is sent through the logic. \n",
+    "\n",
+    "The resource consumption can be extracted from the implementation report (Vivado logic synthesis) and is more accurate then what is quoted in the C-synthesis report. \n",
+    "\n",
+    "In this case we did not run co-simulation (this mainly because important when using CNNs and io_stream), but lets print the rest:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "dbc8b9f2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "C synthesis report (latency estimate):\n",
+      "TargetClockPeriod:   2.50\n",
+      "EstimatedClockPeriod:2.176\n",
+      "BestLatency:         15\n",
+      "WorstLatency:        15\n",
+      "IntervalMin:         1\n",
+      "IntervalMax:         1\n",
+      "BRAM_18K:            1\n",
+      "DSP:                 47\n",
+      "FF:                  5914\n",
+      "LUT:                 17754\n",
+      "URAM:                0\n",
+      "AvailableBRAM_18K:   5376\n",
+      "AvailableDSP:        12288\n",
+      "AvailableFF:         3456000\n",
+      "AvailableLUT:        1728000\n",
+      "AvailableURAM:       1280\n",
+      "\n",
+      "Vivado synthesis report (resource estimates):\n",
+      "LUT:                 8436\n",
+      "FF:                  4511\n",
+      "BRAM_18K:            0.5\n",
+      "URAM:                0\n",
+      "DSP48E:              45\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(\"\\nC synthesis report (latency estimate):\")\n",
+    "print_dict(report[\"CSynthesisReport\"])\n",
+    "#print_dict(report[\"CosimReport\"]) # If also running co-sim\n",
+    "print(\"\\nVivado synthesis report (resource estimates):\")\n",
+    "print_dict(report[\"VivadoSynthReport\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "72c1723a",
+   "metadata": {},
+   "source": [
+    "A latency of $2.5\\cdot15=37.5$ ns, that is not bad! \n",
+    "\n",
+    "Also, the network is using very little resources: 8k out of 1728k LUTs, 45 out of 12k DSPs. This is <1% of the total available resources.  We have a set of HLS files that will be integrated into the CMSSW emulator (```L1TMLDemo_v1/firmware/```) and VHDL that will be integrated into the mGT firmware (```L1TMLDemo_v1/myproject_prj/solution1/impl/vhdl/```). That's next!\n",
+    "\n",
+    "If you did not finish synthesising before the start of the next exercise, you can copy an already synthesised project from here:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "80d4fe1d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# ! cp /eos/home-t/thaarres/cms_mlatl1t_tutorial/L1TMLDemo_v1.tar.gz\n",
+    "# ! tar -xzvf "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "83e0276c",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "mlatl1",
+   "language": "python",
+   "name": "mlatl1"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.14"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}